Wikipedia talk:Verifiability/Archive 66

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 60

←

Archive 64

→

Attributability

Should the nutshell reflect this policy's name, or something else?
No, we should stick with the language from the WP:Attribution proposal.	Yes, we should change the wording to refer to this policy's name, which is WP:Verifiability.
all material must be attributable to reliable, published sources.	all material must be verifiable in reliable, published sources.

I replaced some old language about what's "attributable" with what's "verifiable" in this policy today. The immediate issue is that the policy talks about WP:INTEXT attribution, so it's confusing to have the same word used to mean two different things. The background is that language on attribution is leftover from the failed attempt to merge WP:V with WP:NOR years ago into a streamlined, coherent policy that was to be called WP:Attribution (which, for those who weren't around back in 2007, was and still is a good idea, but it isn't going to happen).

This change was reverted with an edit summary of "verifiable is not enough; it must also be attributable using a citation". Even if I weren't aware that verifiable and attributable (in the WP:V+WP:NOR sense, rather than the INTEXT sense) are (and have always been) exact synonyms in wiki-jargon, I cannot imagine why being verifiable would not be "enough" for complying with a policy on verifiability.

But a sensible editor has re-reverted and asked for a discussion, and so here we are: Does anyone actually object to using the word verifiable throughout this policy, instead of scattering in a synonym that has already confused one editor? WhatamIdoing (talk) 07:24, 28 May 2017 (UTC)

Support. I support the change. We want it to be verifiable, not attributable (assuming that's a thing). When we write "The boiling point of mercury is 630 K", we don't mean that this is attributable to some source, we mean that it is verifiable, probably in many sources. Sometimes, attribution is necessary, but many facts in an encyclopedia to which the WP:V policy applies, do not require specific attribution. Sławomir Biały (talk) 12:11, 28 May 2017 (UTC)

I think in terms of WP policy we want or better require the opposite, that is we want "attributable" rather than "verifiable". As the latter might be understood as proving or being able to prove some content to be true or correct and that is something the policy intentionally stays away from. Policy does not require a content to be true, just that in can be found in reliable external sources (which might be wrong). Having said all that, I agree that in terms of WP policy in particular in the context here verifiable and attributable mean the same. However for a non-Wikipedian verifiable (contrary to attributable) by default might mean something else than the use in WP, which can be a source of confusion confusion.--Kmhkmh (talk) 12:55, 28 May 2017 (UTC)

Yes, but, doesn't "verifiable" connote this more refined concept of "attributable", in the context of this policy? Why use a different word for the same concept? Sławomir Biały (talk) 15:09, 28 May 2017 (UTC)

It's not a "more refined concept" – in Wikipedia terms. It's the same concept, slightly re-branded during the noble (but doomed) effort to merge WP:V and WP:NOR.

Also – this may be a distraction, so feel free to skip this if you don't care – there's a small science-vs-non-science subtext in the choice of names. In the hard sciences, you want "The boiling point of mercury is 630 K", full stop, attributed to no one, but in the arts, you want "According to Alice Expert, the poem's rhythmic devices counterpoint the surrealism of the underlying metaphor". The first requires a citation and no INTEXT attribution (because there cannot be two views on this point); the second requires a citation plus INTEXT attribution (because other experts might disagree with Alice, or might think that the poet's compassionate soul was more important). Whether the real-world concept of attribution (in the sense of "giving credit where credit is due" for an idea) is a good thing or a bad thing for the validity and acceptance of the idea varies quite a lot across academia. But on Wikipedia, they are exact synonyms. The point behind the WP:ATT proposal was to merge the policy pages without changing the meaning or requirements. If a reliable source could be found for the material, then that material is equally verifiable and attributable. WhatamIdoing (talk) 17:38, 28 May 2017 (UTC)

Oppose. This major change alters the core meaning of policy. We should not encourage unsourced content. Verifiability only mean it can be sourced somewhere in the world. We want to encourage source content rather than encourage unsourced content. The word "attributable" and "verifiable" have completely different meanings. QuackGuru (talk) 15:17, 28 May 2017 (UTC)

Well, of course you oppose it, because you reverted it. Other editors may want to look in at your comments on another page right now, which say things like "explain how our readers are going to verify a claim when there is no visible citation" and "Verifiable does not mean as long as a reader can verify the claim somewhere on the web or at a library the text is verifiable." and most recently, when I pointed out that WP:V only requires inline citations for three types of content (direct quotations, statements 50% or more WP:LIKELY to be challenged, and statements that have already been CHALLENGEd; BLP adds a fourth requirement) "I challenge all unsourced statements. Any statement on Wikipedia that is unsourced is now challenged."
Wikipedia doesn't work that way. Despite your claims (on many pages over many years now), verifiABLE (and attributABLE) actually does include statements that someone can verify somewhere on the web or at a library. You don't have to like it, but you do have to accept that's what the policy says. WhatamIdoing (talk) 17:21, 28 May 2017 (UTC)

According to this comment you think policy supports adding unsourced content. You claim it has the same meaning, but different words. No, they are very different. Editors are continuing to violate policy. They are adding content that is original research and unsourced. They are replacing sourced content with text that fails verification. Many editors think policy is irrelevant based on the continued proliferation of original research. Wikipedia doesn't work when any editor attempts to dilute the very meaning of policy. QuackGuru (talk) 18:16, 28 May 2017 (UTC)

Yes, you are absolutely correct: I continue to agree with the policy that if you add material such as "The human hand normally has four fingers and a thumb" or "George Washington was the first President of the United States", then that material does not actually require an inline citation under any policy. I do not agree with you that adding such sentences without a citation is a violation of this (or any other) policy, and I'd be surprised if anyone who follows this policy disagrees with me on that point. WhatamIdoing (talk) 22:45, 28 May 2017 (UTC)

Oppose ~~Support~~. ~~The previous version was acceptable, but WhatamIdoing's version is an improvement.~~ I don't like the scare quotes around the second instance of the change, however, in either version and worry about what they add or take away from the meaning. Best regards, TransporterMan (TALK) 18:09, 28 May 2017 (UTC) Now that I better understand the distinction that WaId is making, I can't support this and my opposition would become even more entrenched were he able to get the amendment she is suggesting in the Defining_what_verifiability_means section below passed. — TransporterMan (TALK) 04:38, 29 May 2017 (UTC)
- TransporterMan, I believe that – in terms of Wikipedia policy, not necessarily a dictionary – attributable and verifiable mean exactly the same thing. WhatamIdoing (talk) 21:04, 30 May 2017 (UTC)
Best practice is to cite the source Adding text that you think is "true" with the expectation that someone else will come along and add a high quality reference is not something IMO we want to encourage.Doc James (talk · contribs · email) 18:49, 28 May 2017 (UTC)
- Sure, but what's that got to do with anything? The question here is whether this policy should say "all material must be attributable to reliable, published sources" or "all material must be verifiable in reliable, published sources". Do you have an opinion on that question? WhatamIdoing (talk) 22:45, 28 May 2017 (UTC)
  - All material must be verifiable in a reliable, published source is supporting unsourced content. We attribute to a reliable source. If it is true it is not enough. QuackGuru (talk) 23:39, 28 May 2017 (UTC)
    - No policy has ever said that all material must be attributED to reliable, published sources. Not one, not ever. The policies have only said that "all material must be attributABLE to reliable, published sources" or "all material must be verifiABLE in reliable, published sources."
      The question at hand isn't the -ABLE vs -ED question. The only question at hand is whether the policy should say that someone must be "ABLE to VERIFY that a claim appeared in a reliable, published source" or "ABLE to ATTRIBUTE a claim to its appearance in a reliable, published source." The -ABLE vs -ED parts aren't contested here; only the ATTRIBUTE vs VERIFY parts are being discussed. Given that whichever word is chosen, it will end in -ABLE instead of -ED (i.e., you do not get what you want, no matter what decision is made here), do you actually care whether it says "atrributABLE" or "verifiABLE"? Either way, the policy will still accept exactly the same amount and type of unsourced content being added. WhatamIdoing (talk) 23:53, 28 May 2017 (UTC)
      - You claim "No policy has ever said that all material must be attributED to reliable, published sources."
      - I will cite policy. See WP:VERIFY: "This means all material must be attributable to reliable, published sources." It means what it means. We must cite reliable sources. We can leave it the way it is and let editors decide what it means. AttributED or attributABLE works for me. QuackGuru (talk) 00:25, 29 May 2017 (UTC)
        Here, I've highlighted the difference between what I wrote and what the policy currently says. See the difference in those last few letters? Those are not the same words. When I said that the -ED word was never present, you have replied that the -ABLE word is present, as if that contradicted my claim. The presence of the -ABLE word does not actually prove that the -ED word is present, does it? WhatamIdoing (talk) 00:37, 29 May 2017 (UTC)
        See WP:VERIFY: "This means all material must be attributable to reliable, published sources." The way this is accomplished is by using inline citations. When an inline citation is used the text is then attributed to a source. QuackGuru (talk) 00:45, 29 May 2017 (UTC)
        Yes, I'm quite familiar with the sentence. We agree on this much: when an inline citation is used, the text moves from being merely possible-to-attribute (or possible-to-verify) into the already-attributed (or already-verified) state. But do you see yet that the policy only requires that it be possible for all material ("attributABLE"), and not that it already be done ("attributED")? WhatamIdoing (talk) 01:07, 29 May 2017 (UTC)
        Now we are arguing over the definition of the word "attributable". The rest of the policy explains the details. I don't have any need to add unsourced content. Why should anyone else want to add unsourced content? If it is possible it can be attributable to a source then it can be attributed to a source. QuackGuru (talk) 01:13, 29 May 2017 (UTC)
        Actually, now I'm going to repeat my question: Do you agree that Wikipedia's policies require only that it be possible to cite a source for all material, or do you still insist that the policies require all material to actually possess an inline citation?
        For example, would you say that this sentence, as a description of what Wikipedia's policies actually require is true or false? "As long as someone can verify the material in a reliable source somewhere on the web, or at a library, then the text is verifiable, even if no source has been cited yet." WhatamIdoing (talk) 01:22, 29 May 2017 (UTC)
        Sounds more like a WP:SKYISBLUE argument. That is like claiming we don't really need a source for a claim as long as it is true. That does not work for me. I don't think policy is that relaxed. If someone can verify a claim then they should have no problem using an inline citation. QuackGuru (talk) 01:27, 29 May 2017 (UTC)
        Providing citations beyond the rock-bottom minimum requirements would be lovely, of course, and it is increasingly common, to the delight of all editors. But I'm asking whether you believe that the policy actually requires it. Do you believe that the word "attributABLE" in the policy means "ABLE to be attributed" (=it is possible to cite a source) or "has already been attributed" (=you have already cited a source)? WhatamIdoing (talk) 01:47, 29 May 2017 (UTC)
        Apparently there are editors who claim we don't need to cite that the sky is blue. But we do cite the sky is blue using inline citations. See "During daylight, the sky appears to be blue because air scatters blue sunlight more than it scatters red.[1][2][3][4]"
        
        I require others to provide a citation. I follow my own advice. If it is "ABLE to be attributed" then don't waste time and go ahead and provide a citation. There is no strong reason to wait to provide a citation later. QuackGuru (talk) 02:00, 29 May 2017 (UTC)
        Sure, everyone knows that you want everything cited instantly. But I'm asking you whether the policy actually requires everything to be cited instantly, or if the policies only require that it be possible to do so. Can you just say, "I agree that Wikipedia's policies do not technically require citations for everything (even though I think it would be better if they did)"? Is that a statement that you could agree to? Or do you still think that the policies actually do require citations for everything? WhatamIdoing (talk) 02:34, 29 May 2017 (UTC)
        Changing it to "verifiable in" is tantamount to supporting original research. Any editor can claim the text is verifiable because it can be "verifiable in" a book at the library without providing a source. In a nutshell, policy requires the content be "attributed to" a reliable source whenever "possible". Policy requires editors to aim to add sourced content. QuackGuru (talk) 13:08, 29 May 2017 (UTC)
        Please quote the exact sentence in any policy that says all content, without exception, must be cited "whenever possible". I can provide you with sentences that require exactly four types of content to be cited (without exception); I cannot find any sentence in any policy that requires all content to be "attributed to" any source at all. WhatamIdoing (talk) 21:06, 30 May 2017 (UTC)
        See Wikipedia:Verifiability: "Readers must be able to check that any of the information within Wikipedia articles is not just made up. This means all material must be attributable to reliable, published sources."
        
        Whenever possible "attribute to" a source rather than claim it is "verifiable in" a source. QuackGuru (talk) 00:01, 31 May 2017 (UTC)
        Quack, I'm sorry, but even though I share your belief that everything here ought to be cited, this policy just does not say that. I agree that it seems a bit self-contradictory on that point, but the "likely to be challenged" language simply cannot be read in any way other than to say that some stuff doesn't have to be cited, at least on initial introduction, and that it's the posting editor's decision whether or not challenge is likely. But the notion of verifiability also cannot be used, under this policy, to justify or allow unsourced information to remain in the encyclopedia in face of a challenge and I'm opposed to making that notion more specific and risk the dilution of the ability to challenge and/or remove unsourced information. Best regards, TransporterMan (TALK) 01:52, 31 May 2017 (UTC)
        Anything "likely to be challenged" must be attributable to a reliable source. QuackGuru (talk) 02:05, 31 May 2017 (UTC)

I agree, that's the reason the right to challenge exists - to get rid of the stuff that may not be attributable. Indeed, even uncited stuff that is actually attributable can be challenged and removed, so long as the challenging editor has a good faith concern that it may not be attributable. But I kinda think I'm still missing your point. Best regards, TransporterMan (TALK) 03:07, 31 May 2017 (UTC)
- Any editor can challenge any unsourced content. I challenge everything. It worked. All the original research and unsourced claims were replaced with sourced content. See Heat-not-burn tobacco product. QuackGuru (talk) 03:42, 31 May 2017 (UTC)
  - Interesting, but I don't see how that has anything to do with this discussion about what this policy means and should mean. I'm still not sure I'm getting your point. Best regards, TransporterMan (TALK) 14:57, 31 May 2017 (UTC)
    - I took it one step further at Heat-not-burn tobacco product. I considered everything failed verification until I checked each citation. The content that was unsourced was eventually sourced and rewritten. A lot of content failed verification. It is easier to edit articles when there are less editors. When there are too many editors it is much more difficult to remove the original research. QuackGuru (talk) 15:16, 31 May 2017 (UTC)
      - Interesting, but I don't see how that information about that specific article has anything to do with this discussion about what this policy means and should mean. I'm still not sure I'm getting your point. Best regards, TransporterMan (TALK) 22:09, 31 May 2017 (UTC)
        If policy changed it would essentially mean that all material no longer has to be "attributable to" a reliable source; now it only has to be "verifiable in" a reliable source without having to provide a source. QuackGuru (talk) 00:51, 1 June 2017 (UTC)
        "Attributable to" does not mean "attributed to". "Attributable to" means a source could be found to which the claim could be attributed; "attributed to" means a citation to a source has been made in the article. Jc3s5h (talk) 10:17, 1 June 2017 (UTC)
        Sarah, could you take a look at this comment from Quack? He appears to honestly believe that a requirement for material "attributable to" a source means that inline citations are required for 100% of content, where as a requirement for material to be "verifiable in" a source means that inline citations are not required for 100% of content. This belief is obviously wrong, and he's pushed it past the point of disruption on article talk pages for years. How would you re-write this policy to acknowledge (1) that these two phrases have identical meaning, (2) that both of them equally require citations for WP:MINREF material, and (3) that neither of them requires every single sentence to be followed by an inline citation? You don't seem to approve of the simplest approach, which is to use one word for that concept throughout, so that claims that the words have different meanings is irrelevant.
        TransporterMan, the answer to your question is two comments above: Quack actually does believe that "verifiability" does not require inline citations, but that "attributability" requires inline citations for "all material". He sincerely believes that the presence of the word attributable in this policy gives him license to demand inline citations for every single sentence, and that removing that word would somehow eliminate the policy requirement for inline citations. We need to adjust this policy in some fashion so that everyone gets the same idea of what's required and what's not. WhatamIdoing (talk) 16:23, 3 June 2017 (UTC)
        WAID, does anyone else support Quack's belief? (I'm still not absolutely sure what he believes and he seems to be being very coy about clarifying his exact position.) If he does believe that then he's clearly wrong, and I don't see the need to make policy modifications to address one editor's error. If that's what he believes, and he has no support — he certainly has none here in this discussion — for that position, continuing to flog it in a way that's disruptive may deserve a trip to ANI for an editing restriction, but doesn't justify a policy change. Best regards, TransporterMan (TALK) 17:57, 3 June 2017 (UTC)
        
        WAID, one of the problems is that "verifiable in a source" isn't good English. Something is verifiable if we can authenticate it, show that it's true. What we're discussing here is whether something is attributable to a reliable source—whether we can say: "someone else said this, and that someone else is reliable". We are stuck with the title; it's part of our culture now. But in this policy we have to explain how we're using the term. There's no point in offering a circular definition: "when we use 'the wrong word' in this context, what we really mean is 'the wrong word used slightly differently.'" It's much clearer to explain: "By verifiable, we mean attributable to a reliable source." And that ties up with other uses of attribution.
        As for attributable versus attributed, you're right: that everything must be attributable doesn't mean everything must actually be attributed. But realistically we do expect sources nowadays for just about everything, especially in the areas that QuackGuru edits. SarahSV ^(talk) 15:12, 4 June 2017 (UTC)

That sentence does not mean what you claim it to mean. TransporterMan is exactly right on that point. Even if we would like every single sentence cited – even if we would personally like every single sentence backed by two high-quality, independent, secondary, academic sources – the policy simply doesn't require it. In particular,

"Readers must be able to check" does not mean "you must attribute (aka "cite") whenever possible".
- If the reader is able to check a fact, despite no citation having been provided anywhere in the article, then that material complies with the requirement that the material be verifiable/attributable.
- Secondly, that requirement has no exceptions for "whenever possible". If it isn't possible for you to name such a source upon request/challenge, then you aren't allowed to add your material, with no exceptions. You must always be able to provide a source, but you do not (always) have to actually provide the source.
"...all material must be attributable to reliable, published sources" does not mean "you must cite a reliable, published source". It means "it must be possible for you to cite a reliable, published source even if you haven't cited a single source (yet)". The word attributable means "able to attribute". It does not mean that you have already attributed it; it only means that you are able to do so. WhatamIdoing (talk) 05:29, 31 May 2017 (UTC)

If an editor is able to provide a source then they can cite the source. QuackGuru (talk) 00:51, 1 June 2017 (UTC)

Yes, those verbs mean the same thing, too. "You can provide a source" and "You are able to provide a source" have exactly the same meaning. Did you perhaps mean to say something like "If an editor is able to provide a source then they ~~can~~ must cite the source"? WhatamIdoing (talk) 16:12, 3 June 2017 (UTC)

Verifiability means all material must be attributable to reliable, published sources. If an editor is competent then they should provide a source for verification (confirmation). QuackGuru (talk) 15:44, 4 June 2017 (UTC)

Here are my questions to you, WaId: What practical difference does it make whether or not verifiability only means capable of being verified? You obviously don't think that the policy currently says that, at least not other than through the plain-English definition of the word, or you wouldn't be suggesting the amendment you are in the Defining_what_verifiability_means section below. What practical benefit would it achieve? I'm afraid of that amendment because I feel that it provides ammunition for arguments opposed to BURDEN. Here's how that argument would go. Editor A: "I challenge that unsourced information because I have a concern that it may not be possible to find a published reliable source. Unless someone adds a source, I'm going to delete it." Editor B: "You don't have the right to remove it because it is verifiable under the WaId amendment." Editor A: "If I remove it, BURDEN says that you can't add it back in without a RS." Editor B: "BURDEN does not apply in this case because you don't have the right to remove it because of the WaId amendment." And that brings me to my final question (and I don't mean to imply your answer to this, it's a genuine and honest question) is it your goal to change the standards or methods by which unsourced material can be removed from the encyclopedia? Best regards, TransporterMan (TALK) 04:30, 29 May 2017 (UTC)
- Actually, I do think that verifiable means "capable of being verified" (specifically, it means capable of being verified by a member of the general public who is not the editor that originally added said material, but is willing and able to spend the money necessary to buy a book, hire a translator, etc.). My goal here is not to change any standards. My goal here is to stop an editor from claiming that "policies" require every single sentence to be cited, even though the policies say no such thing. This editor has been claiming that the use of the word attributable means something different from the verifiable. WhatamIdoing (talk) 21:04, 30 May 2017 (UTC)
Oppose Such a change could force the burden of proof from the person adding content to anyone questioning it. wp:SKYISBLUE handles the obvious. Attributable should be used, not verifiable. Jim1138 (talk) 00:53, 29 May 2017 (UTC)
- What do you think the practical difference is between attributable (=able to be attributed to a source, even if that hasn't been done yet) and verifiable (=able to be verified in a source, even if that hasn't been done yet)? Can you give an example of a situation in which an experienced editor would say, "I am able to 'verify' that this claim about the boiling point of mercury has been published in multiple high-quality sources, but I can't actually 'attribute' this claim to any source at all"? WhatamIdoing (talk) 01:04, 29 May 2017 (UTC)
  - It should be the burden of the editor citing what the boiling point of mercury is. Not mine to search for it and cite it. If they sources are numerous, the editor should not have any problem finding a source and citing it. With your proposed change, the burden becomes "prove it is not". Some cases easy, other cases impossible. Disproving "leprechauns exist" would be impossible as the editor could retort that others haven't looked under every toadstool. Jim1138 (talk) 01:56, 29 May 2017 (UTC)
    - Fortunately, we firmly agree on that the BURDEN for providing a citation is on the person who wants to add material.
      But I'm actually asking you a completely different question. No edits to the BURDEN section have been proposed or even contemplated here. Here is the question I'm actually asking: What is the difference between this policy saying "Material must be verifiable" and this policy saying "Material must be attributable"? Is there actually any practical difference? WhatamIdoing (talk) 02:29, 29 May 2017 (UTC)
Oppose the proposed change. The current writing is better. It's an advantage to use attributable here, and attributed when discussing in-text attribution. It ties things together. SarahSV ^(talk) 01:10, 29 May 2017 (UTC)
- It linguistically ties it to the concept of INTEXT attribution, but it also linguistically distances it from the title of this policy. WhatamIdoing (talk) 02:36, 29 May 2017 (UTC)
I think I support -- reading the two nutshell's of W:V and WP:OR is looking in a fun-house mirror at the moment, they say the same thing. I get that there may not be much difference between the bald "verifiable" (although I would say 'verifiable by' not in) and the bald "attributable" but we should not have two policies that say the same thing. -- Alanscottwalker (talk) 12:25, 29 May 2017 (UTC)
- Alan, they should say the same thing, because they're describing the same issue, and it's important to continue that thread across the policies and guidelines. If you use different words, people will get muddled and assume we mean something different. WAID wants the policies to use the same vocabulary too; she wants to change attributable to verifiable at NOR. WP:INTEXT (part of CITE) also uses the same language: it talks about in-text attribution, not in-text verification. There is no reason after all these years to break that continuity. SarahSV ^(talk) 20:26, 29 May 2017 (UTC)

They are not describing the same issue.

If I write, "Black swans are erratic in their migration patterns." That is verifiable.

If I write, "Black Swan events arrive or appear unpredictably." That is verifiable.

But if I write, ""Black swans are erratic in their migration patterns. Black Swan events arrive or appear unpredictably." I have not violated verifiability, but I have violated OR.

I have imparted or implied something original about black swans and Black Swan events, even though both sentences are verifiable. Alanscottwalker (talk) 23:59, 29 May 2017 (UTC)

That introduces a sub-topic, namely WP:SYN, but at root it is indeed the same issue—assuming those sentences together imply something new (I think they are just confusing together rather than implying something new, but the particular example doesn't matter)—namely whether you have a source for the new implication. The point is that WP:V, WP:NOR and WP:INTEXT are obviously connected, and we've used the concept of attribution as a thread throughout them for several years. SarahSV ^(talk) 00:39, 30 May 2017 (UTC)

WP:ATTRIBUTION is pushing up the daisies (although consensus can change). For those who do not know: In late 2006 and early 2007 an number of prominent editors worked very hard on an integrated policy and implemented it in a big-bang, only to find in a subsequent Poll (closed 7 April 2007; one that I did not participate in) there was not enough support for the new policy and so the old ones were reinstated (The whole saga can be read in WT:ATTRIBUTION although the conversation in those archives is measured by the metre so I doubt most will bother). The use of the attribution in this policy was introduced at the time of this debate see here. It first usage after 7 April 2007 was in a footnote on 13:15, 30 July 2007. -- PBS (talk) 10:47, 29 May 2017 (UTC)
- Didn't see this discussion moved from another thread
- I don't agree that the two words mean the same thing. "...attributable to reliable, published sources" is the definition of verifiable. Its redundant to say something is verifiable if it can be verified and redundancy tends to not give us much in the way of information. Phrases like, "This means..." are weak syntactically and are indefinite. We could fix all of this easily by saying:
  Verifiability means that all material must be attributable to reliable, published sources. Additionally, quotations and any material challenged or likely to be challenged must be supported by inline citations. Readers must be able to check that any of the information within Wikipedia articles is not just made up.
- (Littleolive oil (talk) 18:47, 29 May 2017 (UTC))
  - Defining things by their synonyms is a long-standing problem in dictionaries, and that's what that definition gives you.
    Here's a question: Can you give me an example of something that is not verifiable in a reliable source, but is still attributable to a reliable source? How about the other way around? If there is nothing that is A but not B (or B but not A), then the two words have the same meaning (within the context that we're using them). WhatamIdoing (talk) 21:14, 30 May 2017 (UTC)
Support. Comment: I think this would be improved if we defined "attribution" and "verification". I think of them in the WP context as different: Verification is confirming cited attribution, so verifiability is confirmably citable attribution. [I can attribute a quote to Cicero, but can only verify it if there's a reliable source stating that attribution, cited in a way that lets me find and read that source.]

This process can recurse: A single cite, written out in full, might say something like

"user U claimed [on rev-date] to read article A [as it existed on retrieval-date] by author X published in reliable publication P [presumed to have reasonable editing and fact checking], and the article supports the statement 'Stephen Hawking wrote that black holes are portals to another universe.' "

In that example, I would say the statement is attributed to Hawking, the cite verifies the statement, the cite is attributed to X, the original source A verifies the cite, A is attributed to P, and the choice of cite is attributed to U. – SJ + 21:05, 29 May 2017 (UTC)

I don't quite agree with SJ. Verification, as used in connection with this policy, is the act of a reader obtaining and reading a reliable source and determining that the source supports the claim. A claim is verifiable if a reliable source exists with which verification may be carried out, if any reader cares to do so. The mere presence of a citation to an ostensibly reliable source in an article doesn't verify anything, because the source may not exist, may not be as reliable as the citation makes it seem, or may not support the claim. The citation just makes it easier for the reader to carry out the verification. Jc3s5h (talk) 22:10, 29 May 2017 (UTC)

You are right, good catch. I expanded my definition accordingly. This is the entire reason that I proposed adding 1-2 sentences about accessibility, last week -- if none can access a cite for a claim, it can hardly be used for verification. So we should use standard citation formats, page #s or URLs where possible, and free versions of a source where possible. – SJ + 22:28, 29 May 2017 (UTC)

On your Cicero example, it's true that "Cicero said this" is attributed to Cicero but not verified in a source. OTOH, the policy require "attribution to a reliable, published source", so that would not be attribution as far as the policy was concerned.

NB that I would cheerfully drop this if someone could come up with an example of something being verifiable in a reliable, published source without being equally attributable to that same reliable, published source. So far, I find that I'm frustrated by the unjustified claims that I'm trying to discourage editors from citing sources, or this ongoing nonsense that "possible to attribute to a reliable, published source" means "is followed by an inline citation", while "possible to verify in a reliable published, source" somehow does not. The mean the same thing. WhatamIdoing (talk) 21:35, 30 May 2017 (UTC)

Break 1

Kind of oppose; kind of don't. Although I am for the WP:Preserve policy, as WhatamIdoing knows, we don't want editors adding unsourced material and then stating that it's fine because other editors can verify the material on their own. Ideally, we want that material to be sourced to an inline citation, unless it's a WP:Sky is blue case. The WP:Burden portion of the policy states, "The burden to demonstrate verifiability lies with the editor who adds or restores material, and is satisfied by providing a citation to a reliable source that directly supports the contribution. Attribute all quotations and any material whose verifiability is challenged or likely to be challenged to a reliable, published source using an inline citation. The cited source must clearly support the material as presented in the article." And, in my experience, editors generally challenge any unsourced material; see the "cite everything" portion of WP:Sky is blue for what I mean on that, if you don't already know. I never add unsourced material unless it's a WP:Sky is blue case or unless I'm reverting another editor per WP:Preserve; and in those latter cases, I plan to source the material myself. Regarding the edit that WhatamIdoing made, after a discussion with QuackGuru I don't have an issue with the change to the WP:No original research portion of the policy; that policy is not, as commonly believed, about things simply being unsourced; it's about whether or not a source exists for the content. This is another reason why I think we should not have the WP:Verifiability policy essentially be an alternate WP:No original research policy. As for the first change, made to the nutshell box, I can also see why WhatamIdoing made it...since Wikipedia does allow uncited content in some cases (see WP:Sky is blue is again), but I would change "This means all material must be attributable to reliable, published sources." to "This means material should ideally be attributable to reliable, published sources." You know, instead of changing it to "This means all material must be verifiable in reliable, published sources." But that stated, the nutshell does note immediately afterward: "Additionally, quotations and any material challenged or likely to be challenged must be supported by inline citations." So I'm thinking that people will know that the material should generally be supported by an inline citation. Flyer22 Reborn (talk) 21:11, 30 May 2017 (UTC)
- But does this change actually have that effect? Do you honestly think that editors are going to say "Oh, gee, all material doesn't have to be attributable to a reliable source any longer; now it only has to be verifiable in reliable sources! That means I don't need to cite as much content now!" Is that really a reasonable interpretation of changing "all material must be attributable to a reliable source" to "all material must be verifiable in a reliable source"?
  (Also, there are four kinds of material that are required to have inline citations, not merely the three you listed, and it must be possible to supply a RS – whether or not you do – for all content in every case, not merely "ideally".) WhatamIdoing (talk) 21:21, 30 May 2017 (UTC)

If editors interpret the policy to mean that adding unsourced material and then stating that it's fine because other editors can verify the material on their own, that is a problem. And it is a problem many of us have faced before, which is why WP:Burden states what it does. I have no doubt that newbies and other less experienced editors could get the idea that "it's fine because other editors can verify the material on their own" if the policy were worded in a way that suggests that it is not too concerned with sources being in the article but rather with sources existing somewhere. Not that your above wording suggests this, but your below proposed wording could be an issue. And what "three" and "four" are you talking about? Flyer22 Reborn (talk) 23:47, 30 May 2017 (UTC)

@Flyer22 Reborn: Consider the case of an editor who is an expert in the field that the article falls in. The editor will be perfectly capable of writing about well known facts (which are unlikely to be challenged) without having to look each one up in a source. Such an editor should not be required to dig up a source for every sentence. Jc3s5h (talk) 13:02, 31 May 2017 (UTC)

I did not argue that. I was very clear about "sky is blue" cases and "likely to be challenged" cases. What is common knowledge to an expert may not be common knowledge to a layperson and the "likely to be challenged" content should be sourced. If the expert doesn't source it, it is likely to be tagged for needing a citation. And, for the record, as seen here and here, I didn't place my comment under a break section either. But I guess I will take one for the team. Flyer22 Reborn (talk) 14:38, 31 May 2017 (UTC)

We do not have an "expert" exception to WP:V. Kmhkmh has given (26 May) a good statement why we have to show the basis of all content. That an editor is an expert and presumably correct and even unbiased (but not always!) is a good reason to listen to them, but does not show the basis of any content such an editor might add. In that regard even experts are "required to dig up a source for every sentence."

Though I think it is a waste of a valuable resource to have experts writing copy for us. Where an expert is willing to assist, I think they are most useful in giving us an overview of a topic and any principal points and issues, explaining the mainstream expert point of view, pointing us to sources, and even providing sources. Once we have distilled that into text (according to WP principles) they can also review it for errors. So in the end perhaps Jc3s5h is correct: experts "should not be required" to dig up sources, etc., because they should not be required to write text. But all content requires sourcing, regardless of who writes it. ~ J. Johnson (JJ) (talk) 21:17, 31 May 2017 (UTC)

@J. Johnson: "...we have to show the basis of all content" and "In that regard even experts are "required to dig up a source for every sentence." This is a misstatement of the verifiability policy. It is not necessary to provide a source for all content, only direct quotes and content that is challenged or likely to be challenged. Jc3s5h (talk) 21:25, 31 May 2017 (UTC)

You misapprehend my point. It is conceivable for any content to be questioned. Whether a source is required in any given instance depends on the nature of the content (even of the context), not on the nature of the editor. I believe the pertinent statement is: "Any material that needs a source but does not have one may be removed." Note the full-stop (period): There is no exception for experts. ~ J. Johnson (JJ) (talk) 21:34, 1 June 2017 (UTC)

I agree there isn't any exception for experts. However, if there a requirement that every claim be supported by a citation, the requirement would be have a greater effect on experts, since they are more likely to be adding content (of the type that isn't likely to be challenged) without a source right by their sides. Jc3s5h (talk) 22:52, 1 June 2017 (UTC)

You say you agree, but you are still troubled by some supposed "greater effect on experts", and want a softer interpretation of the policy. Which in the end amounts to – an exception for experts. But if you want to delve deeper into this perhaps we should move this discussion elsewhere. ~ J. Johnson (JJ) (talk) 20:34, 3 June 2017 (UTC)

By "experts" what we are talking about in practice here is people with first-hand knowledge, e.g. "I was there", "I am his daughter", or even "I am the subject of this article". This is especially true in cases of breaking news. We've had cases where someone has died and the first report is an anonymous update to their Wikipedia article. I try to find a WP:RS somewhere. Hawkeye7 (talk) 22:18, 3 June 2017 (UTC)

From what I see, "experts" is being argued as those who are scholars or otherwise very knowledgeable on a topic. Above, Jc3s5h stated, "Consider the case of an editor who is an expert in the field that the article falls in." Flyer22 Reborn (talk) 02:47, 4 June 2017 (UTC)

Yes. Merely having special knowledge about something, such as being an eye-witness to something, does not make one an expert. Nor is WP about "first report"; that would be journalism. ~ J. Johnson (JJ) (talk) 02:34, 5 June 2017 (UTC)

Um... Even real scholars and acknowledged experts have to cite their sources - when the information they add is "challenged or likely to be challenged". Most of our experts understand this... but problems arise because experts often have a different expectation of what is likely to be challenged than non-experts have. Experts sometimes get upset when information they assume is unlikely to be challenged... actually is challenged (tagged or removed). Blueboar (talk) 11:22, 5 June 2017 (UTC)

I suspect that is largely because they are used to an environment where there is a high level of professionalism, with greater trust in others' work, and an expectation of reciprocal trust. Questioning one's edit(s) thus has a flavor of questioning one's authority. On the other hand, if we move away from "citation when challenged" towards "citation as routine", there will be fewer challenges.~ J. Johnson (JJ) (talk) 20:08, 5 June 2017 (UTC)

Support Whatiamdoing presents a well founded and historically accurate representation of Wikipedia practice. The four situations she has listed are the only requirements in policy that require inline citations. Content must be verifiable. That is true. The burden also lies, and will remain to lay, on the person adding content to demonstrate that it is verifiable. Most of our articles do not contain content such as "The sky is blue" and so most of our content ends up being cited with inline citations because they are statements of fact that are likely to be challenged. That is the nature of the encyclopedia. Changing "attributable" to "verifiable" does not change the authority or purpose of this policy, nor does it shift any burdens.--v/r - T P 13:48, 1 June 2017 (UTC)

Rather than say what TParis said, "Most of our articles do not contain content such as 'The sky is blue'..." I would say that nearly all of our articles contain a mixture of statements, some of which have no more need of a citation than "the sky is blue", and some of which are likely to be challenged. So only some of the statements in a normal article require citations. For uncontroversial statements that are well known in the field, and have not (yet) been explicitly challenged, a general reference with a good index may suffice. Jc3s5h (talk) 15:32, 1 June 2017 (UTC)

To me so-called "general references" seem to say either "I'm sure I saw 'x' some where in this source, but I can't be troubled to look for it, find it yourself", OR "Eeek, I want to cite this source more than once but I don't know know how to handle multiple page numbers so I'm going to cite this source "generally"." I'm fine pointing a reader to a source "generally" for background information or such, but specific content should be specifically cited. As stated at WP:BURDEN (emphasis added): "Cite the source clearly and precisely (specifying page, section, or such divisions as may be appropriate)." ~ J. Johnson (JJ) (talk) 21:39, 1 June 2017 (UTC)

You can read the WP definition here: WP:GENREF. In the light of the Verifiability policy, though, such "general references" are absolutely useless, because WP:V is geared toward inline citations to MINREF/BURDEN content. Material "cited" to general references fails these just like completely unreferenced material does. – Finnusertop (talk ⋅ contribs) 22:08, 1 June 2017 (UTC)

Prefer "Attributable" - The problem with "verifiable" is that it is all too easy to confuse the term with a related term: veracity (ie truth). Often what we are trying to demonstrate with a citation is the fact that a reliable source says X... not that X is necessarily accurate or true. For example, we can not cite any source to verify the statement "Jesus is the Son of God"... however we can attribute that statement to expert theologians ... and use that attribution to verify the statement "Christians believe that Jesus is the Son of God." Blueboar (talk) 16:57, 5 June 2017 (UTC)
Prefer "Attributable" - this is more directly actionable in terms of wikipedia content in the case when it really matters: when content is actually challenged. Per policy, the correct answer to such challenges is to provide attribution. Therefore IMO no need in "middleman" terms. Staszek Lem (talk) 00:58, 6 June 2017 (UTC)

WP:Citation overkill RfC

Opinions are needed on the following matter: Wikipedia talk:Citation overkill#Should this essay be changed to encourage more citations?. A WP:Permalink for it is here. Flyer22 Reborn (talk) 00:55, 9 June 2017 (UTC)

Military history project: FA/GA discussion

MilHist project is hosting a discussion on GA / FA articles that have been tagged as possibly having unreliable sources. The discussion can be found here:

MILHIST: GA / FA articles.

Interested editors are invited to participate. K.e.coffman (talk) 21:59, 10 June 2017 (UTC)

Recent change to Template:Refimprove

We need some opinions on the following matter: Template talk:Refimprove#Or better. A permalink for it is here. Flyer22 Reborn (talk) 01:44, 17 June 2017 (UTC)

Discussion of interest

You may be interested in this discussion, which relates to an interpretation of this policy. Wikipedia:Miscellany for deletion/Wikipedia:Citation underkill. Scribolt (talk) 18:57, 17 June 2017 (UTC)

It goes both ways

Withdrawn by proposing editor, see below. — TransporterMan (TALK) 22:47, 19 June 2017 (UTC)

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

See here. QuackGuru (talk) 18:38, 11 June 2017 (UTC)

I would have thought you would have known better than to make such an edit without establishing a consensus in favor of it first. It troubles me that you would link to it without clearly establishing that you were also the editor responsible for it. No surprise to me that it was reverted. DonIago (talk) 19:02, 11 June 2017 (UTC)

Agree with the revision. It is easy for those wanting to include contested information to meet the onus... all you need to do is explain why you think the information should be included (if you can't do this, then it is likely that the information doesn't belong in the article). Having easily met that onus, it is then up to those who want to exclude it to refute that explanation. Blueboar (talk) 21:16, 11 June 2017 (UTC)

The revision does not make it clear that an onus to show a consensus to remove information only applies if there was a consensus to add it in the first place. Blueboar's statement is a perfect example of how the revision could be misinterpreted: "all you need to do is explain why you think the information should be included...." No, that's not enough; it's also necessary that other editors express approval of the edit, or allow the edit to remain for a substantial time. Jc3s5h (talk) 21:30, 11 June 2017 (UTC)

I agree with Jc3s5h. Whether or not the onus is also on those seeking to exclude content is irrelevant in the context of this section since the onus being referred to is that which is being addressed in the first two sentences of the section and, most importantly, in the phrase "verifiability alone is not a reason for inclusion". The entire purpose of this section is to refute the argument that material must be included in an article if it is proven to be verifiable. Best regards, TransporterMan (TALK) 22:22, 11 June 2017 (UTC)

I somewhat agree with you, Jc3s5h. It does not hurt to try again. QuackGuru (talk) 22:38, 11 June 2017 (UTC)

Making it more clear is even better. The entire purpose of this section should not be one-sided. QuackGuru (talk) 22:38, 11 June 2017 (UTC)

As per User:TransporterMan. I'm confused as to what problem is trying to be solved here, and in particular how it relates to the fact that 'Verifiability is not a reason for inclusion'. Scenario A: Unverifiable content in an article. Editor A looks at it and decides to either tag it or remove it. These actions are described in the policy already. Scenario B: Verifiable content is in an article. Editor A looks at it and decides for other reasons it should be removed (NPOV, UNDUE, BLP etc). Other editors either object or don't and a discussion ensues. Firstly could someone explain how Scenario A isn't covered by the existing wording in this policy. Secondly, could someone explain why the second scenario relates to Verifiability, and if it does, why it is best addressed in this particular section. Scribolt (talk) 06:36, 12 June 2017 (UTC)

Scenario A is covered. The second scenario relates to Verifiability only to the extent that a lot of editors think that if something is verifiable, it should not be removed. They don't understand that something can be verifiable and yet still be removed... indeed it can be removed simply because there is a consensus to remove it. It might be verifiable... it might pass every other content and policy and guideline... but if it does not pass WP:CONSENSUS, it can be removed. Blueboar (talk) 17:08, 12 June 2017 (UTC)

Good, my understanding mirrors yours. We're trying to fix a mis-understanding with other Wiki policies with badly written throwaway comment here. Even if the text was clarified, I still think it's a bad idea. Scribolt (talk) 06:01, 13 June 2017 (UTC)

The onus is on inclusion alone. Show how the reliable sources demonstrate that the material is encyclopedic. If reliable sources mention this information in passing, or if its inclusion is of uncertain weight, there is no reason to include it. Bright☀ 19:48, 12 June 2017 (UTC)

The onus is also on exclusion unless editors support whitewashing articles. WP:CON on Wikipedia is grounds for any editor to exclude anything. QuackGuru (talk) 19:53, 12 June 2017 (UTC)

It's quite funny how you mirror my comment's style but not substance. While I use consensus and Wikipedia policy, you present a minority opinion that is not accepted. Bright☀ 19:54, 12 June 2017 (UTC)

Consensus is tantamount to whitewashing articles, presenting a majority opinion that is allowable? QuackGuru (talk) 19:59, 12 June 2017 (UTC)

If you have a problem with WP:WEIGHT go discuss it there. WP:ONUS, the section that you are editing, says, in so many words "you found some verifiable information and put it in a Wikipedia article? That's good, but mere verifiability is not the same as being encyclopedic, and you should consider the quality of the source and the extent of which the verifiable information's importance is apparent in it."

It just looks like you're fighting for WP:FRINGE, when, in general, new and exciting information does not belong on an encyclopedia, which covers old and accepted knowledge.

Remember, Wikipedia is not everything. Content should be excluded unless it's encyclopedic, not the other way around (included unless shown it's not encyclopedic). Bright☀ 20:08, 12 June 2017 (UTC)

It only seems as if QG fighting for WP:NPOV. Apparently, Wikipedia covers everything QG added. See Talk:Safety of electronic cigarettes/Archive 3. Mass content was deleted and then restored. ONUS is being used to delete mass content. QG does not like making minor changes. QG likes to makes mass changes. Half-baked articles are allowed, but QG likes them well done. What happened next? Sourced content was replaced with WP:OR. QuackGuru (talk) 21:12, 12 June 2017 (UTC)

What is that all about?

I concur fully with Bright* that "[c]ontent should be excluded unless it's encyclopedic, not the other way around". Which is why (as Blueboar says) "something can be verifiable and yet still be removed". The persistence of this issue seems to arise from editors who think WP will be boring if it is not sensationalistic. At some point we should work out a complete and compelling argument settles the matter. ~ J. Johnson (JJ) (talk) 22:44, 12 June 2017 (UTC)

The continuation of this problem appears to stem from editors who believe WP can be censored if they keep on reverting. That is what that is about. QuackGuru (talk) 23:28, 12 June 2017 (UTC)

A lot depends on the specific situation, but Wikipedia can in fact be "censored"... if there is a clear consensus to censor. We even explicitly enshrine one specific form of "censorship" in policy (see WP:UNDUE). Blueboar (talk) 00:48, 13 June 2017 (UTC)

Objecting for no legitimate reason is whitewashing and can be seen as censorship. Consensus can be used as a form of censorship. Wikipedia is not supposed to be censured. There should be no specific form of "censorship" (see WP:NOTCENSORED). QuackGuru (talk) 15:13, 13 June 2017 (UTC)

Have you read, with any understanding, the section you cite? WP:NOTCENSORED – a.k.a. WP:CENSORED – is about conforming, or not, to reader's notions of acceptability. The second paragraph begins: "Content will be removed if it is judged to violate Wikipedia policies". (I.e.: censored!) The topic there is about not removing content that is otherwise encyclopedic. The topic here is more about whether non-encyclopedic content can be freely added (and inversely, freely deleted) in the first place. So your reason ("censorship") for objecting to "[o]bjecting for no legitimate reason" is itself invalid, and your objection is a prime example of what you object to. Hilarious.

The reality is: there IS, and rightly SHOULD BE, certain specific forms of censorship. However, they are separate from the general issue of inclusion. It appears that the particular cases that concern you were about inclusion, but being unable to make an argument for that you keep yelling "censorship". Which is simply not applicable here. ~ J. Johnson (JJ) (talk) 23:21, 13 June 2017 (UTC)

Where is the specific part of policy or guideline that addresses editors deleting pertinent content for no logical reason? It is not one editor. It is a small group. They all have something in common. If they can't delete content then they do it another way. They all have replaced sourced content with original research on the same page. That is like deleting content. QuackGuru (talk) 05:11, 14 June 2017 (UTC)

The pertinent policies are WP:Ignore all rules and WP:CONSENSUS. Anything can be included or excluded if there is a consensus to include or exclude it. Whether there is a consensus (or not) for including or excluding some specific bit of information is another matter... to be discussed on the relevant article's talk page (and through an RFC if need be).

As for people removing information "for no legitimate reason"... that begs the question: who determines whether a reason is "legitimate" or not. The answer (once again) is "consensus". Even, "I just don't think it we should say this" can a valid reason, if consensus agrees. Blueboar (talk) 12:10, 14 June 2017 (UTC)

Yes.

QG: what seems to be driving you is a relatively small number of specific edits that you do not understand. (Note that your saying these were "for no logical reason" goes no further than showing that you see no such reason.) If these all involved a single article (or even a group of related articles) then it is probably a content dispute, and should be discussed there. If it was, but you were not satisfied, then it slides into being a personal problem of understanding, for which your talk page would be a more appropriate venue.

You ask for a specific applicable policy, but that is not possible lacking the specific details of where you are coming from. To address this angst you feel you should provide (but not here!) a list of all the edits that you object to, including their context and any previous discussion about them. Lacking that we have no referents, and you are just waving wildly in the dark. ~ J. Johnson (JJ) (talk) 20:56, 14 June 2017 (UTC)

If I Wikipedia:Ignore all rules than I am following the rules. Wikipedia has contradictory policies and rules. It is unlikely it is a content dispute when sourced content is replaced with content that failed verification. This is being done by a group of editors on the same topic. This is crazy. QuackGuru (talk) 01:58, 15 June 2017 (UTC)

If you're going to make such claims, you really need to provide diffs rather than simply talking about it in the abstract. Give us something we can look at and respond to, rather than vague claims that there's a problem without any supporting evidence. DonIago (talk) 06:39, 15 June 2017 (UTC)

Yes: really. And I really mistrust your unsupported assertion these unidentified matters are not a content dispute, as they sure quack like one. I strongly suggest you collect diffs of these mystery edits on your talk page (so you don't further bog-down the discussion here), and then perhaps some kindly editors with some spare time might help you sort out how to best soothe your feelings. ~ J. Johnson (JJ) (talk) 18:15, 15 June 2017 (UTC)

I collected most of the diffs in my e-mail box. Uninvolved admins supporting OR was the last straw. QuackGuru (talk) 18:36, 15 June 2017 (UTC)

I'm glad to hear that you have them collected, and I look forward to your posting them on your Talk page or in another location where we might review them to gain a better understanding of your concerns. DonIago (talk) 18:38, 15 June 2017 (UTC)

Editors are not allowed to focus on others on Wikipedia. I will have to find another way to fix the avalanche of original research. QuackGuru (talk) 18:54, 15 June 2017 (UTC)

At this point, I have to call "Bullshit" ... there is nothing wrong with compiling a short list of difs to illustrate a potential problem. You would be focusing on the edits not the editors who made them. Indeed, if the potential problem really is as wide spread as you claim, you should be able to come up with examples from different articles, made by different editors. Time to "put up or shut up". Blueboar (talk) 19:15, 15 June 2017 (UTC)

In QuackGuru's defense, there are a lot of bad reverts on Wikipedia... however I find that policy is against these sort of bad reverts, for example WP:EPTALK, WP:TALKDONTREVERT, and WP:SOURCES, as well as WP:ONUS as it's currently phrased. Still, this discussion seems to be a case of barking up the wrong tree. Bright☀ 13:43, 16 June 2017 (UTC)

Dose the existing consensus mean the article improved? Consensus does not remove original research. Consensus does not do anything to improve any article. WP:ONUS as part of consensus is tantamount to obstructionism. What is really consensus? It is a way to tell others to shut up. QuackGuru (talk) 14:52, 16 June 2017 (UTC)

If you really want to improve articles, add information from high-quality reliable sources. I have found that consensus always favors information from high-quality reliable sources, and arguments simply disappear if there are high-quality reliable sources available. Bright☀ 19:05, 16 June 2017 (UTC)

I did add information from high-quality reliable sources. Consensus does not favor sourced content or unsourced content. It favors which side has a larger group. QuackGuru (talk) 22:56, 16 June 2017 (UTC)

Quack, I have a really hard time understanding what you're trying to say, and I wonder if it's not because of a fundamental difference in the way we regard the encyclopedia. It seems at times you're opposing the very concepts that Wikipedia is based upon. When you say that, "Consensus does not do anything to improve any article" you're saying that the idea of Wikipedia being a wiki doesn't work. If so, you're not the first one to feel that way, but tinkering with policies ain't gonna cure your concerns. The other folks participating in these discussions (including me) believe, or at least accept, that the wiki system based upon consensus works and is, indeed, in the long run the best way to improve articles and the encyclopedia as a whole. Regards, TransporterMan (TALK) 23:03, 16 June 2017 (UTC)

The wiki system based upon consensus works? How does consensus stop original research or remove original research? QuackGuru (talk) 23:11, 16 June 2017 (UTC)

QG: Your statements – such as your "Consensus does not do anything to improve any article" – are so off-base, so divorced from the reality the rest of us know, as to be meaningless. Your views are so antithetical to our fundamental model for editorial decision-making, and your apparent notions of what an encyclopedia is and how to build it so defective, that your presence here begins to look nothing short of disruptive. Perhaps you should reassess that. ~ J. Johnson (JJ) (talk) 23:17, 16 June 2017 (UTC)

How does consensus stop original research That's an easy one. Point 1: by massive consensus a policy was established that OR is disallowed. Point 2: a local consensus cannot override the policy lightly. Step 3: if by some quirk one runs into a walled garden with weird local consensus, we have procedures for dispute resolution. Staszek Lem (talk) 23:21, 16 June 2017 (UTC)

In case you need a link, WP:Dispute resolution -- if you cannot work it out in normal editing or talk it out on the page, generally get others editors involved appropriately. But sometimes, under this system we have, one is bound to be disappointed from time to time, but that reality should be apparent and come as little surprise. -- Alanscottwalker (talk) 23:33, 16 June 2017 (UTC)

"One is bound to be disappointed from time to time"... All too true. For example, one editor may be convinced that an edit is original research, but other editors (consensus) disagree with that one editor. The one editor is going to go away disappointed. Blueboar (talk) 00:43, 17 June 2017 (UTC)

Please show me where you added a high-quality reliable source and it was struck down by consensus. Please mind that your notion of "high quality reliable source" may not meet Wikipedia's description of a high-quality reliable source. Bright☀ 10:03, 19 June 2017 (UTC)

This is not the appropriate venue to discuss this. It was odd after the reviews were added more than one editor added sources that were not reliable to counter the reviews. It was also strange the copyright violations and mass original research. One editor saw the problems on one of the pages and was quickly reverted. I think it is best to let it go for now. Hopefully they will eventually stop. QuackGuru (talk) 14:39, 19 June 2017 (UTC)

You brought it up repeatedly and when repeatedly asked for diffs you said it's not appropriate to discuss here. Gotcha. Bright☀ 15:08, 19 June 2017 (UTC)

"This is not the appropriate venue to discuss this", you say, and then – you proceed to discuss this. But whining without any substantiation is just annoying everyone, and seriously undermines your credibility. You are just wasting everyone's time. ~ J. Johnson (JJ) (talk) 21:34, 19 June 2017 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Template:no source under discussion

Currently, Template:no source and Template:nosource, both of which redirect to template:di-no source, are nominated for discussion at Wikipedia:Redirects for discussion/Log/2017 June 14#Template:No source, where I invite you to comment. --George Ho (talk) 03:57, 20 June 2017 (UTC)

Recent changes to policy about verifiability as a reason for inclusion

The following discussion is an archived record of a request for comment. Please do not modify it. No further edits should be made to this discussion. A summary of the conclusions reached follows.

This page in a nutshell:

The result of the discussion was No Consensus.
The policy is reverted to the January 2017 version.
The community is asked to re-start an RFC in no sooner than two weeks to fully resolve the issue

Details---

Altering policies on Wikipedia requires a considerably higher standard of consensus than what is necessary/required for most other discussions. The main discussion in this RFC was a statistical tie numerically, and none of the arguments presented were clearly superior to the others or easily refuted.
None of the proposals in this discussion-- the two texts in the RFC question or the compromise proposals that arose toward the end of the discussion (which appeared to be promising but attracted relatively little attention) gained the desired levels of consensus and thus failed to make it to the policy.
The recent change which is the subject of this RFC was installed in the policy as a result of a local consensus. While the execution of the changes was procedurally perfect, the changes have failed to find a consensus in this much larger discussion.Since site-wide consensuses supersede local consensuses, it would be plainly unsatisfactory to leave the current wording in place. Therefore, the policy should be restored to its longstanding wording (i.e. the wording that was in place in January of this year).
There is clearly some discontent with the longstanding version of this policy, and we are only restraining from prescribing a change due to the lack of consensus for any other option in this RFC. In an attempt to find a better resolution in going forward, the community is encouraged to create a new RFC in no less than two weeks from now; (two weeks should be sufficient time for everyone to approach the wordsmithing with fresh eyes.) The new RFC shall ideally contain both the wordings initially considered in this RFC, as well as a couple of alternate/compromise proposals such as the ones that were debated towards the end of this RFC.

Signature--

Winged Blades^Godric 09:25, 1 July 2017 (UTC)
Primefac (talk) 18:34, 1 July 2017 (UTC)
Tazerdadog (talk) 18:40, 1 July 2017 (UTC)

Recently a talk page discussion on this policy page led to this change.

Q: Was this a good idea? Please express a view below.—S Marshall T/C 23:12, 9 April 2017 (UTC)

Before change	After change
Verifiability does not guarantee inclusion	Verifiability is not a reason for inclusion
While information must be verifiable in order to be included in an article, this does not mean that all verifiable information must be included in an article.	While content must be verifiable in order to be included in an article, verifiability alone is not a reason for inclusion, and does not guarantee that any content must be included in an article.
Consensus may determine that certain information does not improve an article, and that it should be omitted or presented instead in a different article. The onus to achieve consensus for inclusion is on those seeking to include disputed content.

Note: the same discussion also prompted this related change to the lede, also relevant to this Request for Comments. Diego (talk) 13:06, 10 April 2017 (UTC)

Yes

Reyk _YO! 14:30, 9 April 2017 (UTC)
Yes, for many many reasons delineated in the thorough discussion that led to the change, and also the many reasons delineated in the more recent large discussion discussion. Briefly:

WP:ver in essence imposes a requirement for inclusion, nothing more. Nowhere does it say anything about meeting wp:ver being a full or partial reason for inclusion. Some folks have been misreading this. This merely means that once wp:ver has been met, wp:ver "steps aside" and allows all other policies and practices to determine what happens.
As detailed at length in the discussion on this, a common problem (rooted in a false urban legend in Wikipedia) is mis-use or mis-reading of this policy to claim that meeting wp:ver is a force or reason for inclusion. For example, every time an editor says the ubiquitous "undo removal of sourced material" in an edit summary they are in essence making this claim. Nobody makes the claim that it is an absolute mandate for inclusion, but it is quite common to make the false claim that it is a strong force or reason for inclusion. This false urban legend is extensively used by wiki-lawyers to POV articles. Flatly saying that meeting wp:ver is not a reason for inclusion in the lede is (unlike "not sufficient") an accurate summary of of the body of the policy, and directly addresses the false urban legend.
IMHO many folks (including, going by memory, one or more current or ex Arbcom members) have said that Wikipedia has an overly byzantine set of confusing and overlapping / conflicting policies. This is a perfect example of sitting on the edge of that precipice, and avoiding the abyss. Regarding content, WP:VER sets a requirement for inclusion, nothing more unless we let it be a mess. It should not try to wade into the immensely complicated question of "sufficiency" for inclusion. To start with, "sufficient" requires not being in violation of dozens of policies and core guidelines. If those are complied with, there is an immensely varying additional bar for inclusion. At the low end might be an uncontested inclusion at an obscure article where the additional bar is essentially zero. At the other end of a spectrum, the additional bar for inclusion (on a contested major item) might be getting a strong consensus in an RFC. WP:VER should stick to imposing the verifiability requirement, not start reflecting on or overlapping with the hugely complex area of what constitutes sufficiency for inclusion.
For new timid editors, this falsely implies that there is some unknown-to-them wiki-rule about needing a compelling reason to include something. This imposes a false hurdle and also conflicts with new editor objectives (and wp:bold)
The RFC just asks for an overall opinion on collectively several different changes made in February, without asking or proposing any specific actions on any of them. If anyone is thinking that it is asking about any particular action, it is malformed with respect to that. But the work below on a compromise action is showing some promise.North8000 (talk) 21:03, 2 May 2017 (UTC)
The current wording was arrived at in a 27,000+ word discussion that ran for 23 days (Feb 7th - March 2nd) with no objections at the end of the process. North8000 (talk) 22:01, 18 April 2017 (UTC)

To be fair, S Marshall's first objection was raised only 18 days after the discussion you point us to ended (he started to object on March 20th) and we have been debating his objections ever since. Given the short time span between the change and his first objection, this all needs to be seen as one single ongoing discussion that started in Feb, and has not yet ended. Blueboar (talk) 22:23, 18 April 2017 (UTC)

I don't agree, although if it would resolve that aspect I'd agree to a compromise that considers them to be on equal footing. North8000 (talk) 01:45, 19 April 2017 (UTC)

North8000 (talk) 14:43, 9 April 2017 (UTC)

The idea that a fact being covered by a reliable source provides a reason to try to include it in the article is not a "false false urban legend in Wikipedia", it is how many of us see the policies and the core way that Wikipedia is built. If a strongly reliable and high-profile source covers a fact in depth, that's a very good reason for that fact to be included in a relevant article; but the new wording makes this coverage look not-at-all important. Surely there may be other reasons why the content might be excluded, but this doesn't mean that being verifiable provides no reason at all for using it, not even a partial one, as the new text says.

With the new version, if you have two facts where one is non-verifiable and the other is covered in-depth by multiple reliable sources, both would have the same reason for inclusion according to this policy, which is none at all. In fact, now the only remaining reason in policies to include content is editors have consensus to include it; in an attempt to remove from WP:V the weakly verified content, the really strong verifiable has been eliminated with it, thanks to the poorly and overbearing way the new wording is expressed. Diego (talk) 12:40, 10 April 2017 (UTC)

I point you to NPOV. That's the policy that requires looking into allot of strong sourcing (thus ,expressly making that a non-V reason). The language under discussion says "verifiaability alone", which just reinforces the long-standing introduction that points to other policy and again says, 'you know, you do have to come up with other reasons than you have one RS.' Alanscottwalker (talk) 15:07, 10 April 2017 (UTC)

Do you mean this wording in the lede? "However, while verifiability is required for including something, it is not a reason for inclusion." It has been there since last february, at the same time that the section was changed; there's no long-standing introduction saying "you have to come up with other reasons". By the way, NPOV introduction says: "It is also one of Wikipedia's three core content policies; the other two are "Verifiability" and "No original research". These policies jointly determine the type and quality of material that is acceptable in Wikipedia articles, and, because they work in harmony, they should not be interpreted in isolation from one another.". Diego (talk) 16:09, 10 April 2017 (UTC)

I think the lead would best follow the body of the policy,("verifiability alone"), but no I expressly pointing to the links to other policy including NPOV in V's lead, which says you have to look at more than V to determine content. Alanscottwalker (talk) 19:00, 10 April 2017 (UTC)

@Diego Moya:, I'm with you on your overall view, but what you don't realize is that the February wording is also with you on your overall view. The February wording merely accurately summarizes wp:VER; once wp:ver is met, it steps aside, allow other practices (such as normal article building) and policies take over. In fact, per my last of four points, the February wording actually usually works on the inclusionist side. North8000 (talk) 21:22, 10 April 2017 (UTC)

Yes, Verifiability ≠ nobility/inclusion. - Mlpearc (open channel) 14:52, 9 April 2017 (UTC)
Yes, per North, but also because "does not guarantee inclusion" implies that it is a reason for inclusion...and it isn't. Nobody should be adding information to an article if their sole reason for doing so is "it's verifiable". DonIago (talk) 15:53, 9 April 2017 (UTC)

Question: why shouldn't someone add information if their sole reason is "It's verifiable"? (Note: adding information is not the same as objecting to someone else removing it) Blueboar (talk) 16:11, 9 April 2017 (UTC)

For various reasons outlined in the policies WP:NOT and WP:POV, and in the guidelines like WP:RS and WP:TECHNICAL, and even style guidelines like WP:EMBED. Bright☀ 20:48, 9 April 2017 (UTC)

Consider this... suppose there is information that you think might improve an article... you are fairly sure it is accurate, but you hold back on adding it because you don't have a source. You search and search... and Then you finally find one. You can now add the information... why? The reason is because you can now show that it is verifiable.

Now, this verifiability does not guarantee that the information will stay in the article... we all agree that there are many reasons why it might get removed... but the verifiability does allow you to add it in the first place.

In other words, verifiability can be a reason to include information... But verifiability is not enough to protect information from removal (i.e. exclusion). It's a subtle, but important distinction. Blueboar (talk) 22:36, 9 April 2017 (UTC)

"you think might improve an article" -- you already have a reason, and it is not that it is verifiable because (in your scenario) you are not even sure it is verifiable. You should already be able to articulate why you think it improves the article. -- Alanscottwalker (talk) 22:49, 9 April 2017 (UTC)

That hypothetical person already felt the information improves the article even before they knew it's verifiable, so the information must have had some other value. Then, once it's verified and added, that reason can be considered by other editors, hopefully according to Wikipedia policies and guidelines. Bright☀ 23:57, 9 April 2017 (UTC)

It must have some other value than verifiable, as you say, and that reason existed before it was added. Alanscottwalker (talk) 12:14, 10 April 2017 (UTC)

WP:WEIGHT says "Neutrality requires that each article or other page in the mainspace fairly represent all significant viewpoints that have been published by reliable sources (yadda yadda prominence of each viewpoint)", so apparently some other policy does consider verifiability as a reason to include at least some kind of content, in some circumstances. Diego (talk) 13:51, 10 April 2017 (UTC)

No. That other policy, expressly makes that an NPOV reason not a V reason. And it's actually focusing on, examining and representing the wide literature - that is not V's issue, which is 'we require a single RS'. Alanscottwalker (talk) 14:26, 10 April 2017 (UTC)

And per NPOV, the reason why a topic is prominent enough to have due weight is ... drum roll ... because it is verifiable in several RSs! The assertion "Verifiability is not a reason for inclusion" at the very least mischaracterizes this part of WP:DUE. Diego (talk) 15:10, 10 April 2017 (UTC)

The policy says: "Verifiability alone is not a reason for inclusion." Alanscottwalker (talk) 15:31, 10 April 2017 (UTC)

But the section title says: "Verifiability is not a reason for inclusion." Diego (talk) 15:38, 10 April 2017 (UTC)

Yes. I see altogether too many partisan editors who insist that an encyclopedia article has to include today's trivial tempest in a teapot solely "because it's got reliable sources" (and because it advances their agenda). — Malik Shabazz ^Talk/_Stalk 17:01, 9 April 2017 (UTC)

This is a typical way how articles are expanded:

A reliable source describing some fact about a topic provides support towards including the fact in an article.
Some editor provides a reason why this fact should not be included, based on another policy.
The strength of the source and the other policy are weighted, and a decision is made by consensus whether to include the fact or not.

The blunt recent change to policy amounts to forbidding step 1, i.e. including a fact in an article while giving as reason that you have found a reliable source supporting it; but this is overkill. The right approach to solve your stated problem instead would be to emphasize step 3, i.e. that in case of dispute you need consensus even if the fact is verifiable. Diego (talk) 14:02, 10 April 2017 (UTC)

"Does not guarantee" is the wrong idiom. At issue is not whether anything is guaranteed, but whether it is a good idea. "Is not a reason" is much better, but might be amended for clarity to read "is not itself a reason". (Insofar as non-verifiability is a sufficient reason for exclusion, verifiability is a necessary but not sufficient reason for inclusion.) ~ Ningauble (talk) 19:18, 9 April 2017 (UTC)
Per above, Malik Shabazz and Ningauble, whom I assume in talking about 'good idea', includes in that the general fulfilling content policies/guidelines (not just V) . I also think this is better, in part, because Verifiability policy sets its minimalist requirement at only "a" (single) RS. It makes little sense to suggest, 'I gotta source' is enough reason. -- Alanscottwalker (talk) 21:53, 9 April 2017 (UTC)
Yes, "for many many reasons delineated in the thorough discussion that led to the change" (see #Where verifiability is not an adequate criterion, above). ~ J. Johnson (JJ) (talk) 22:12, 9 April 2017 (UTC)
Mostly yes, only that I feel the wording should be "Verifiability is not the only reason for inclusion." which is stressed more in the explaination but could be misread (without "only") as that we don't consider verifiability as one of the reasons for inclusion. --MASEM (t) 01:05, 10 April 2017 (UTC)

One of the sentiments in the original discussion is that WP:V should not be a reason for inclusion, only a requirement; that once something meets this minimal bar WP:V should have nothing further to say. If something is questioned verification is required. If that can't be done, then it is out, end of story. If it can be verified, that is as far as WP:V goes; inclusion depends on other considerations. ~ J. Johnson (JJ) (talk) 06:53, 10 April 2017 (UTC)

I recognize you're talking the diff between a requirement and a reason, and I fully appreciate that, but I see the current suggested change as something that an inexperienced editor is going to take out of scope since it does not capture this subtle difference. (And yes, the body text covers it more, but keep in mind how inexperienced editors often take one line of policy to justify a point). It would be better that the sentence include this, even something like "Verifiability is a necessary but not sufficient reason for inclusion" or the like. --MASEM (t) 13:17, 10 April 2017 (UTC)

Yes. Many of those who argue for "no" insist on drawing on other reasons why verifiable by high quality sources tends to be worth including in some article. I believe each policy should strictly stick to the goal of the particular policy, and not trespass on the territory of other policies, otherwise we end up with six or seven policies that all say the same thing, and it becomes impossible to change any of them because we have to change all of them at once. If the wording surprises people, good; we should vigorously shove the concept that each policy has its own territory in editors faces. Jc3s5h (talk) 13:56, 10 April 2017 (UTC)

I believe each policy should strictly stick to the goal of the particular policy, and not trespass on the territory of other policies. I could agree with that sentiment. The problem is that the recent change does not achieve that goal; it makes WP:V step in the territory of WP:WEIGHT, describing reasons why some content should or should not be included in an article. I recognize that the change intended to isolate WP:V from reasons for and against inclusion, but the way it's worded provides some reasons not to include content instead. Diego (talk) 14:11, 10 April 2017 (UTC)

No. The Weight reference has been in that paragraph forever (and was never changed), and it's function is to point away from V to NPOV (by link), that is expressly not stepping into NPOV. Alanscottwalker (talk) 14:36, 10 April 2017 (UTC)

(Uh, I don't understand to what paragraph do you refer? "Weight" is mentioned only in the introduction, though there's a See also link to UNDUE at Other issues). With "describing reasons why some content should or should not be included in an article", I refer to the wording "Verifiability is not a reason for inclusion", not any outside link. In any case, Weight links back to WP:V as a way to decide whether some fact has enough weight. Diego (talk) 15:10, 10 April 2017 (UTC)

The paragraph section under discussion has a big Undue (WEIGHT) link right there. The policy language is Verifiability alone is not a reason for inclusion, in other words, you are going to have to have more reason then just "I gotta RS", so look at these other policy and guidelines. Alanscottwalker (talk) 15:28, 10 April 2017 (UTC)

In that case, it should rather say "you have to have X reasons to include it", rather than "Verifiability is not a reason for inclusion". The change is very poorly worded. Diego (talk) 15:37, 10 April 2017 (UTC)

I strongly agree that each policy should stick within it's defined scope. Which is violated if WP:V gets into the other reasons for inclusion (as in "you must have ..."). The intent of the change (aside from any issue of implementation) is to remove WP:V from inclusion debates; that it speaks to (and is used in) only the narrow issue of a minimum requirement for inclusion. ~ J. Johnson (JJ) (talk) 20:06, 11 April 2017 (UTC)

Yes: Makes sense for newbies who are confused about verifiability and creating their first articles. KGirlTrucker81^{huh? what I've been doing} 23:38, 10 April 2017 (UTC)
Yes The language itself may need tweaking, but the previous wording did not stress the distinction between the minimum threshold for inclusion and editorial judgement regarding all other policies concerning inclusion. WP:V is not the be-all and end-all of inclusion. I've seen this mistaken before, particular when trying to include some minor negative material into BLPs of people some editor doesn't like.--v/r - T P 00:25, 14 April 2017 (UTC)
Yes When I remove a line from an article for a reason, even if subjective, it is not an appropriate response to re-add it simply because it is verifiable. "Do not remove sourced information", as has been used in reversion of my edits, is not a good explanation to why content should be kept in an article indefinitely. Reywas92^Talk 00:18, 17 April 2017 (UTC)

But it is appropriate to re-add it when it has been removed without a reason, which is all too common. Reverting a removal of sourced information is an adequate edit when the removal was done without an edit summary nor a comment at the talk page. It is also a good way to prompt the removing editor to explain themselves and explicit the reason they had for deleting the information. Diego (talk) 10:17, 17 April 2017 (UTC)

So given an addition (inclusion) "without a reason", and a subsequent removal (deletion) "without a reason", you are saying favor the inclusion. Which is contrary to WP:V:

[First four points not applicable. See below. ~JJ]

~~"all material must be attributable ..." (nutshell)~~;
~~"Any material that needs a source but does not have one may be removed" (lede)~~;
~~"The burden to demonstrate verifiability lies with the editor who adds or restores material..." (first section; emphasis in the original)~~;
~~"Any material lacking a reliable source directly supporting it may be removed and should not be restored without an inline citation to a reliable source" (following; emphasis added); and~~ not the least:

"The onus to achieve consensus for inclusion is on those seeking to include disputed content (WP:ONUS)."

Is any of that unclear? ~ J. Johnson (JJ) (talk) 22:21, 17 April 2017 (UTC)

Reywas92 was talking about sourced (and thus verifiable) information. You are quoting the policy on unsourced information. It is unclear what relevance your comment has over the comments it replies to. Diego (talk) 08:39, 18 April 2017 (UTC)

Point taken: the first four points hinge on a lack of verifiability. And this discussion is entirely on the inclusion or exclusion of verifiable material. However, the last point still holds: in content disputes consensus has to be obtained for inclusion.

Note that in the particulars you describe – where material has been removed without a reason, which is to say, without an edit summary – I actually favor a revert. But that is on the basis of BRD, not anything to do with WP:V. (I also favor reversion of additions that lack an edit summary.) At any rate, the removal, addition, reversion, or whatever that is done without a reason is irrelevant here, because the issue is whether verifiability IS a "reason" for inclusion. The particulars I envision are after the preliminaries of Boldly adding/removing/whatever verifiable material and Reverting, whether verifiability alone, without any other reason, is sufficient to "guarantee" inclusion. ~ J. Johnson (JJ) (talk) 23:57, 18 April 2017 (UTC)

whether verifiability alone, without any other reason, is sufficient to "guarantee" inclusion. See here, there's again this essential misunderstanding I see in that statement. No one here is defending that verifiability alone, without any other reason, is sufficient to "guarantee" inclusion (not even S Marshall). Yet the change in the policy wording is a gut punch against our position that verifiability is a sign leaning towards including the content, if there are no reasons to exclude it, which was compatible with the old guideline but not the new one.

The problem comes from the insurmountable ambiguity of the words "not a reason", which can mean either not "a guarantee" but also not "support towards". We'd better off if the change had said "not enough reason" or "not a reason sufficient to...", which only rejects the notion that verifiability gives a free pass for inclusion (which nobody argues), and not the idea that verifiability provides some indication of due weight (which I support). Diego (talk) 10:19, 19 April 2017 (UTC)

The problem (as has been addressed elsewhere) is that terminology like "not enough reason" or "not sufficient reason" implies that verifiability is a reason for inclusion (in the sense of "support"). As to whether that should be a free pass, that is the sense of guarantee, as well as the plain meaning of Marshall's own words, which I have quoted elsewhere. But that is an argument we are having elsewhere. Whether "verifiability provides some indication of due weight" would be a good discussion to have, but should be started afresh on its own thread. ~ J. Johnson (JJ) (talk) 02:25, 20 April 2017 (UTC)

Then use a wording that does not endorse that sense of support, but that also don't endorse its nonexistence. At this point, I'd be in favor of removing the sentence as a whole, since now it clearly doesn't express anything that the community as a whole agrees to, nor provides any guidance. I'd rather have the paragraph reduced to what the community actually agrees: Consensus may determine that certain information does not improve an article, and that it should be omitted or presented instead in a different article. The onus to achieve consensus for inclusion is on those seeking to include disputed content. BTW there's a whole world of difference between "providing support" and "being a free pass", and you would do well in recognizing it in order to stop misrepresenting the argument of the other part. Diego (talk) 09:04, 20 April 2017 (UTC)

Yes. It helps fix a not uncommon problem in ideologically contentious articles. Earnest editor A adds widely published, verifiable information that may well be important for everyone to know but that relates only indirectly to the subject. B accordingly removes it. A restores it "per WP:V"... --Dervorguilla (talk) 19:52, 18 April 2017 (UTC)
Yes, for many reasons, but the main one would be that the current summary text better summarizes the text in the box below. LK (talk) 01:54, 22 May 2017 (UTC)
Yes as this seems to be a reoccurring problem, particularly with new editors and POV warriors. Dennis Brown - 2¢ 14:28, 23 May 2017 (UTC)

No

—S Marshall T/C 14:19, 9 April 2017 (UTC)
While not as adamant about this as S Marshall seems to be, I do think the recent change in language was a mistake. The thing is, verifiability actually can be a reason to add information (ie include it) at least initially... The problem is that verifiability often isn't enough for us to keep added verifiable information in an article. There are lots of reasons why we might remove (i.e. exclude) verifiable information from an article. Some of these reasons are outlined in other policies (for example, mentioning it may give undue weight to a fringe view)... some are not based on policy (for example, editors may simply feel that the information is too trivial to be worth mentioning.) In essence, what we want to tell our editors is this... Both adding and removing verifiable information is allowed... both are part of the normal editing process. I think the original "not a guarantee of inclusion" language expressed this concept better than the current "not a reason for exclusion" does. Blueboar (talk) 15:37, 9 April 2017 (UTC)

I think you mistake the issue. It is not that [b]oth adding and removing verifiable information is allowed" – that is given. The issue is more in cases where there "are lots of reasons" to remove, and verifiability is not just "a" reason among others for retention, but is the sole reason put forward for retention. Where there are "lots of reasons" for retention, verifiability is not needed. Where there are lots of reasons against, verifiability alone should not be superior ('guaranteed'). Part of the argument against the term "guarantee" is that it seems more formal, and simply denying any "guarantee" would leave plenty of scope for mischief ~ J. Johnson (JJ) (talk) 22:23, 9 April 2017 (UTC)

That's not what it says. If that's what it's meant to mean, then that's what it should say.—S Marshall T/C 22:29, 9 April 2017 (UTC)

Let me give a hypothetical to show how verifiability can be a reason for inclusion... Suppose an editor remembers reading a bit of information in a book, but hesitates to add it to a relevant article because he can't remember what the book was (ie he can not show that the information is verifiable). He searches high and low... and finally, after perhaps years of searching, he finds the book... and joy of joys, it qualifies as reliable! Now he can finally add the information to the article. Why can he add it? The reason is solely because he can now show that the information is verifiable.

HOWEVER... Once added (due to verifiability), the information might still end up being removed. There are a host of reasons to remove information (even verifiable information). So, while verifiability can be a reason to add (include) information, verifiability does not "protect" information from being removed.

You know, the more I think about this, the more I am thinking that the underlying problem isn't with the words "guarantee" or "reason"... the problem is actually with the word "inclusion". The section under discussion isn't really about inclusion... it's about removal. Perhaps it would be more accurate to rephrase: "Verfifiability on its own does not protect information from being removed." Blueboar (talk) 23:49, 9 April 2017 (UTC)

In your second scenario, it's still not the only reason or even the main reason -- they have presumably read countless verifiable things in their life, they must have some overweening reason other than verifiability for wanting to add that there. -- Alanscottwalker (talk) 00:49, 10 April 2017 (UTC)

Nope... in my scenario the editor read it, thinks the article should include it, but waited until he found a source to verify it. That's it. Lack of Verifiability, was the only reason for the wait and verifiability was the only reason for the addition. I hope you are not trying to argue that he can't add it. Are you saying he shouldn't add it? Blueboar (talk) 01:10, 10 April 2017 (UTC)

No. I am saying that is not the reason - why did your hypothetical editors think it belonged there? That's the reason. Alanscottwalker (talk) 03:09, 10 April 2017 (UTC)

I think you may be assuming a motivation that is not there. In my scenario, the editor reads an article... sees that information he remembers reading about is missing from the article, and thinks the article should mention it ... no more, no less. Blueboar (talk) 10:59, 10 April 2017 (UTC)

Why does the editor think that should go there? Why does that information improve the article? It's not just that a source exists (out of the multitudes of RS or verifiable information, they have known) and because it is not just that a source exists, you are beyond V, as reason. Is it relevant (or due)? Does a knowledge of the subject suggest that it is needed for context? Why? Does the fact that it is missing, render the article somewhat more confusing, or unclear about the topic? Why and how? If it is not added, is the article skewed? How? And it goes on. V alone - a source exists - (or often, at all) is not the answer for reason(s) for it going there. Alanscottwalker (talk) 11:43, 10 April 2017 (UTC)

In your scenario the editor reads it, but waits until he found a source? Ah, surely you have not forgotten WP:SAYWHEREYOUGOTIT. Where he reads it is the source. Perhap he doesn't know how to cite it in proper form, but if he can't point to where he got it, then there is a keen question of just where this comes from. Of course, it appears that sometimes editors see really interesting stuff in non-reliable sources, so they hunt for some barely adequate source to give them cover, but that is going about this backwards. Material for which an editor cannot provide both a reliable source and a reason for inclusion does not belong in the 'pedia. ~ J. Johnson (JJ) (talk) 07:28, 10 April 2017 (UTC)

Read the scenario again... the editor remembers something he read some time ago, but can't remember the source he read it in... so he does exactly the right thing - he searches for the source, and does not add anything until he finds it (in other words, he is following SAYWHEREYOUGOTIT). Blueboar (talk) 10:59, 10 April 2017 (UTC)

Blueboar, I think you are worrying about a non-problem, and missing the real problem. This merely means that wp:ver "steps" aside after wp:ver has been met. Nobody uses wp:ver to exclude wp:verifiable material. But many mis-use the urban legend to try to us meeting wp:ver to force inclusion, usually in POV maneuvers. North8000 (talk) 12:23, 10 April 2017 (UTC)

But this is not what the new wording implies; had the text been something like "after it is satisfied as a minimum requirement, verifiability has no bearing on whether something should be included or not", you might have a point. However, the new wording strongly implies that it doesn't matter that a fact is verifiable, even by very strong references, it should not be included, or at least not until you jump through hoops to explain how it matches every other policy under the sun. Diego (talk) 13:00, 10 April 2017 (UTC)

No. "Verifiability alone" does not imply that. "Very strong sources" is rather meaningless, without answering the 'for what?' The 'for what' answer goes into other reasons (This policy expressly already says in the intro: "Verifiability" is no reason for adding sourced text that makes the article OR, that renders it POV, that makes it CVIO, that makes it a BLP problem by giving 'false light/undue, etc.) "Strong sources" is an NPOV or OR or BLP reason, V requires a single source. - Alanscottwalker (talk) 13:48, 10 April 2017 (UTC)

But the recent change forbids using WP:V as a reason for inclusion even if there are many in-depth reliable sources. If the change had said ""Verifiability is no reason for adding sourced text that makes the article OR, that renders it POV, that makes it CVIO, that makes it a BLP problem by giving 'false light/undue", we wouldn't be having this conversation. Diego (talk) 14:06, 10 April 2017 (UTC)

Plainly false, one of the things in that very paragraph says, among others, is look to weight(due) and points to NPOV policy, making it an NPOV problem and reason (not V).Alanscottwalker (talk) 14:17, 10 April 2017 (UTC)

And per NPOV, the reason why a topic is prominent enough to have due weight is because it is verifiable in several RSs. V points to NPOV, but WEIGHT in NPOV refers back to V. The two policies collaborate, NPOV does not take the full responsibility. Diego (talk) 15:08, 10 April 2017 (UTC)

Thus, you have just stated, "variability alone" is not enough - and that's what the the policy says. Alanscottwalker (talk) 15:17, 10 April 2017 (UTC)

I've said elsewhere that I could live with the "verifiability alone" wording. But the section title says "Verifiability is not a reason for inclusion", not "verifiability alone". I tried to change it to make both appearances consistent with each other, but it was reverted. Diego (talk) 15:22, 10 April 2017 (UTC)

Oh? I see. I agree that the wording of the policy would be better in the title. Alanscottwalker (talk) 15:37, 10 April 2017 (UTC)

I agree with Blueboar. Verifiability certainly is a reason for inclusion, but other policies often override it for exclusion. Sławomir Biały (talk) 11:17, 10 April 2017 (UTC)

And I say it's not a "reason" for inclusion. (Merely a requirement.) Which goes to show that we're all arguing about terms, but we don't yet have consensus on the underlying principles. Even if we could agree to specific terms, we would still have different theories of how it interpret them, so the confusion and arguments would continue unabated. That is why this RfC is fundamentally flawed: it doesn't have an agreed on basis such that we are all starting from the same place. ~ J. Johnson (JJ) (talk) 20:47, 10 April 2017 (UTC)

I don't dispute that you do not find it a reason for inclusion. However, my perspective is that, if something is "verifiable", then all other things being equal, it would be better to include the thing in an encyclopedia than not to. More precisely: if our prior is that the other policies and guidelines are neutral, then I believe it is marginally better to include a verifiable fact than not to include that fact. From my point of view, being verifiable does create a reason for inclusion (although that is a necessary but not sufficient condition of inclusion), and this is in conflict with the way the guideline is currently written. Of course, it is often not the case that other things are equal. But I think that for most mundane facts in the encyclopedia, other things are, in fact, equal. Determining how each individual edit sits with respect to the entire edifice of WP:PAG is simply not a reasonable requirement. Sławomir Biały (talk) 00:17, 11 April 2017 (UTC)

Most verifiable information does not belong in this, or any other, encyclopedia. Every entry in every phone book, the information from every birth, marriage, and death certificate for everyone who is not living, the details of every benchmark maintained by the U.S. National Geodetic Survey, etc., do not belong. The fact that we seldom encounter editors anxious to add such trivia is a credit to our pool of editors, but nonetheless, the vast majority of available verifiable information does not belong. Jc3s5h (talk) 02:28, 11 April 2017 (UTC)

I did not say that "most verifiable information ... belong[s] in this ... encyclopedia". This seems to neglect the very specific prior that other policies and guidelines should be neutral. In particular, the considerations of WP:NOTDIRECTORY already exclude the kind of content that you mistakenly believe that my view condones. Sławomir Biały (talk) 09:27, 11 April 2017 (UTC)

That is one of the key issues here: does "being verifiable ... create a reason for inclusion"? Consider this: in the universe of ALL reliably sourced material, a great deal (such as phone directories) is excluded by various polices (WP:TRIVIA, WP:BLP, etc.). But there is an immense amount of material NOT excluded, much more than could be included in the entirety of Wikipedia. What we are concerned about is where some editor, WITHOUT ANY OTHER REASON, insists that because something is verifiable it MUST be included in THIS (some) article. That WP hasn't burst with such information is because editors do have reasons for not including such stuff. The argument here is that inclusion should be based on thoseDoe reasons, not on mere verifiability alone. ~ J. Johnson (JJ) (talk) 20:49, 11 April 2017 (UTC)

Why not? Wikipedia is not paper. See "#Inc vs Del". Staszek Lem (talk) 21:08, 11 April 2017 (UTC)

Yes, it does create a reason for inclusion. When an editor adds a small fact here or there, for example what is the vapor point of mercury, the primary consideration is not and should not be whether it meets NOT, NPOV, NOR, BLP, or a host of other policies and guidelines. For most of the routine facts that appear in an encyclopedia, the verifiability condition is both necessary and sufficient, without having routine facts legalistically dissected noticeboard style. I am not saying here that other policies do not create a reason for excluding some kinds of content, but verifiability itself most definitely creates a reason for including something. And for the record, I think the Inc vs Del argument is a red herring. I am not an "inclusionist", nor do I believe that any verifiable content should be added to the encyclopedia. But I do think that, if something is not excluded under some other policy or guideline, then verifiability creates a reason for inclusion, absent any reasons for exclusion. It seems like some folks aren't getting that. And if people are saying that WP:V doesn't create a reason for inclusion, I challenge them to find any clear "reason for inclusion" in any policy. It seems to me that our background position is one of inclusion, with exclusion usually being dictated by one or more PAGs. Sławomir Biały (talk) 21:14, 11 April 2017 (UTC)

re red herring - yes you are an inclusionist ("centrist-leaning"), because for you everything is to be included unless there are reasons for exclusion. Staszek Lem (talk) 21:45, 11 April 2017 (UTC)

No, I am not. I actually prefer to delete things that are poorly sourced, and insist on high quality references for things that remain. However, there quite simply are no "criteria for inclusion". In order to have an encyclopedia at all, it is necessary to operate on the presumption that information is worth cataloging and including. If the default assumption is one of exclusion, our guidelines would need to be rewritten entirely to provide guidance on what we should include, but I read them as primarily exclusion. So if our default position were not to include information, and our guidelines further stated certain kinds of things cannot be included, then that leaves no content suitable for an encyclopedia. That is not what deletionists believe: we don't wish to destroy the encyclopedia, we just want our guidelines to be interpreted strictly (e.g., only scholarly sources of the highest quality should be allowed for an "encyclopedia", although that has become quite an unpopular position). Sławomir Biały (talk) 22:22, 11 April 2017 (UTC)

Yes you are. Because you include everything unless it must be excluded because it is poorly sourced. The presumption "that information is worth cataloging and including" is inherently wrong: who decides what is worthy? Personally, I would wipe all pornstars from wikipedia. Of course, there is a wide spectrum of inclusionists, from "extreme left" ("include it because my dad told me it was so") via "far left" ("include every person and every pizza hut outlet into wikipedia") to "moderate left" ("include everything verifiable") to "centrist" ("include all verifiable and notable"). Staszek Lem (talk) 23:42, 11 April 2017 (UTC)

"Yes you are. Because you include everything unless it must be excluded because it is poorly sourced." Wrong. Please check your reading comprehension. If something is well-sourced to scholarly literature (for example, a fact sourced to a standard textbook on quantum chromodynamics), then of course it should be included. I defy you to say otherwise, and if you do I will seriously begin to question your good faith in this discussion, and indeed whether you are here to build an encyclopedia at all. Sławomir Biały (talk) 00:30, 12 April 2017 (UTC)

Slawomir: Seemingly, your "presumption" could only begin to work if we had one single article entitled, "Everything." We do not, we have millions of articles. "Cataloging", requires more reason than, 'I have a source', it also requires reasons on where and when to include and exclude, and if it's worthy of inclusion at all. So: 1) There are millions of articles we will undoubtedly exclude a particular 'fact', even though it is verifiable. 2) There are a comparative handful of articles where we may include that 'fact' but not just because 'a source exists' (ie. not just, it is Verifiable). 3) There are some number of articles where someone might try to include that 'fact' but in the process create, POV, OR, UNDUE, in-context-trivia, or otherwise skew the article; and 5) there are millions of verifiable facts which will NOT be included, at all. Thus, there is not a single article where variability alone, works, and saying it works contradicts all the Wikipedia content policies, including WP:V itself, which in its introduction says you cannot just consider WP:V in isolation for determining article content. -- Alanscottwalker (talk) 00:17, 12 April 2017 (UTC)

I will reiterate that, if the prior is that other policies and guidelines are neutral, then verifiability does create a reason for inclusion. So, if information is not excluded under WP:NOT, WP:NPOV, WP:OR, WP:BLP, then of course verifiability creates a reason for inclusion. To rephrase, if there is no reason in policy at all not to exclude the information, then of course being verifiable creates a reason for inclusion. I am astonished that anyone would seriously suggest otherwise. For example, if a fact is printed in a standard textbook on Quantum Electrodynamics, you are of the opinion that that information should not be included, because we don't have a policy that says that facts from standard physics textbooks are the kinds of things that should be printed in an encyclopedia? Nonsense. The presumption is that such facts do belong in an encyclopedia. And I defy anyone here to seriously suggest otherwise. Sławomir Biały (talk)

Other policies and guidelines are never "neutral", they are all in operation simultaneously. What you mean by 'neutral' appears to be, all the other policies support inclusion, and thus you are far beyond Verfiability alone supporting inclusion. As for that fact on Quantum Dynamics, there is no doubt that fact should not be included in millions of our articles (almost all of our articles, in fact). Alanscottwalker (talk) 00:37, 12 April 2017 (UTC)

I don't think there is anything in conflict here with what I am saying. Verifiability creates a "reason for inclusion", along with neutral point of view, BLP, OR, etc. No one has suggested a binary decision here. But the current wording of the policy is not acceptable: verifiability most certainly is a reason for inclusion. There may be other reasons too, either for inclusion or exclusion. The purpose of the sentence should be to emphasize the need for the policies to work together, not to exclude verifiability from that particular calculus. Furthermore, I still maintain that most of the boring facts that are added to our encyclopedia (e.g., the vapor point of mercury), do not require a complicated calculus of policies to justify their inclusion: verifiability is usually enough. Sławomir Biały (talk) 01:02, 12 April 2017 (UTC)

No, or at least certainly not this way. Regardless the merits for a change, the new wording is a terrible one. For a start, the new section title (not a reason for inclusion) doesn't match the new body description (not alone a reason for inclusion), in a way that makes for different and in some way conflicting meanings; I for one could live with the text in the body, but the section title is unacceptable.

Procedurally, the new wording was achieved with a short discussion that was not publicized, therefore achieving limited feedback that could have got us a better text. Diego (talk) 12:19, 10 April 2017 (UTC)

P.S. Policies and guidelines are descriptive, not prescriptive, and their purpose is to explain the principles and best-agreed practices of the community; the new wording imposes constraints based on a principle that a large amount of editors don't agree to and don't regularly follow, so it is not a description of agreed upon practices. We'd be better off reverting to the previous version, unless we can agree to a new improved wording thanks to this RfC. Diego (talk) 12:59, 18 April 2017 (UTC)

P.P.S. The original discusion was a local consensus, without an attempt to advertise the changes beyond this page or open a formal process. WP:CONLEVEL says that "Wikipedia has a higher standard of participation and consensus for changes to policies and guidelines than to other types of pages. [...] Changes may be made without prior discussion, but they are subject to a high level of scrutiny. The community is more likely to accept edits to policy if they are made slowly and conservatively, with active efforts to seek out input and agreement from others."

This high level of scrutiny didn't happen, therefore the limited consensus achieved for the change is not enough to gain the "stability and consistency" required at policy text. Diego (talk) 12:17, 20 April 2017 (UTC)

No While it is certainly not the reason, it is a reason. While likely unintentional, the new wording suggests a false equivalence between verifiable and unverifiable information. – Philosopher ^{Let us reason together.} 12:47, 10 April 2017 (UTC)

Very much ^this.

The people that suggested the new wording so far seem to fail to acknowledge this important point; while I don't see anyone rejecting most of the reasons for the change, the new wording has so many unintended and problematic implications that it should be changed to something more agreeable. Diego (talk) 12:56, 10 April 2017 (UTC)

I agree with user:Sławomir Biały's statement, and think that the unaltered wording was clearer. -- PBS (talk) 12:54, 10 April 2017 (UTC)
No. The suggested section title may be confusing, at least for people with limited command of English an/or logic. The function/purpose of "Verifiablitity" may be may be described variously; e.g., IMO the descriptor "confirmation" IMO is more appropriate, rather than "reason", because "WO:V" is nearly always a secondary step: Someone adds a piece of info not because they think it is verifiable, but because they think it is useful for encyclopedia. That said, from the discussion I have an impression that the actual purpose of this statement is 'Verifiablity cannot be the only reason for inclusion", to which is the change in the text body actually approaches: "verifiability alone is not a reason for inclusion". And this is why I once reverted this change for being self-controversial: the crucial word here is "alone", which is what many 'ay'-!voters say, but they fail to see the text body contradicts the title. Staszek Lem (talk) 16:32, 10 April 2017 (UTC)
No. This change would affect a lot of articles. Also, this would lead to more endless debates on any content. ~~Worse would be more AfDs, especially some failed ones.~~ George Ho (talk) 18:21, 10 April 2017 (UTC)

George, the result that seems to concern you depends very much on how you interpret WP:V, and the details of the situations it is applied to. This is where examples are needed, but in the preceding discussion no examples were presented where deletion or retention came down to verifiability alone. How this could affect an entire article is quite unclear (examples?), and I think quite unlikely. ~ J. Johnson (JJ) (talk) 21:46, 10 April 2017 (UTC)

Sorry, I just got emotional about the AfD, so I'll strike that out. Examples I can think are Cold War II (whose topic is very contentious) and some other biographies and other political topics. --George Ho (talk) 21:55, 10 April 2017 (UTC)

I understand. Do you perchance have any examples where verifiability alone, lacking any other consideration, was the issue in whether to include or exclude any material? ~ J. Johnson (JJ) (talk) 19:44, 11 April 2017 (UTC)

I can think current events, like United Airlines Flight 3411 (not an accident or crash actually), which is currently AfD-ed, which NOT#NEWS is cited for deletion. Also, Vince (rhinoceros) was nominated for deletion but then kept due to improvements made. George Ho (talk) 02:20, 12 April 2017 (UTC)

At the Flight 3411 AfD verifiability was not raised at all, and WP:V mentioned only as a qualification passed by numerous articles (thus establishing notability). There was no issue of WP:V at Articles_for_deletion/Vince_. Both articles were "keep" without needing any help form WP:V. ~ J. Johnson (JJ) (talk) 21:10, 12 April 2017 (UTC)

Other than notability concerns, I can't think of any other issues concerning inclusion based on verifiability alone. That's all. George Ho (talk) 21:27, 12 April 2017 (UTC)

Which is par for the course. Several editors have expressed concern that broad swaths of text, even whole articles, might be deleted on the basis of the current language, but examples have not panned out. Cases where people vaguely recall some element of verifiability turn out to hinge on other issues, and it appears this change does not "affect a lot of articles." ~ J. Johnson (JJ) (talk) 21:28, 13 April 2017 (UTC)

It applies to every article where an unexplained deletion has been reverted with the comment "don't remove verifiable content without explanation", which North8000 above agreed is a common occurence. (While North8000 calls this a "false claim", many of us look at it as a perfectly valid practice that the new wording forbids, thus the disagreement). Diego (talk) 13:04, 18 April 2017 (UTC)

George said "[t]his change would affect a lot of articles", I asked for examples, and it turns out that the ones he had in mind did not actually involve verifiability. It is not clear what you are referring to in "a perfectly valid practice that the new wording forbids," If you are talking about using verifiability for retaining material for which there is no other reason for retaining, well, we have been over that elsewhere. This sub-thread (commenting on George's "no") is not the proper place for zinging off into another direction. ~ J. Johnson (JJ) (talk) 00:50, 19 April 2017 (UTC)

I was replying mostly to your assertion "it appears this change does not affect a lot of articles", with an example of how it does. I agree it is tangential to your request for examples of large discussions where verifiability alone has played a primary role, though this illustrates that those examples are not the only possible way by which articles are affected by the policy change. Diego (talk) 10:23, 19 April 2017 (UTC) P.S By "a perfectly valid practice" I refer to reverting unexplained removals with a comment like "restore verifiable information", which North8000 dislikes, and with the new wording is "not a reason" for the revert. Diego (talk) 10:29, 19 April 2017 (UTC)

As to other possible ways "by which articles are affected", I don't recall that anyone has suggested any, even hypothetically. As to the reversion of unexplained deletions, WP:BRD would be the better basis. And I would note that one implication of WP:V is that ALL material added (or re-added) is implicitly verifiable, so a rationale like "restore verifiable information" is actually redundant. As I have been saying, it is not necessary to invoke verifiability for restoring "unexplained removals". ~ J. Johnson (JJ) (talk) 02:37, 20 April 2017 (UTC)

The old wording was better than the current wording. Verifiability is a reason for inclusion. Several examples have been given above, and I do not find the attempted refutations of those examples convincing. Verifiability is not the only reason for inclusion, and it can be trumped by reasons for exclusion, but it remains a reason for inclusion. The policy should not flatly say that it is not a reason for inclusion in the section header, nor in the lead. “Verifiability does not guarantee inclusion” is probably the best wording in terms of accuracy, striking the right balance for weighing verifiability, and elegance, but I am open to other qualified statements such as “verifiability is not sufficient for inclusion”.--Wikimedes (talk) 22:17, 19 April 2017 (UTC)

No -- the prior version is less wordy and clearer. K.e.coffman (talk) 02:42, 20 April 2017 (UTC)
No - The phrase "is not a reason for inclusion" is inaccurate and misleading and counter to what WP:V actually is about. It not only is a reason to include content, but it's actually the first, most fundamental criterion for inclusion we have. Changing it was a horrible idea. Swarm ♠ 03:27, 20 April 2017 (UTC)

This is why a number of editors feel that the correct term is requirement, not reason. DonIago (talk) 03:47, 20 April 2017 (UTC)

I think some of the argument is whether that should be correct term. But either way, that is a critical distinction that seems not entirely appreciated. ~ J. Johnson (JJ) (talk) 20:00, 20 April 2017 (UTC)

No I'll admit it took me a while to parse out what this discussion was about. That said, I think the prior wording is clearer, though I think I largely am okay with the ideas behind either. Hobit (talk) 01:34, 27 April 2017 (UTC)

No. Per Bluboar, and the wording itself in the change is simply put, less precise. Guarantee does mean the same as no reason. There is reason to include verifiable content that's why we have a verifiable guide, but there may be not be a guarantee that content qualifies further for inclusion per other policies and guidelines. Our policies and guidelines work in consort with all other policies and guidelines so there is never(almost) a definitive no. I prefer to have agreement before changes are made to policies.(Littleolive oil (talk) 18:04, 1 May 2017 (UTC))

Making the change was not a good idea

Making the change was not a good idea. No consensus existed for the change: The proposal for wording change on the table in Wikipedia_talk:Verifiability/Archive_65#Proposal was:

Original Proposal: following the third sentence of the lead ("Even if you're sure something is true, it must be verifiable before you can add it."), this text: "However, being verifiable is not sufficient for including something."

This had support and was also considered to be minor enough not to require broader community input.

Supporting: J. Johnson (support inferred as proposer), BrigtRoundCircle, Kmhkmh (with minor alteration), Staszek Lem, North8000, Blueboar, Alanscottwalker, K.e.coffman

Opposing: Someguy1221 (with comment specifically opposing another part of the proposal not related to wording change),

Uneasy about the change: WhatamIdoing.

Two amended changes were then put forward:

Proposal 2a: Change title: "Verifiability ~~does not guarantee~~ alone is not sufficient for inclusion

Proposal 2b: Change first sentence: "While ~~information~~ content must be verifiable in order to be included in an article, verifiability alone is not sufficient for inclusion, and does not guarantee that any content ~~this does not mean that all verifiable information~~ must be included in an article."

Supporting: J. Johnson (support inferred as proposer), BrightRoundCircle, North8000.

J.J. Johnson then made [1] this proposal (in a threaded conversation):

               Okay. How about if my proposed addition is replaced with: "However, while verifiability is required for including something, it is not a reason for inclusion."

               For my additional proposals we can replace "not sufficient for" with "not a reason for". (Though in the text I'm still inclined towards "not a sufficient reason for".) I think most everyone here is generally agreeable, and not going to sweat the details. If you and I agree on the details I'll make those changes.

North8000 strongly supported. [2]

J. Johnson made the change.[3]

Total number of supporters out of ten participants: 2. Total time elapsed from proposal to change: about 13 hours. There was both retroactive support for and opposition to the change. However, making a change under discussion without consensus was not a good idea. The change will not break the encyclopedia, but User:J._Johnson: please put it back the way it was until consensus exists for the change.--Wikimedes (talk) 22:46, 19 April 2017 (UTC)

Excellent analysis of this fly-by-night change to the lead of Wikipedia's most fundamental policy. Sławomir Biały (talk) 22:52, 19 April 2017 (UTC)

Bullshit. And a gross misrepresentation of the situation. The discussion leading to the change grew out of a comment by Kmhkmh on 02:10, 8 Feb. ("Verifiability alone is not (and never was) a sufficient criteria for mandatory coverage ....") (see #Where verifiability is not an adequate criterion, above), and went through various proposals, discussion, and amendments over 20 days. There being consensus for the change in the general sense, North8000 and I worked out the final details with the tacit approval of the rest. Your characterization of this as a quick trick by only two editors is not only flat out wrong, it manifests bad faith. ~ J. Johnson (JJ) (talk) 03:25, 20 April 2017 (UTC)

I did not characterize, nor do I think of it as a quick trick. The fact remains that the change you made was not the change that had consensus. You might read again the post by Kmhkmh that you linked; it does not say that verifiability is not a reason for inclusion, it says that verifiability alone is not a sufficient reason for inclusion. I have read most of your posts above (not all below) and I believe that you believe that this is the same thing as saying that verifiability is not at all a reason for inclusion, but I hope the discussions over the past several weeks have convinced you that many editors disagree with this equivalence.

As for tacit approval: 13 hrs is not enough time to gauge tacit approval. The fact that several participants in the discussion have since disagreed with your change indicates that the approval was misjudged. Such a misjudgement is easy to make after a long discussion on a topic one feels strongly about, especially if some of the other editors don't follow the same logic that you do.

I'm OK with tacit approval in principle: In the "original proposal" I mentioned above you did the right thing and asked if a formal RfC was necessary. One person said that it was not, and the other respondents did not mention it. You had tacit approval for making a change without a formal RfC. Latecomers might have disagreed with the change, or the fact that it was made without wider community input, and I might have been one of them, but I would not have faulted you for making the change.--Wikimedes (talk) 07:31, 20 April 2017 (UTC)

What a fine piece of wikilawyering. If we had had a full-on RfC, running a couple of months with dozens of editors, resulting in a specific block of text, and if someone being directed to make that change noticed there was some kind of trivial error, and said: "Oh, it looks like a small detail needs to be corrected; it looks uncontroversial, but I'll wait a couple of hours in case anyone objects", and then, no objections being raised, and making the change: would you then object to that change as being made by one person with only a "couple" of hours of notice? In fact we had 20 days of discussion that considered and then narrowed the scope of what we were trying to achieve, with tacit approval for North8000 and I to work out the final details. No one has complained, either then or since, that we went beyond that scope. For you to characterize the final result as "13 hrs" is totally superficial, and such a blatant mischaracterization as amounts to a knowing and willful deceit. Ten editors participated in the original discussion, out of some 2,000 page watchers who presumably were aware of that discussion, and my recollection is that there were only two dissents. That sure looks like consensus.

Which all is actually rather beside the point. While Marshall originally complained of the process by which this change was done, in this RfC he asks only if the idea is good (or not), not if making the change was good or not. So it appears you are off-topic. ~ J. Johnson (JJ) (talk) 19:55, 20 April 2017 (UTC)

Ouch. Well, you've stated your position twice and I've stated mine twice, and we can leave it there. We still have one unaddressed point of disagreement over whether the change procedure is within the scope of this RfC; I think that although of secondary importance to the content of the change, it is in scope. But we're unlikely to convince each other, and I'm OK with your refactoring, so I'll leave it at that.--Wikimedes (talk) 21:45, 20 April 2017 (UTC)

I am not okay with your gross misrepresentation. Your "13 hours" characterization is not a mere difference of viewpoint, nor an "alternative fact", it is flat out wrong.

As to the general question of process: note that I am not saying that it cannot be raised; I am saying only that it is not within the scope of this RfC as formulated. Which could be taken as a fault of the RfC for having inadequate scope. ~ J. Johnson (JJ) (talk) 21:36, 21 April 2017 (UTC)

Discussion

This diff is germane.—S Marshall T/C 22:17, 9 April 2017 (UTC)

Hardly. That diff is your comment at WP:ANI#User:Spacecowboy420 blanking articles [now archived], where you make a distorted and faulty interpretation that no one else buys into. It seems like you are trying to generate an issue there (where I don't believe there was any issue of WP:V) which you can cite here. It rather seems like you are citing yourself as an example of how people can misinterpret WP:V. ~ J. Johnson (JJ) (talk) 23:28, 9 April 2017 (UTC)

I have noticed something... the policy uses the term "inclusion" (and thus, through omission, its opposite: "exclusion")... but in our discussions most of us keep using the terms "retention" and "removal". What is the difference? ... retention/removal is a more specific sub-set of inclusion/exclusion that occurs only after information has already been included. Is that, perhaps, the root of the disconnect here? We are not actually focused on inclusion, but on removal? Blueboar (talk) 00:51, 10 April 2017 (UTC)

It is unclear what we are focused on here because a certain editor has generally failed to make those distinctions. ~ J. Johnson (JJ) (talk) 07:56, 10 April 2017 (UTC)

My observation was based on the comments by multiple editors (including my own)... not those of any one particular editor. Blueboar (talk) 10:32, 10 April 2017 (UTC)

Likewise. None of us are perfect in using terms and concepts precisely and consistently. It's when these are left ambiguous, or even unacknowledged, from the start that we keep going around and around and .... ~ J. Johnson (JJ) (talk) 20:35, 10 April 2017 (UTC)

The title of the RfC section should be changed. The "yes" and "no" sections are inverted relative to the title question. Sławomir Biały (talk) 11:19, 10 April 2017 (UTC)

I've changed the title to avoid this problem, it no longer is a yes/no question itself. Diego (talk) 13:33, 10 April 2017 (UTC)

RfC design discussion

Gotta say I'm surprised at the neutrality of the RfC, I was expecting it to be biased and was pleasantly surprised. However, "was this a good idea" is the canonical bad way to present an RfC. The all-or-nothing approach is the very reason the discussion above is stymied. A better approach would be tackling the issues one at a time (like the original discussion that led to the change) instead of swallowing it whole as "good idea"/"bad idea". Bright☀ 14:27, 9 April 2017 (UTC)

If you can come up with a third option that would satisfy everyone, please suggest it. Blueboar (talk) 15:39, 9 April 2017 (UTC)

"Satisfy everyone" is a tall order, but I've already suggested tackling the changes one by one (as they were originally tackled) instead of all-or-nothing. Anyway this RfC is already rolling so let's just roll with it. Bright☀ 20:51, 9 April 2017 (UTC)

The problem is not just in being presented in a simplistic either-or form, but (independent of the number of options) in the failure to provide any background or other information to give the respondents any basis for an informed decision. There are reasons why this change was made (and it was both well-considered and consensual), but Marshall has not even provided a link to that prior discussion. There is also question of whether this is a good time for this RfC, as the last month's contentious and unproductive discussion on this has led to discussion fatigue, so any extended discussion on this now will likely be of sub-standard caliber. ~ J. Johnson (JJ) (talk) 22:15, 9 April 2017 (UTC)

Well, It is fine SMarshall raises an RfC, and feel free to link to that prior discussion. Alanscottwalker (talk) 22:19, 9 April 2017 (UTC)

Let him clean up his own mess. ~ J. Johnson (JJ) (talk) 07:05, 10 April 2017 (UTC)

S Marshall, since you initiated this RfC, please could you fix the RfC opening statement in two ways: (i) to not include that table; (ii) so that a signature (or a valid timestamp, at the very least) appears before the "Yes" section. To see why, have a look at how the RfC appears at Wikipedia:Requests for comment/Wikipedia policies and guidelines (Permalink). Both of these are also explained at WP:RFC. --Redrose64 🌹 (talk) 22:48, 9 April 2017 (UTC)

I've added the timestamp. I don't see a provision prohibiting tables at WP:RFC, and I note that it's one of the recommended options at Wikipedia:Requests for comment/Example formatting. However, you're welcome to fix it so that it displays correctly at Wikipedia:Requests for comment/Wikipedia policies and guidelines.—S Marshall T/C 23:17, 9 April 2017 (UTC)

It's in WP:RFC#Statement should be neutral and brief (which was linked by BrightR (talk · contribs) above): "For technical reasons, statements may not contain tables or complex formatting, although these may be added after the initial statement (i.e., after the first date stamp)." If you follow the permalink that I provided, you'll see that nothing is displayed after the opening brace that starts the table. --Redrose64 🌹 (talk) 23:22, 9 April 2017 (UTC)

I've moved the signature above the table as requested. If this doesn't fix it, feel free to make whatever change you need to resolve the transclusion issue. My wikimarkup-fu is weak, and you admins are supposed to be good at this. :)—S Marshall T/C 23:29, 9 April 2017 (UTC)

While verifiability is required for inclusion it is not the only thing required for inclusion (content must also be notable). Doc James (talk · contribs · email) 23:35, 9 April 2017 (UTC)

Actually, Doc... according to our WP:NOTABILITY guideline, content does not need to be notable... article topics need to be notable. Perhaps what you mean is "content must also be worth noting". (it is a subtle, but important distinction). Blueboar (talk) 01:00, 10 April 2017 (UTC)

@S Marshall: that this novel wording requires community consensus and in the absence of consensus, should be reverted to the stable version. There was no absence of consensus, you are wording it backwards. There was consensus for the current version, and you objected on the basis "I missed the discussion", which is not a valid reason. This has already been pointed out to you. This RfC is about achieving a new consensus, as the current consensus already favors the change, even if you weren't there for the discussion. As it stands, from what I see, there is no consensus for the revert, and the current version is preferred anyway. Bright☀ 19:28, 19 April 2017 (UTC)

I propose that the guideline should say what I think people mean when they say "Verifiability is not a reason for inclusion" in this discussion. What they apparently mean is that "Verifiability alone is not a sufficient reason for inclusion." Is that an acceptable alternative to the current (unclear and literally wrong) wording? Sławomir Biały (talk) 19:55, 19 April 2017 (UTC)

Of course, I entirely disagree with BrightR, and of course, I join issue with him. I understand why his position is that I'm challenging a consensus that was established after due process, but I see his position as not closely aligned with the facts. My position is that this was a major edit to a policy; that was made without the RfC that should normally accompany a major edit to the policy; and I challenged it reasonably promptly after it was made. I think custom and practice is on my side here. I also concur with Blueboar when he says we need to let some other people have their say, as we're mostly repeating ourselves now.—S Marshall T/C 20:26, 19 April 2017 (UTC)

It was felt that the change was not "a major edit to policy", but a clarification. The perceived significance of the change depends on various and differing interpretations of the wording, the policy, and the effects. A major defect of this RfC (and why it has become such a muddle) is in failing to identify the caveats on which it is based, to the result that we still don't have any common understanding of just what we are discussing. ~ J. Johnson (JJ) (talk) 03:30, 20 April 2017 (UTC)

While I feel J. has allowed personal feelings to interfere with some of his postings on this matter (no offense intended), I strongly agree with the last part of this sentence. This RfC appears to be tackling multiple issues with multiple interpretations simultaneously, to the point where I don't even know how it could reasonably be closed other than "no consensus for any change to current wording" unless we rein things in and more clearly elucidate both terminology and intent. DonIago (talk) 03:56, 20 April 2017 (UTC)

On the contrary, the RfC is a very simple Yes/No question. There was a change made to a core policy through a WP:LOCALCONSENSUS, that was challenged just after. The process here should determine whether there's community-wide support for that change in the form of approval through a RfC. While it's true that a "no consensus" outcome would not be able to change the wording, it is clear that it would make it unenforceable and useless as guidance (since "editing a policy/guideline/essay page does not in itself imply an immediate change to accepted practice").

If a single thing is obvious at all in this discussion, it is that the change does not have the "high level of consensus" required by WP:PROPOSAL for new rules in policies and guidelines. The proponents of the change failed to follow the "Good practice for proposals", which is to "create a proposal that has a solid chance of success with the broader community, [then] start an RfC for your policy", and at WP:PGCHANGE "major changes should also be publicized to the community in general"; and now are fighting tooth and nail to keep their entrenched version against legitimate opposition. There is nothing good coming from that attitude. Diego (talk) 09:58, 20 April 2017 (UTC)

I disagree on several levels. First, the wording of the RFC is what it is (flaws and all), not new things created in this talk area. Next, the current wording (instituted March 2nd) merely accurately summarizes the actual policy, it does not change it. Third, the current wording was discussed with 27,000+ words for 23 days (Feb 27th - March 2nd) and instituted March second (at the time with no objections) and is the current wording; phrasing it as a "proposed change" is not correct. Sincerely, North8000 (talk) 10:21, 20 April 2017 (UTC)

the current wording (instituted March 2nd) merely accurately summarizes the actual policy The idea that the wording is an accurate summary of the policy has direct opposition in this RfC.

it does not change it The idea that the wording is merely a minor change or no change at all has direct opposition in this RfC.

the current wording was discussed with 27,000+ words for 23 days - by a small amount of editors, with no announcement to the wider community; and as described by Wikimedes above, the current wording was supported by a total of two editors, with several other editors agreeing to different wordings, not this particular one. Therefore, the idea that the previous discussion was enough level of consensus to satisfy the requirements for a change in policy has direct opposition in this RfC, when seen in the light of WP:PGCHANGE.

A honest good-faith attitude would be to acknowledge that the changes have failed to gain the wide support required by policy, agree to change to a neutral version (either the previous long-standing one or a full removal of the disputed sentence), and work towards finding a new wording acceptable to roughly all editors participating in the RfC. Diego (talk) 11:09, 20 April 2017 (UTC)

It does not really matter whether the February change had a proper consensus or not... because WP:Consensus can change... Even if we say that there was a consensus for the "Reason for inclusion" wording back in February, what is clear now (in April) is that the "Reason for inclusion" wording no longer has consensus. What is not clear yet is whether we should return to the old "not a guarantee of inclusion" wording, or go with some other (new) wording. That may take a follow up RFC. Blueboar (talk) 13:19, 20 April 2017 (UTC)

Having not commented here for almost a week, it is diappointing there has not been movement toward a compromise, such as the "Alternative" below. There is much agreement among the 'sides', but it keeps coming back to extended disagreement, unfortunately. Alanscottwalker (talk) 13:37, 20 April 2017 (UTC)

I agree that consensus can change. But note the second sentence there: "On the other hand, proposing to change a recent consensus can be disruptive.". (Especially when poorly thought-out.) Note especially that this RfC does not propose to change any consensus, nor any language; it asks only if the change was a good idea. If it was, then presumably there is no need to undo anything. If not, then presumably we might proceed with what to do, but that is not actually addressed here.

My deep objection here is not to possibly changing anything, but to any attempt to force a change based on such wretched discussion, where we still do not have any definition of terms, identification of concerns (let alone any examples of those concerns) or relevant sub-issues, and certainly not any statement of how the issue is to be resolved and what should ensue. This RfC is defective, which is the main reason why the discussion has been so disruptive; it would be an extremely poor justification for any result. ~ J. Johnson (JJ) (talk) 20:36, 20 April 2017 (UTC)

Blueboar: This RfC is worded in the canonical bad way to present an RfC. "Was this a good idea? Please discuss" and all-or-nothing approaches almost never lead to consensus. The lack of consensus on this RfC does not nullify the consensus of the previous discussion. A new consensus needs to be reached, not "no consensus". If and when this RfC reaches its "no consensus" conclusion, it should not lead to a revert, but to further discussion, hopefully after a break. Bright☀ 11:55, 21 April 2017 (UTC)

Which is why I said that a follow up RFC will be necessary. My take on S Marshall's question is that he is essentially asking "Does the current language (still) have consensus?" And looking at the replies, it is clear to me that the answer is "no, the current language does not have consensus"... However, having answered that yes/no question, we are left with another question: What next? I agree that we have yet to reach consensus on what to do next. Should we simply revert to the old text? Should we reword to something new (and if so, what)? Should we remain silent on the issue? The question of what to do next still needs to be addressed, but it will take a follow up RFC to address it. Blueboar (talk) 12:21, 21 April 2017 (UTC)

That is undoubtedly Marshall's view, but it is not what he asked. And to grant credence to that which is not stated is to blind us as to what the issue is. It would be like asking a jury for verdict, and only afterwards telling them what their decision applies to.

As to any lack of consensus: the thrust of all these discussions is to show a loss of support for the prior language. At least not as some people implicitly interpret it. To resolve that will take a broader (and better structured) discussion. Unfortunately, the ground is so chewed up now that any discussion that is simply an extension of this one is just going to get mired. What is needed next is break – a rhetorical fire break — so that the next discussion can proceed on its own terms, without carrying over the deep flaws and bad feelings of this discussion. ~ J. Johnson (JJ) (talk) 21:45, 21 April 2017 (UTC).

I'll be gone and off the grid & wiki for 4 days. Everyone here is simply doing what they think is best for the good of Wikipedia. I think that the current wording is clearly the best of all considered but am open to better ideas or a compromise.North8000 (talk) 12:34, 20 April 2017 (UTC)

Moving forward

If this RfC ends with "no consensus", what then?

There was previous consensus for a change (in a discussion spanning about 20 days, which three weeks later was contested).
The policy WP:PGCHANGE suggests (but doesn't explicitly state) that the existing consensus was reached correctly, since the result of the discussion was clear.
The policy WP:NOCON suggests (but doesn't explicitly state) that when there's no consensus, no action should be taken except in the specified cases.
The policy WP:EDIT favors "existing consensus" or "present consensus", and does not mention "static version"/"long-standing version"/"stable version".

The RfC is interpreted by its proposer in a manner inconsistent with these policies. Even if the previous consensus does not reflect current consensus (it's certainly possible), a "no consensus" result does not mean a revert to the "previous-previous" consensus. Bright☀ 12:55, 21 April 2017 (UTC)

This is the most fundamental policy on Wikipedia. It is important that every aspect of it have solid consensus. As already observed, the consensus for the original edit in question was not particularly solid. It was not advertised to the wider community in an RfC, and it changed this policy in a significant and fundamental way. The old consensus wording should be restored. If there truly is solid consensus for the new wording, then that consensus should be demonstrated in a request for comment in which the entire Wikipedia community is invited to participate. Sławomir Biały (talk) 16:34, 21 April 2017 (UTC)

"What then" is not up to any of us. We need to let the closer deal with that. I have suggested before that this RfC should be closed by a triumvirate of experienced admins, and I would like to repeat that suggestion now.—S Marshall T/C 17:12, 21 April 2017 (UTC)

Sławomir Biały, I've noted above that there's at least one fundamental aspect of this policy ("Consensus may determine [...] include disputed content") which was never discussed, before or after it was put in the policy. It's been inserted, reverted, and then inserted again, with consensus through silence. Claiming that the 20-day discussion is any less "solid" than the discussion for most other parts of this policy is disingenuous. There was more discussion about it than many other parts of the policy, including the wording of that section before that discussion (which lasted for less time and included less participants). By your reasoning, the old consensus needs to be rejected in favor of something that was RfCd... but this section was never RfCd, the most thorough discussion it had was the one you're contesting. Feels... disingenuous. Bright☀ 13:19, 22 April 2017 (UTC)

As S Marshall correctly points out, what to do next will depend on what the closers have to say to us in their closure. Obviously, those of us who support the current language want them to say "Yes, the current language had consensus, and thus should be kept", while those who support the older language want them to say "No, the current language did not have consensus, and thus the policy should return to the older language"... but despite the yes/no wording of the RFC, the closers have more options available to them than just "yes" or "no". They might tell us that neither has consensus, and that we all need to compromise and find some new consensus language (which will require a follow up RFC). Or, they may tell us to do something that none of us has thought of yet. In other words... it is premature to ask "what next". We need to have some patience... we won't know what to do next until we hear what the closers have to say.

That said, given the heated nature of the exchanges so far, I fully endorse S Marshall's call for a three admin closure. Blueboar (talk) 14:06, 22 April 2017 (UTC)

The proposed wording of the new second sentence of the guideline was made in this post to WT:V, and then it was implemented 13 hours later. There was a discussion that preceded that edit, with a number of perhaps better options, but given the lack of community support now that this has been subjected to an RfC, the only solution is to go back to the previous stable RfC version. The lead of the WP:V was the subject of this formal RfC. If the community wants to change the WP:V policy in a fundamental way as to exclude verifiability as a possible reason for including information in an article, then it should also be through the RfC process, not an apparent local consensus among various options only one of which is the current wording. Indeed, it is actually difficult to find any discussion of the present wording of the guideline, let alone find a place where to !vote against its inclusion. In particular, in that discussion I do not see much that I would claim is "consensus" for the wording that "Verifiability is not a reason for inclusion", which is what the current policy says. Perhaps this is all obvious to those few editors who were involved in that discussion, but if an outsider can't even find it, I find it hard to swallow that it represents any kind of Wikipedia-wide consensus in the way that the result of the previous RfC can claim to be. So, in conclusion, without a similarly strong consensus for this radical change to the policy, I do not see how a 13 hour period between a proposal and edit can possibly be placed on equal footing with the result of a widely-publicized RfC that led to the former stable version of the guideline, particularly not now that the changes are contested by quite a few editors. I think the only reasonable solution is to go back to the stable version. If there is consensus for a change to the policy from the one established at that RfC, then it should demonstrate a similarly high level of support in which the whole community is invited to participate, not a 13 hour period between proposal and implementation based on a (fairly weak) local consensus only. Hopefully this clarifies my position, and I have no doubt that a panel of administrators will unanimously agree with this assessment of the situation. Sławomir Biały (talk) 14:34, 22 April 2017 (UTC)

The RfC you point to is about the lead, not WP:ONUS. There is no "previous stable RfC version" for WP:ONUS. The wording in WP:ONUS was created in this discussion, not an RfC, where the wording was this, which slowly changed to this, without an RfC, and the last sentence without even as much as discussion, only consensus through silence. If you seriously suggest all consensus other than RfC is invalid, there is no reason to revert to that particular version, but to delete the section entirely. However RfC is not the only valid consensus, and a discussion among ten people is larger than many discussions for any previous similar alterations (some of them without discussion at all!). The claim that this is a "previous stable RfC version" is disingenuous, to put it mildly. Bright☀ 15:19, 22 April 2017 (UTC)

I am not being disingenuous, and remind you please to assume good faith. This RFC discussion concerns changes made to the lead of the verifiability policy. There is a very clear diff of that change in the statement of the RfC. I do not believe that there was adequate consensus for that change, to override a previous RfC on the lead. The fact that this RfC fails to demonstrate any consensus for it merely shows that there was none to begin with. Our default position should therefore be to revert to the previous stable version if the lead until a new (and much clearer consensus) forms, probably as a result of a new RfC, since stating that verifiability is not a factor in determining whether something is suitable for Wikipedia seems to be a fundamental change that would require broad community-wide support if it to be implemented. Sławomir Biały (talk) 15:55, 22 April 2017 (UTC)

~~{{hat}}~~

You are being disingenuous. And also rather one-sided where you ask others to assume good faith even though you do not do so your self (e.g., here). I will remind you that under Behavior that is unacceptable in the WP:Talk page guidelines "No personal attacks" is followed by "Misrepresentation of other people, which you also continue to do. Your comments do not further any resolution of this matter, and violate WP:TPG.

I note in passing that your questioning of the process by which the text was changed is actually off-topic. As I noted above, this RfC only asks if the idea is good, not if making the change was good. ~ J. Johnson (JJ) (talk) 20:57, 22 April 2017 (UTC)

Do you claim to be special, that you can not only insinuate bad faith in others and misrepresent them, but you can freely judge the relevance of, and hat, comments you don't like? (As you just did to mine.) Are the rest of us allowed to play by these new standards you seem intent on setting? If so, then I should be allowed to hat some of your comments, right? ~ J. Johnson (JJ) (talk) 23:12, 22 April 2017 (UTC)

I don't think this even merits a reply. But I will say that I would not expect to continue to receive aggressive replies from someone who already wrote me off as too incompetent to be worth spending any more time on. I shouldn't think that what such a person writes would matter in the least to you. Anyway, I think from the beginning of your post in this section, this discussion should be hatted as completely off topic. I know I'm certainly done caring what you think. Sławomir Biały (talk) 01:32, 23 April 2017 (UTC)

I did indeed wonder (20 Apr.) if it might be beyond your competence to "just line up the original with what I quoted and show us the difference" (this in response to your accusation that I had misquoted you). A simple demonstration could easily settle that, perhaps even the allegation of misquotation; the lack of same speaks for itself. As to taking offense: you set the standard with your imputation on the 14th of "either a WP:CIR failure on your part, or is just bad faith trolling." If you play rough expect to get bumped. ~ J. Johnson (JJ) (talk) 18:47, 23 April 2017 (UTC)

I'm unclear why you persist in talking about quoting me out of context, since it seems to bear no relation to the subject under discussion. Sławomir Biały (talk) 21:35, 23 April 2017 (UTC)

The discussion was about the disingenuousness of your argument. Following your lead we are now talking about your tendency to question the good faith and competence of other editors when you don't like their comments. Is that clear enough? ~ J. Johnson (JJ) (talk) 20:43, 24 April 2017 (UTC)

I did not question the competence of BrightR. I did question your competence after repeatedly bringing up the same refuted misrepresentation, and quoting me out of context. But thanks for the clarification anyway. Sławomir Biały (talk) 01:45, 25 April 2017 (UTC)

You are such a sly, slippery fellow. Of course you did not question BrightR's competence, you only reminded him "please to assume good faith", as if he were the one that needed reminding. And you have slipped around the little fact that you introduced "competency" into this discussion (at 01:32, 23 Apr.). And I see that you, having failed to demonstrate in the previous discussion any actual misquotation, are now sliding into "quoting me out of context." Such adroitness feels (as BrightR said) ... disingenuous. ~ J. Johnson (JJ) (talk) 23:38, 25 April 2017 (UTC)

~~{{hab}}~~

For the second time Sławomir Biały has hatted my comment (with the reason "Offtopic personal attacks"), which I am unhatting. Sławomir, it is not for you to censor someone else's remarks when you don't like them. Your style of argument was criticized, to which your response was to impugn the good faith of another editor, just as you have done previously. We can hat this section when you acknowledge the criticism and abandon your incivility. ~ J. Johnson (JJ) (talk) 21:14, 26 April 2017 (UTC)

For the record, this not just whining of J.Jonson. I was a subject of several personal attacks of Sławomir in this page as well. This really poisons the discussion. Slawomir, please refresh your grasp of WP:NPA ASAP. Staszek Lem (talk) 17:10, 27 April 2017 (UTC)

Slawomir, IMHO the February change was not a policy change; it was aligning summation type wording with the actual policy. North8000 (talk) 20:59, 24 April 2017 (UTC)

Yes, aside from a tiny matter that there seems to be differing views on what the policy actually is. Or should be. ~ J. Johnson (JJ) (talk) 21:38, 24 April 2017 (UTC)

To return to the question of what next: the exact and entire question asked in this RfC (in regard of the contested change) is: "Was this a good idea? Please express a view below."

Nothing more is asked. While some participants here have an anticipation that disapproval should result in removal, there is no basis for that in the RfC. The arguments about a supposed lack of consensus for the change are outside the scope of this RfC.

What the views expressed have shown are some differences in interpretation and even fundamental assumptions (which certainly predate the change, and do not follow from it). As these issues evoke strong reactons, it would be a useful (if possible) next step to sort out and clarify them. Even more so, examples of the rather speculative concerns that have been raised should be collected and examined to see just what the problems are, and if their derivation or remediation depends on or requires any particular interpretation of WP:V.

What we do not need is another poorly formulated RfC. I think what we need most is no RfC for several months, lest it become just a continuation of the current debacle. (Serveral months might be needed just to prepare, as described above, while lack of preparation would make it senseless.)

For those who insist the current wording will have all kinds of dire consequences: if so, then six months should provide plenty of examples, which would greatly strengthen your argument. ~ J. Johnson (JJ) (talk) 21:44, 24 April 2017 (UTC)

What I see moving forward: two things are very clear:

One, there are existing Wikipedia policies that exclude some verifiable material—WP:NOT. Those who think all verifiable information belongs on Wikipedia are at odds with current policy.

Information should not be included in this encyclopedia solely because it is true or useful. A Wikipedia article should not be a complete exposition of all possible details, but a summary of accepted knowledge regarding its subject. Verifiable and sourced statements should be treated with appropriate weight.

Two, in case of an editing dispute, consensus is needed for the inclusion of the disputed material—the (not-currently-disputed) last sentence of WP:ONUS. Those who think all verifiable information should be included until there is consensus to remove it are at odds with current policy.

Consensus may determine that certain information does not improve an article, and that it should be omitted or presented instead in a different article. The onus to achieve consensus for inclusion is on those seeking to include disputed content.

These two ideas can be summarized as verifiability does not guarantee inclusion or verifiability is not a reason for inclusion or verifiability is necessary but not sufficient for inclusion. The last one, "necessary but not sufficient", has been suggested and supported by people on both sides of the RfC, so it seems like the best WP:CONSENSUS to use it. Bright☀ 20:31, 25 April 2017 (UTC)

I could go with "necessary, but not sufficient". Perhaps better if that could be offered as a clean RfC. ~ J. Johnson (JJ) (talk) 21:17, 26 April 2017 (UTC)

Alternative suggestion

IMO a better way is to combine the two versions (as explained in my 'nay'!vote above):

Title : Verifiability does not guarantee inclusion

Body: While content must be verifiable in order to be included in an article, verifiability alone is not a reason for inclusion, and does not guarantee that any content must be included in an article.

Staszek Lem (talk) 16:44, 10 April 2017 (UTC)

Support. Per my prior comments. Alanscottwalker (talk) 18:49, 10 April 2017 (UTC)
The current version is better but I'm worn out and in the mood to cave on a compromise if that would settle it. North8000 (talk) 19:08, 10 April 2017 (UTC)
Oppose. Partly because I'm getting cranky and disinclined to yield anything to a poorly-organized and ill-advised RfC. But largely because the "verifiability alone" phrasing suggests that (with other arguments) verifiability is a reason for inclusion. The point here is that verifiability should only open the possibility of inclusion. ~ J. Johnson (JJ) (talk) 20:18, 10 April 2017 (UTC)
You are precisely correct.North8000 (talk) 20:28, 10 April 2017 (UTC)
I could live with it... but would prefer it to say: "Verifiability is not a guarantee of retention", and "verifiability alone is not a reason to retain problematic material". That's really what we have been discussing. Blueboar (talk) 20:47, 10 April 2017 (UTC)
In my understanding of English "inclusion" covers "retention"; in any case both must be covered, so that people thingk twice before dumping trivia into articles. Staszek Lem (talk) 19:48, 11 April 2017 (UTC)

for those of us who are getting fatigued (and I include myself in that)... perhaps it is time to stop trying to convince each other, and to instead sit back and let others comment for a while.Blueboar (talk) 20:41, 10 April 2017 (UTC)
The discussion that led to the current language wasn't this fatiguing. And faulty rhetoric is pernicious, it should be resisted at every step. ~ J. Johnson (JJ) (talk) 19:38, 11 April 2017 (UTC)

re: verifiability should only open the possibility of inclusion - please explain the difference between "opening a possibility" and "being a reason", so that we can arrive at a better mutual understanding and hence a consensus. Also explain why you think this distinction is important for the policy (eg what bad consequences may come out of their confusion). Staszek Lem (talk) 19:42, 11 April 2017 (UTC)

P.S. It dawns upon me that this issue is a watershed between "inclusionists" and "deletionists": For an "inclusionist" , verifiability is a reason for inclusion, and the OP has to work hard to prove why something must be excluded. For a deletionist it is vice versa: you have to have really solid reasons to include something, so that when it is in, no policy can kick it out. Staszek Lem (talk)

I'm an "inclusionist" and I believe verifiability is not a sufficient reason for inclusion. Bright☀ 01:34, 12 April 2017 (UTC)

That could use some explaining. ~ J. Johnson (JJ) (talk) 01:54, 12 April 2017 (UTC)

I'm also an "inclusionist" and I believe verifiability is not a sufficient reason for inclusion. But that is not what this about. The current wording simply says that after wp:ver has been met, wp:ver "steps aside" on the include/exclude question. Sufficient reason for inclusion can be something as simple as "it's a plus for the article" or "we're building an article"

Not quite. Your characterization is spoiled by your caricaturization of the alternatives. On one side, it is the nature of a guaranteed inclusion that "no policy can kick it out." (Roughly speaking. "Policy" is not the right word here.) On the other side, no one is claiming that "you have to have really solid reasons to include something". But you miss the point: what we are saying is that all the argument about why something should be included (or excluded!), the reasons for doing so, and how "strong" those reasons have to be, should be entirely outside of mere verifiability.

I don't understand why you seem unable to grasp something so simple, but I will try one more time. First, if something is not verifiable, it does not belong in the encyclopedia at all, and lack of verification is therefore sufficient grounds for removal. Which is to say, verifiability is required for inclusion, and sufficient for exclusion. Okay? The issue is whether any material, having met this minimum requirement of verifiability (and not excluded by any other policies), is an automatic home run. I am saying that verifiability only puts you on first base, where you then have a possibility of scoring a home run. Or if you want a deeper analogy: you have to be on the eligibility list before you can swing at the ball. Is that clear enough? ~ J. Johnson (JJ) (talk) 02:02, 12 April 2017 (UTC)

Yes, I agree that "The issue is whether...". However you are not answering my question on the difference between "opening a possibility" and "being a reason" (notice the indefinite article; your "home run" analogy is synonymous with "the reason", not "a reason"). Staszek Lem (talk) 17:43, 12 April 2017 (UTC)
I think what you describe as "something so simple" is perfectly understood by all in this discussion, and is in fact accepted by us opposing the change. No one here is saying that a verifiable fact must be included for being verifiable.

What we oppose is the idea that, when a fact has been found verifiable, that has no bearing at all in our weighting of its merits for inclusion. It's all a matter of degree; you're saying here that verifiability should not put the dial for inclusion up to 100%, to which we all agree (in fact, the old wording "not a guarantee for inclusion" described this much better). But the change has instead moved the dial all the way down to 0%, which is the source of the disagreement; many of us think that finding a fact supported by a reliable source provides some value for inclusion in between 0 and 100%, with the amount being larger depending on the higher quality of the source.

That recent change does not reflect how many editors work, and thus is failing to describe the actual practices of the community - which is what the policy should be doing. It should be worded in a way that reflects this disagreement, or at least remaing ambiguous about it, not taking any position. Diego (talk) 06:25, 12 April 2017 (UTC)

No one is suggesting that WP:V creates an "automatic home run". But the current guideline does not say that verifiability does not create an "automatic home run". Instead it says that verifiability is not a reason for inclusion, which is to say that it is not a reason (among others) for including material, and instead can only be used to exclude it. There is a huge gap between "a reason" and "a home run". In court, lawyers on both sides of an argument will discuss reasons for things, but surely you're not suggesting that anything that has a reason behind it is always therefore a home run. This is, in fact, the same kind of fallacious black-and-white thinking of which Stanisław is guilty. Sławomir Biały (talk) 12:25, 12 April 2017 (UTC)

of which Stanisław is guilty I strongly protest against personal attacks and misrepresenting my position. Please strike it out. Here I am, starting a subsection in an attempt to find a middle ground, and suddenly I am an extremist. Staszek Lem (talk) 16:27, 12 April 2017 (UTC)

It may help to discuss not only what we think the policy should say... but why we are saying it. Looking back to when we first added the paragraph, the purpose was this:

WP:V is very often the first policy that new editors are pointed to (usually because they didn't realize that we want citations, and they added a fact that needed a citation). The new editor quickly discovers that if you don't provide a source, someone might come along and challenge your contribution. It might be removed. They assume (incorrectly) that if "Don't provide a source and your contribution might be removed" is true, that the opposite must also be true - "provide a source and your contribution won't be removed". They don't realize (at this point in their wiki-history) that there are other reasons, besides verifiability that might cause their contribution to be removed. And they often get upset when (having provided a source) they find that a contribution is still being challenged and removed.

So... we added the "not a guarantee" language ... simply to say - WARNING: Verifiability is just the beginning of the conversation when it comes to deciding whether to keep or remove a contribution... it isn't the end of the conversation. There are other policies and guidelines that might cause your contribution to be removed. Blueboar (talk) 12:46, 12 April 2017 (UTC)

So ... why don't we just say that? " Verifiability is just the beginning of the conversation when it comes to deciding whether to keep or remove a contribution... it isn't the end of the conversation. There are other policies and guidelines that might cause your contribution to be removed., followed by the consensus line. – Philosopher ^{Let us reason together.} 14:53, 12 April 2017 (UTC)

well... it's a bit clunky... and when we first added the "guarantee" wording we thought we had come up with a way to get the same warning across using phrasing that sounded a bit snappier. But if everyone prefers the clunkier (but perhaps more direct) language, you will get no objection from me. Blueboar (talk) 15:47, 12 April 2017 (UTC)

Maybe we have something here, but first we should clarify exactly what the idea is...what (if anything) would come out from where, and what would go in where.North8000 (talk) 16:10, 12 April 2017 (UTC)

Do we all agree about the inherent non-symmetry: non-verifiablity is a sufficient condition for exclusion, but verifiability is not a sufficient condition for inclusion? Staszek Lem (talk) 18:05, 12 April 2017 (UTC)

I don't think any compromise is possible. We're at RfC because some users want the policy to say that verifiability isn't a reason for inclusion and will not accept any change that removes that wording. In fact it's the sine qua non for inclusion. If there are any reasons to include things in Wikipedia at all, then verifiability is the very topmost item on the list. If there are no reasons to include things in Wikipedia then why are we here? We might as well just switch off the servers. "Verifiability isn't a reason for inclusion" is a novel interpretation that can't remain in this policy unless there's a community consensus that this really is what editors want.—S Marshall T/C 18:10, 12 April 2017 (UTC)

I think that compromise is possible once that you realize that you are working against your own goals. :-) I think that you and I are both inclusionists. The main reason for inclusion is simply to build good, informative articles, wp:ver is merely a constraint on inclusion. Isn't building good informative articles enough of a reason? Why would you need to say that compliance with one of many constraints is a reason for inclusion? You are supporting wording that (among st other things) implies than more than that is required to include something. Sincerely, North8000 (talk) 18:31, 12 April 2017 (UTC)

It doesn't say that, though. If it said any of that then my objections would be much reduced, although I remain firmly of the opinion that collecting verifiable information is what Wikipedia is for, and therefore, in case of dispute, verifiability is the single most important reason to include a disputed thing.—S Marshall T/C 19:10, 12 April 2017 (UTC)

Most of the policies that are mentioned when there is a disagreement about whether to include something, such as WP:V, WP:OR, WP:NOT, and WP:NOTE mostly describe constraints on what can be included. The policy that seems to say more about what should be included is WP:Editing policy, which says "Wikipedia is here to provide summaries of accepted knowledge to the public...." (Emphasis added). The word "summaries" tell us the majority of accepted knowledge should not be included, because if we try to include everything, it isn't a summary, and it isn't an encyclopedia. Jc3s5h (talk) 19:20, 12 April 2017 (UTC)
@S Marshall: I agree, but let's dissect that sentence. Even by your own wording in that sentence, "verifiable" (in its position as an adjective that narrows the term "information") is a restriction on inclusion, not a reason for inclusion. Implicitly (and I agree that it should be said explicitly in some prominent place) the reason for inclusion is to build good informative articles.North8000 (talk) 19:30, 12 April 2017 (UTC)
verifiability is the single most important reason to include a disputed thing At the first glance "collecting verifiable information is what Wikipedia is" is a reasonable stance. However please clarify you position to me: the fact "Lil Nigga Ho used the word 'f...' 11 times in his new album, including 'f... you' 5 times and 'f... him' 3 times", if true, belongs to wikipedia, yes or no? Staszek Lem (talk) 19:36, 12 April 2017 (UTC)

No, it doesn't belong to Wikipedia. It probably doesn't belong in Wikipedia either, which is why the policy wording I want to restore says verifiability is not a guarantee of inclusion. The place to talk about verifiable information that doesn't belong on Wikipedia, though, is in the relevant policy, which is WP:NOT.—S Marshall T/C 20:04, 12 April 2017 (UTC)

Well, IMHO the statements "verifiability is the single most important reason to include" and "verifiability is not a guarantee of inclusion" kinda contradict each other, unless one recognizes that the words "reason" and "guarantee" are not exactly synonymous (which does not always happen, and people keep miscommunicating due to speaking about different things). Therefore I suggest to change the wording to less ambiguous a bit below. Staszek Lem (talk) 20:16, 12 April 2017 (UTC)

Please notice that I attempted to slightly change the terminology, using a less ambiguous term "condition" instead of "reason". My, now bolded, statement above is logically equivalent to: (A) "verifiability" is a necessary condition for inclusion; (B) it is not an unconditionally sufficient condition for inclusion. Are the statements (A) and (B) correct? Staszek Lem (talk) 18:45, 12 April 2017 (UTC)

I would agree with both of those. DonIago (talk) 20:07, 12 April 2017 (UTC)

Not as good as the current version, but looks good to me if that would settle it. A mouthful but logically sound with respect to this discussion. The proposal should be made clear if it goes further. E.G what comes out from where, what goes in where. North8000 (talk) 20:40, 12 April 2017 (UTC)

I think there is general acceptance of A being correct, but not B. While B might be strictly accurate, the double-negative is confusing, and it allows the implication that verifiability can be sufficient, which is the core issue here. Also, I think it misses a key point: once A is met, WP:V should be no longer applicable. ~ J. Johnson (JJ) (talk) 20:53, 12 April 2017 (UTC)

P.S. I am more inclined to a more succinct variation of your previous bolded phrase: verifiability is required for inclusion, non-verifiability is sufficient for removal. ~ J. Johnson (JJ) (talk) 21:04, 12 April 2017 (UTC)

Actually, the statements "verifiability is required for inclusion" and "non-verifiability is sufficient for removal" are logically equivalent. Staszek Lem (talk) 21:21, 12 April 2017 (UTC)

They would be logically equivalent in a binary world where the only outcomes are true or false. But in Wikipedia, a passage may be necessary to satisfy policy (for example, inline citation associated with a direct quotation), policy may say a passage is required to be removed (controversial statement about a living person not supported by citation to a reliable source), or allowable but not required (most passages). To say "non-verifiability is sufficient for removal" means it would be unfair to apply disciplinary measures for removing the statement, but it does not say that all those editors who read it and left it there are negligent. Jc3s5h (talk) 21:31, 12 April 2017 (UTC)

Yes, I tried to formulate as "binary" as possible, because several people already noticed miscommunication issues due to different understanding of natural language, so essentially people do not hear each other. As for your argument, the policy is not intended to be a basis for reprimanding or otherwise slap-tagging wikipedians (comment on text, not on editor). We are volunteers and demands to be perfect are forbidden. Therefore your "unfair" argument is irrelevant. Staszek Lem (talk) 02:16, 13 April 2017 (UTC)

It depends on how one squints. "Inclusion" could refer to the act of inserting material (insertion), or it could refer to the state of retaining such material (retention). One could be permitted but not the other. There are additional considerations, such as process. E.g.: given some case where both insertion and retention are "wrong", and therefore exclusion (as both act and state) is proper, there may be issues in how that is done/achieved. Part of the problem we have here is that absence of a statement regarding the flip-side – even though it is tautologically equivalent and perfectly obvious – is taken by many editors as silence, and silence as permission. ~ J. Johnson (JJ) (talk) 22:29, 12 April 2017 (UTC)

No. We are here because one user (you), having missed the original discussion leading to the change and not liking the result, re-opened it. But not getting any satisfaction there (for various reasons, broadly attributable to your non-cooperativeness), you insisted on an RfC. Which, as agonizingly evident from all the directions this RfC keeps darting off into, is so inadequately formulated that we still don't have consensus of the underlying issues and terms. Undoubtedly you think the "yes or no" is simple enough, but that is entirely superficial. ~ J. Johnson (JJ) (talk) 20:57, 12 April 2017 (UTC)

Sure, the whole problem here is me and my uncooperative, muddleheaded behaviour. I'm a terrible, terrible person for upsetting you in this way. We do know that you think this. :)—S Marshall T/C 22:03, 12 April 2017 (UTC)

Histrionics aside, the problem is your opinion that collecting verifiable information is what Wikipedia is for. WP:NOT clearly states that this is not the case: "merely being true, or even verifiable, does not automatically make something suitable for inclusion in the encyclopedia." Even your preferred version of WP:V clearly states that this is not the case: "The onus to achieve consensus for inclusion is on those seeking to include disputed content." Both these statements very clearly show that while verifiability is required, it is not sufficient for inclusion. Bright☀ 16:50, 16 April 2017 (UTC)

And that's quite consistent with what I've said. "Almost all content that's verifiable to a genuinely reliable source belongs somewhere on the encyclopaedia" does allow for exceptions. Can we now please refocus on restoring "verifiability is not a guarantee of inclusion".—S Marshall T/C 10:26, 17 April 2017 (UTC)

A large number of editors expressed to you time and again why it's not consistent with what you said and why your wording is less desirable. So no, at the very least this RfC is going to close with "no consensus", mostly because it was never about working towards consensus, it was an all-or-nothing revert attempt. Bright☀ 15:01, 17 April 2017 (UTC)

Sorry, false. Please take a look at the headers of the subsections of this RFC. Staszek Lem (talk) 17:39, 17 April 2017 (UTC)

Well, let's set expectations correctly. The purpose of this RfC is to determine whether there is community consensus for the recent, drastic change to the longstanding stable version of the policy. That's a yes or no question, so it's a binary RfC with two possible outcomes. If there is consensus, then okay, that's silly but I will accept the result. If there isn't, then the recent change will be reverted, the longstanding stable policy wording will be restored, and then we can talk about whether there's a rewording that will gain community consensus. But let's let the RfC play out first.—S Marshall [f [User talk:S Marshall|T]]/C 22:02, 17 April 2017 (UTC)

Well, the wording of the RFC is just what it is, not something written down here. What you just wrote I would have objected to as malformed because it implies that a consensus is needed to retain the current version. Sincerely, North8000 (talk) 01:30, 18 April 2017 (UTC)

To be fair, a consensus *is* needed to keep and apply the currently disputed version. The text won't change until we reach a new consensus, but it will keep having the "this part of the policy is being discussed for lack of agreement on its wording" tag unless we can agree on a better wording. In addition, there are several editors now around here that don't like the wording achieved in the previous discussion, not just S Marshall. The above comments centered in blaming Marshall for the existence of the current discussion do not lead to solve the disagreement. Diego (talk) 12:48, 18 April 2017 (UTC)

To be fair, I'm certainly to blame for all this discussion. ;-) From these editors' point of view, I've forced an RfC about something they would much have preferred to slip quietly through based on a talk page discussion. My position is very clearly and unambiguously that this novel wording requires community consensus and in the absence of consensus, should be reverted to the stable version. I certainly don't expect these editors to agree with my position on this! The point is for the closer to decide and, based on the course of the RfC so far, it's likely to be the point that determines the final outcome. In order to ensure the point is fairly and correctly decided, I would suggest that when the RfC is nearing its end, we post on WP:ANRFC asking for a triumvirate of experienced admins to perform the close.—S Marshall T/C 20:34, 18 April 2017 (UTC)

You misstate the matter, and insinuate lack of good-faith. The strong suggestion that you should not immediately start an RfC was because the rest of us wanted to take a break and come back fresh, during which break someone might work with you to clarify your concerns and better formulate an RfC. You were not making any progress in the preceding discussion (in part because you refused to clarify your ambiguous statements), and proceeding straight to this RfC without resolving any of the underlying issues has only made for a bigger muddle. ~ J. Johnson (JJ) (talk) 01:15, 19 April 2017 (UTC)

The whole point of RfCs is to gather opinions of people previously not involved... Telling someone not to start a RfC because one will not be able to participate is opposite to the spirit of WP:Dispute resolution. Diego (talk) 09:25, 20 April 2017 (UTC)

You misunderstand the situation and intent. Marshall has shown a certain ineptitude in even explaining what his concerns are, let alone advancing the measures he thinks necessary. It was entirely in the spirit of WP:Dispute resolution that he was urged to not proceed on his own, but to seek the assistance of another editor. Not to censor him, but to keep him from mucking up his own best interests. ~ J. Johnson (JJ) (talk) 22:20, 21 April 2017 (UTC)

Bright* is exactly right. In the preceding discussion Marshall's statements did not ~~could not~~ unambiguously state his objections, and he refused to clarify his statements when requested (while demanding a universal disproof from the other editors). Quite unsurprisingly he ~~could~~ did not obtain any concurrence. His response was simply to bring in more editors via an RfC, without remedying any of the defects that have plagued this discussion since he joined it. Others have tried to help him, but he seems adamant to any suggestions. And ~~it seems to me he may have a memory problem, as~~ for reasons unclear to me we keep having to go over the same points with him. ~ J. Johnson (JJ) (talk) 22:26, 17 April 2017 (UTC)

[Revised language that might be taken as touching on another editor's abilities. ~ J. Johnson (JJ) (talk) 01:32, 19 April 2017 (UTC)]

STOP - comment on the proposals, not the other editors. Blueboar (talk) 20:41, 18 April 2017 (UTC)

I think several of us (I include myself) have reached the point where we are suffering from "last word disease"... where we are so wrapped up in the debate that we simply can not let the "other guy" get in the last word. We end up either repeating what we have already said (multiple times) or resort to personal attacks. May I suggest that those of us who have been the most outspoken in this dispute sit back and let others comment for a while. Blueboar (talk) 02:24, 19 April 2017 (UTC)

Sorting out a few fundamantals

Regarding a possible compromise, there are a few areas which might be helpful to sort out. I would like to ask this question of the folks that don't like the current wording. Let's talk about an item for potential inclusion. Let's say that it met all "9" of the clear cut policy constraints (requirements), including the "biggest" one,, #9, which is wp:verifiability. Now comes step #10, the question comes whether or not to include it in a particular article. And, practically speaking, this means that there is some sort of debate or dispute about its inclusion. You might imagine a "scale" during step #10 where the reasons for and against inclusion are piled on the two sides of the scale to see which way it tips.

One school of thought is that it had to meet the "9" requirements to even be considered for inclusion, and so all of those requirements (such as wp:ver) already had their effect, and that having met any of those requirements (such as wp:ver) should not be usable a second time, not be pile-able on the "scale" of arguments during step #10.
Another school of thought is that meeting other requirement should be so usable, usable a second time and be pile-able on the scale during step #10.

Number 2 could be a reason why you oppose the current wording, but it is not the only possible reason for doing so. Would you say that #1 or #2 matches your thoughts? (PS: I'll be gone for 3 1/2 day days) Sincerely, North8000 (talk) 14:49, 20 April 2017 (UTC)

My position is that verifiability is both a prerequisite for inclusion, and also a factor in the "scale" in the balancing exercise you envisage. I think after reading all this verbiage we might be looking for wording that says: (1) that only verifiable information can be included on Wikipedia; (2) verifiability is one of several considerations that affects the decision on whether to include information; (3) information that's backed up by an inline citation to a reliable source is more includable than information that isn't (or in other words, information that's actually verified right now is more includable than information that editors think is verifiable from somewhere); (4) verifiability doesn't trump WP:NOR or WP:NOT, so information that fails either of those policies should be removed even if it is verifiable; and (5) Flyer22's law: removing verifiable information from established articles requires care and thought, and should not be done recklessly or against consensus.—S Marshall T/C 23:28, 24 April 2017 (UTC)

OK thanks. That does mean that we have a fundamental difference and not just a matter of differing impressions from reading it. :-( It would also be useful to hear the answer from other folks who do not like the current wording. Sincerely, North8000 (talk) 02:29, 25 April 2017 (UTC)

What if we remove the word "inclusion?

I am getting the feeling that many of us are getting hung up on the word "inclusion"... mentally investing that word with all sorts of different (nuanced) significances and meanings. Hell, I'm not sure if we even agree on why we are using the word "inclusion" in the section in the first place. If I am right, then the only way forward may be to rephrase in a way that omits that word. So, I am curious to see if we can do so. let's try to come up with a few formulations that do not use the word "Inclusion" at all. Blueboar (talk) 22:30, 12 April 2017 (UTC)

Rather than removing, as I understand, we must try to split it into several meanings encountered here. So far I see the 'trinity' of distinct actions: addition, retention and removal (and a nit-picking symmetrization with a 'non-action' of "non-addition"). Now, the question is do we need separate rules/flavors to govern these? Staszek Lem (talk) 02:21, 13 April 2017 (UTC)

I think you are making this more complex than it needs to be... rephrasing without the word may be easier. Blueboar (talk) 11:01, 13 April 2017 (UTC)

I am just diagnosing a cause of disagreement. This even may help removing the offending word. Otherwise there we may try to insert even more "offending" one. Staszek Lem (talk) 16:01, 13 April 2017 (UTC)

I tried analyzing it and IMHO at the root root of it there is no actual underlying conflict. The problem happens at the next level up which is wording. I may have this all wrong but here is how I dissect the differences:

First there exists what I've been calling an urban legend, but basically a (lets call it) "floating concept" that, when there is a dispute over inclusion / exclusion of material, that meeting wp:ver is a force for inclusion, or an argument for inclusion. This concept is not specified by any policy, but there are words here and there which implicitly might have supported that concept a bit. This includes the wording that we got rid of a few months ago.
We tend to go by our experiences. I think @S Marshall: is mostly concerned about / focused on people trying to keep out good material. I think that many of the rest of us are also concerned about wiki-lawyering to force in bad material.
I don't know whether or not S Marshall is making the logical distinction (regarding meeting wp:ver) between such fulfilling the main requirement for inclusion and such being a force or reason for inclusion (e.g. during disputes). If they are not, IMHO that is a logical error. Even if he is in favor of both of them, he should acknowledge that they are two different things. And, respectfully, that logical error would conflict with those of us who are logically correct on this point. :-) :-) But if even they did acknowledge this distinction, that would only remove 1/2 of their motivation. While he would then have to admit that the "floating concept" has no solid basis in wp:ver, he still might prefer supporting that "floating concept".
In the change made (after a long discussion) a few months ago, we did two things regarding this "floating concept" We took out the wording that provided a tiny amount of implicit support for the "floating concept"; we also put in wording which clearly says that the "floating concept" does not exist. That is probably the second area of "conflict".
This might relate to the broader feeling that we're missing the prominent policy which says "our mission is to build good articles with good material, and that's a plenty good reason to put a piece of good policy-compliant material in". (as a sidebar, so no "floating concept" is needed) S Marshall would probably like wp:ver to do that a little. Most us feel that wp:ver and the wording under discussion is not the place to do that.

Respectfully, North8000 (talk) 20:31, 13 April 2017 (UTC)

I'm looking at real world disputes about what to include in our encyclopaedia, and the first thing I note is that they're usually, at heart, disputes about what to cut out of our encyclopaedia. I feel that when we're deciding whether to cut something out or not, the very first question is, always, what do the sources say. If the disputed information doesn't appear in sources that are genuinely reliable, then I think that's an extremely strong argument for cutting it. Conversely, if the disputed information does appear in genuinely reliable sources, then yes, that's a factor supporting the case for not cutting it. I think that North's "floating concept" has always been implicit in WP:V. If the information is genuinely based on truly reliable sources, then it should normally be kept, although it might be moved. There's a presumption that verifiable, reliably-sourced information belongs on Wikipedia. A presumption is not a guarantee, and there are still situations in which verifiable information should be cut or moved. All of these situations are fully covered in WP:NOT or WP:NPOV. Those are the appropriate policies to discuss that.—S Marshall T/C 21:49, 13 April 2017 (UTC)

I agree strongly with the above statement. If content is verifiable, and supported by high quality reliable sources, then that by itself is indeed a very clear indicator that it belongs somewhere in an encyclopedia. The examples I gave above, of standard facts from textbooks, do not normally require any WP:PAG gymnastics to justify their inclusion: in practice, verifiability is sufficient. Sławomir Biały (talk) 22:05, 13 April 2017 (UTC)

That any and all verifiable content "belongs somewhere in an encyclopedia" is one of the fundamental opinions here, and really deserves its own discussion. As it is, we have covered this, but it has been so disorganized and scattered about that nothing is retained, so it seems we must cycle through it again. ~ J. Johnson (JJ) (talk) 22:18, 13 April 2017 (UTC)

I will point out one last time, that no one anywhere in this discussion has said that "any and all verifiable content 'belongs somewhere in an encyclopedia'." This is either a WP:CIR failure on your part, or is just bad faith trolling. But it stops now, since we are now absolutely clear that no one in this discussion is saying that. Not me, not S Marshall, not anyone. You can continue to believe that, but please keep these beliefs to yourself in the future. And please don't ever misquote me again. Sławomir Biały (talk) 00:04, 14 April 2017 (UTC)

It is one of the fundamental opinions which was flatly rejected since early days of wikipedia, as covered in multiple WP:NOT bullets. Staszek Lem (talk) 22:53, 13 April 2017 (UTC)

I agree that it is not the case that "Any and all verifiable content 'belongs somewhere in an encyclopedia'." In fact, there already is a Wikipedia policy about that very question. So, we are apparently all violently in agreement about this point. Perhaps it is time to stop attacking strawmen that have nothing to do with the discussion? Sławomir Biały (talk) 00:10, 14 April 2017 (UTC)

Bławomir Biały: Cool your jets; you are getting unnecessarily intemperate. Your insinuation that I lack either competence or good-faith is personal attack, and an apology is in order.

Nor have I misquoted you. That is a false charge, and even a misrepresentaton of what I have said. As to the matter at hand, I quote exactly what you said just above (copied from the diff): "If content is verifiable, and supported by high quality reliable sources, then that by itself is indeed a very clear indicator that it belongs somewhere in an encyclopedia." (22:05, 13 Apr). Your objection seems to be to my qualifying this "verifiable content" term we are all waving at with "any and all". In fact, no one has qualified that term beyond "reliable sourced and not excluded by other considerations". If you really meant something more specific you should have said so. But you didn't, so it is a fair inference that whatever you meant does not exclude anything less than any and all.

What IS being said by some one – notably the chief architect of this RfC, S Marshall – is that "verifiable content", without further qualification other than "probably" and "almost always", "belongs somewhere" in the encyclopedia. E.g.:

01:47, 23 Mar: "it probably belongs somewhere in the encyclopaedia". And: "Everything verifiable should be somewhere in the encyclopaedia".

00:49, 25 Mar: " I did say that verifiability should be a virtual guarantee of inclusion somewhere in the encyclopaedia ...."

12:11, 27 Mar: "almost always belongs somewhere".
15:35, 28 Mar: "almost always belongs somewhere".

With a strong inference that "all" is the qualifier meant. Now you might object that these comments occurred not in this discussion (meaning the RfC, opened 10 Apr.), but in a prior discussion. But this discussion is a straight-forward continuation of the prior discussion, without any break, and this discussion is informed by the past discussions.

So as I said before: whether "verifiable content" (however qualified) "belongs somewhere in the encyclopedia" is one of the fundamental opinions here, and really deserves its own discussion. ~ J. Johnson (JJ) (talk) 23:08, 15 April 2017 (UTC)

I find it slightly creepy that you've collected the diffs so carefully. I'm taken aback. I realise that my views on this are heretical, blasphemous and beyond the pale, so any objection I might have to J Johnson's damning condemnation is mere quibbling to be dismissed out of hand; but if I can interrupt the thundering from the pulpit just for a moment, I would like to point out that all this is in fact a distortion of my position. I've consistently said "almost all content that's verifiable to a genuinely reliable source belongs somewhere in the encyclopaedia", which means that:- (1) In logic, "All" cannot possibly be the qualifier meant; and (2) One or two hypothetical counterexamples do not make me wrong.—S Marshall T/C 10:21, 17 April 2017 (UTC)

Your repeated histrionics aren't helpful toward building consensus. Additionally, there's not "one or two hypothetical counterexamples", there's an entire policy called WP:NOT, and a process called WP:CONSENSUS, for when verifiable content should not be included. The wording that you support gives this caveat ("Consensus may determine [...] include disputed content") so really if you want to know what "makes [you] wrong" it's those policies. Bright☀ 15:14, 17 April 2017 (UTC)

@S Marshall:, I don't think that your views are any of those nasty things, and I have a lot of respect for you as someone who simply wants Wikipedia to be the best that it can be. And, at the root of it, I thin that we're both inclusionists for good material. But I also think that you are mistaken in this case, meaning that what you argue for would be a net minus to Wikipedia. Respectfully, I think that it comes from one perspective error (you are overly thinking about some rare relevant cases, and not about the common ones) and a couple of logical errors. Respectfully, North8000 (talk) 16:34, 17 April 2017 (UTC)

Oh, were those statements "taken back"? Marshall, is it really necessary to have to explain to an experienced editor the use of strike-out text to indicate retracted statements?

But it is not at all evident that you have actually "taken back" that sentiment when just this week you stated (21:49, 13 Apr):

if the disputed information does appear in genuinely reliable sources, then yes, that's a factor supporting the case for not cutting it. ... If the information is genuinely based on truly reliable sources, then it should normally be kept ....There's a presumption that verifiable, reliably-sourced information belongs on Wikipedia.

And then just a day later (just below here, at 14:19, 14 Apr), you said:

Yes. "If it's verifiable to a genuinely reliable source that meets WP:RS then it probably belongs on Wikipedia somewhere" happens to be my opinion ....

Which you top off with your statement above: "almost[emphasis added]all content that's verifiable to a genuinely reliable source belongs somewhere in the encyclopaedia".

These are your own, unretracted, words, and if they be distortions of your views the fault is your own. ~ J. Johnson (JJ) (talk) 23:20, 17 April 2017 (UTC)

And I don't retract them. Our sister project Wiktionary has a definition of "taken aback" which, apparently, it would help you to read. It's not my fault you don't understand my position; you've made no effort to do so at all, and as you've just very neatly demonstrated, you skim what I've said at very high speed and take wild guesses at its meaning. I can tell you what I think but I can't force you to read it attentively or with your brain engaged. When your only reaction to everything I've said is to scour it for things you can attack or take umbrage at, it's hardly surprising that your understanding of it is so limited.—S Marshall T/C 19:42, 18 April 2017 (UTC)

I have made great effort to understand your position (but how would you know?), and I find the greatest difficulty to be: your own statements of your position and concerns. Impugning my competence does not explain your position, nor address your concerns. ~ J. Johnson (JJ) (talk) 01:54, 19 April 2017 (UTC)

Regarding J. Johnson's quoting out of context, I suppose it is possible that it was a honest mistake. However, I will not apologize for pointing out that he is misrepresenting things, when he continues to do so with this completely false statement: In fact, no one has qualified that term beyond "reliable sourced and not excluded by other considerations". If you really meant something more specific you should have said so. But you didn't, so it is a fair inference that whatever you meant does not exclude anything less than "any and all". I have made it very clear from the beginning of the discussion that I am talking about standard, routine facts, that are sourced to standard scholarly literature ([4], [5], [6], [7], [8], [9]. I gave as examples the vapour point of mercury, and other kinds of routine facts sourced to standard physics textbooks. My opinion is that for routine facts of this kind, with high quality scholarly sources, verifiability alone does create a reason for inclusion somewhere in the encyclopedia. Since I have been very clear about this throughout the discussion, I can only conclude that User:J. Johnson does not believe that standard facts sourced to reliable scholarly literature ipso facto belong somewhere in an encyclopedia. Is that an accurate summary of J. Johnson's position? Sławomir Biały (talk) 13:01, 19 April 2017 (UTC)

You misinterpreted what I said, proceeded to misrepresent it as a misquotation, and on that basis then questioned here my competence and good-faith. If that does not warrant an apology then you have abandoned the "civil" usually required of "discourse". (Incidentally, you have not shown that I misquoted anything – if I have, just line up the original with what I quoted and show us the difference – nor quoted out of context, but as this may be beyond your competence I am disinclined to spend any more time on you.) ~ J. Johnson (JJ) (talk) 04:02, 20 April 2017 (UTC)

This now seems offtopic. Your quotation was obviously out of context, since you managed to completely mischaracterize my position as that of 'any and all verifiable content "belongs somewhere in an encyclopedia"'. This was, I should add, after I explained to you the difference "a reason" for something and "an automatic home run", which you still don't seem to grok. Obviously, if you continue to raise the same points that have already been demolished, then it will lead to questions of competency or WP:IDHT. Still, we are now very clear that no one in this discussion is saying that. You can choose to disengage with me, but if you continue to attack strawmen, I will continue to point out your errors. Sławomir Biały (talk) 11:37, 20 April 2017 (UTC)

I agree, there are certain facts or information bits, that are "obviously" relevant and hence verifiability alone would justify their inclusion. "Obviously" means, that it is fair to assume the community considers them as worth of inclusion and/or so did traditionally other encyclopedias.--Kmhkmh (talk) 13:30, 19 April 2017 (UTC)

That the community would consider them "obviously" relevant and be able to point to other encyclopedias that include these bits makes them irrelevant to this discussion, as they're not being added solely because they're verifiable. DonIago (talk) 13:43, 19 April 2017 (UTC)

This debate is not about facts being "added solely because they're verifiable", but about "verifiability not being a reason for inclusion" which is the wording of the policy. Being verifiable is a factor in the community considering the facts relevant and worthy of coverage, and thus a reason for inclusion, despite what WP:V currently says. Besides, you can't cut out WP:V from the debate of content inclusion: content policies themselves say that you need to take all of them into account, not in insolation; that "added solely because WP:V" is a strawman. Diego (talk) 14:48, 19 April 2017 (UTC)

And my feeling is that verifiability is a requirement for inclusion, not a reason for inclusion. "This is relevant information" is a reason; "this is verifiable information" is not, and editors should never be defending additions to articles with, "it's verifiable so it belongs here". Whether a fact is verifiable should be immaterial to its inclusion beyond meeting the minimum threshold of "we can verify it...but is it relevant, significant, etc...". "Added solely because WP:V" may or may not be a strawman, but it's a frequent defense for retaining information in articles. It's that kind of reasoning that led to, for instance, WP:IPCV, which raised the bar on "In popular culture" material to specify that sources must demonstrate not just the existence of a pop culture reference but also it's significance, precisely because people would defend all kinds of trivial inclusions on the grounds that they could be verified. DonIago (talk) 17:11, 19 April 2017 (UTC)

Being verifiable in a reliable scholarly source implies significance, contrary to the current wording of the policy. Sławomir Biały (talk) 17:30, 19 April 2017 (UTC)

[Added] "because WP:V ... is a frequent defense for retaining information in articles". Because it is a good defense. Appearing in a reliable source discussing the article topic makes the fact relevant, so as long as no policy opposes it (such as WP:NOT or WP:UNDUE or WP:SIZE or WP:SYNTH), there is no reason *not* to include it. WP:IPCV is an example of something that the community decided not to do; but lacking such prior consensus, why impose arbitrary restrictions to inclusion on facts that reliable sources have decided are relevant? Wikipedia is WP:NOTPAPER.

Moreover, "verifiability is no guarantee for inclusion" or "not a reason for inclusion" is one of the most used arguments used to try to WP:CENSOR verifiable information that someone just doesn't like, and the new wording is giving fodder to people trying to game the system this way.

my feeling is that verifiability is a requirement for inclusion, not a reason for inclusion. And given that several of us feel that verifiability is a reason for inclusion (maybe not a definitive reason, but a reason nonetheless), the policy should take a neutral wording on the matter, rather than the current taking a stance. Diego (talk) 18:05, 19 April 2017 (UTC)

And here we have the crux of at least two points of contention between the multiple parties who've contributed to this discussion (smile). DonIago (talk) 18:29, 19 April 2017 (UTC)

At least from my perspective, I am beginning to see part of the divide. Smarshall has described a "context" issue, imo, but verfiabiliy is not about context, it is about "proving-up" a discrete piece of information. Forgive me for stressing again, V's minimalist requirement is 'a single RS', not multiple sources, as SMarshall's comment discusses. Our policies on the various issues concerning context are WP:NPOV, WP:OR, WP:PAGEDECIDE. I am looking at this from another end then SMarshall, not his macro view, but as I see it, the real, in the act of editing a particular article, where the practical inclusion question, "does this belong, here, in this context, now?" is answered. (For those who love imperfect analogies: we are building a city of various brick-structures, and all verifiability is, is how to make good bricks, not about how to build the many different structures. And as every builder knows, some of the bricks will never be used.) I think, it's best to keep the discrete and practical focus for this policy - the bricks. (To extend my analogy, NOT is all that brush we have to clear away - someone else can use those sticks to build their city. :)) -- Alanscottwalker (talk) 00:18, 14 April 2017 (UTC)

Whereas I've always thought that verifiability is a matter of principle. For many years this policy said that verifiability rather than truth was the threshold for inclusion on Wikipedia --- which language I, and North8000, both went to a great deal of trouble to change; the two of us are in fact longstanding allies and partners on this point. It's novel for me to be on a different side from North8000. The case we both made in 2010-2011, and which ultimately prevailed at RfC in 2012, is that verifiability is in fact one of several requirements for inclusion. Blueboar enshrined this in the stable wording, "Verifiability is not a guarantee of inclusion", which I endorse.

But if you would prefer to focus on how this new wording will work at the "brick" level then I'm eager to play. You've inserted wording into a policy that greenlights removing verifiable information from the encyclopaedia. Can you think of any verifiable information that, for example, climate change deniers, intelligent design advocates, scientologists, moon landing conspiracy theorists, or pro-Palestinians might want to remove?—S Marshall T/C 01:11, 14 April 2017 (UTC)

I can think of multiple NPOV and OR reasons why they just can't remove whatever they want. But, it is precisely those types you describe who will say 'We gotta RS. It's V', so it has to be here. --Alanscottwalker (talk) 01:19, 14 April 2017 (UTC)

Sure, Alan, and if the only situation you envisage is when the white hat good guy is trying to stop the black hat from adding wrongbad information to the article, then the new wording is obviously ideal. I do get that. But can we please look at some of the other situations that this change affects?—S Marshall T/C 14:19, 14 April 2017 (UTC)

@Alanscottwalker:, the only situations I can think of where the OR policy would prevent the removal of information would be the removal of citations, while leaving the claim in the article, or removing part of the claim in the article so it no longer agrees with the supporting citations. Except for this, the OR policy is about excluding and removing information, not adding or retaining it. Jc3s5h (talk) 14:33, 14 April 2017 (UTC)

OR is also about the correct juxtaposition of facts, that creates a non-original article presentation of the subject - if you remove a critical fact, leaving an original impression or original implication regarding the subject, you violate OR. Alanscottwalker (talk) 14:45, 14 April 2017 (UTC)

Getting back on-topic: I think "inclusion" is the best, most appropriate word, and should be retained. I agree with Staszek overall (sorry Blueboar: it is complex) but quibble a bit with his wording that "we must try to split it into several meanings" – I think the usage is already split, and we need to clarify those usages. I think WP:V is (and should be) essentially one rule, but may need explicit explanation in how it applies in these different cases. ~ J. Johnson (JJ) (talk) 22:25, 13 April 2017 (UTC)
- Yes, that's what I meant under "split": explicitly cover use cases which people may distinguish/confuse. Staszek Lem (talk) 22:53, 13 April 2017 (UTC)

S Marshall's examples "Can you think of any verifiable information that, for example, climate change deniers, intelligent design advocates, scientologists, moon landing conspiracy theorists, or pro-Palestinians might want to remove?" are about accurate, sourced material that should be in the article, but which POV warriors will want to remove. Just brainstorming, but @S Marshall: if we could come up with something that helped in that area but didn't address your "if it's verifiable it belongs in Wikipedia somewhere" assertion, might that be a suitable compromise? North8000 (talk) 13:05, 14 April 2017 (UTC)

Yes. "If it's verifiable to a genuinely reliable source that meets WP:RS then it probably belongs on Wikipedia somewhere" happens to be my opinion, but it was never in the policy in the first place. From my point of view what's important is to restore "verifiability is not a guarantee of inclusion", which was removed in a drastic change to the policy that would require community consensus.—S Marshall T/C 14:19, 14 April 2017 (UTC)

I was thinking of something that is more than you asked for in one area and less than you asked for in another. My first lame attempt: Current wording plus: "Editorial practices and the objective of building good articles also affect inclusion or exclusion of material, but compliance with wp:verifiability and other core policies and guidelines means that the material has completed important requirements for inclusion." Just brainstorming. I realize that others might have objections to this, but right now I'm just starting with the toughest guy in the room. :-) :-) North8000 (talk) 15:51, 14 April 2017 (UTC)

Unfortunately, that leaves "verifiability is not a reason for inclusion" in the policy.—S Marshall T/C 17:45, 14 April 2017 (UTC)

No, that is (in some senses) the whole point of the exercise. ~ J. Johnson (JJ) (talk) 23:10, 15 April 2017 (UTC)

I concur with Diego's observations, passim. I do not concur that the change was a good idea and agree that it is for the reasons S Marshall and J. Johnson (JJ) concur (amazingly) upon: failure to resolve whether "verifiable info" is, ipso facto, a reason for inclusion. Regretting that I did not participate in the earlier, apparently exhaustive debate, I come down in favor of the argument that verifiability is a reason for inclusion. Pragmatically, that should mean that if an insertion is challenged and restored, a new consensus is needed to exclude it. My reasoning is that the fact that the datum was included in a publication meeting our standard for RS means that it has already been deemed worthy of being publicly shared by a respectable source, so that is an implicit "!vote" in support of retention, that can be overcome by at least two objections -- but not by one. FactStraight (talk) 04:19, 25 April 2017 (UTC)

But that is the requirement for everything to be included in any article. IMHO why should simply meeting the universal requirement be an argument (e.g. in a dispute) for inclusion in a particular article? Sincerely North8000 (talk) 12:38, 25 April 2017 (UTC)

With the current wording, Verifiability wouldn't be an argument for inclusion in any article.

Meeting the universal requirement of being verifiable is a good reason for inclusion in articles where the information is relevant, provided it isn't undue weight, anything in WP:NOT, or other reasons for excluding it. Diego (talk) 13:25, 25 April 2017 (UTC)

I think that my argument in that area goes like this: A rule is "Only human beings are allowed to win a Nobel prize." When they meet to decide on whether or not a person gets the prize, should "he's a human being" be usable as an argument for giving them the prize? North8000 (talk) 15:04, 25 April 2017 (UTC)

Following along with the metaphor: if the Intergalactic Council is making the choice among a pool of humans and aliens being taken into consideration, and it isn't a formal requirement that winners show proof of their humanity, and there are hundreds of thousands of categories of Nobel prizes, then "he's got an official certificate showing he's a human" would indeed be a good reason to award the prize for a category where that human shows some particular talent and no other human is making a competing claim. Diego (talk) 15:37, 25 April 2017 (UTC)

@Diego Moya:Besides being a pleasure to work with everyone here (because everyone here is working for the good of Wikipedia), it's a particular pleasure to have this exchange with you because you have been elucidating precise thoughtful logical arguments in the area of dispute. I think I see your point. One is that "thousands of" is a better analogy to Wikipedia than the "few Nobel prizes".....agreed. However, I think that I can note some important relevant flaws in your analogy. This is not to pick apart your good analogy, it is because I think it deals with core issues. First, your analogy is to sourcED material in Wikipedia. This whole debate is about material which complies with wp:VER which merely requires sourcABLE material, not sourcED material. So the material under discussion does not (necessarily) have that extra imprimatur (certificate). Second in in your inclusion discussion, you "needed" to give a (another) good reason for inclusion ("shows some particular talent") which somewhat indicates that an actual :-) reason for inclusion is needed. :-). But most importantly, the situation that you describe (in essence undisputed editing) is the situation where this whole concept (old or new wording) never comes into play. Overly-briefly, the situations where it comes into play are all "disputes"; S Marshall's are where people are trying to keep out good material, and "ours" are where meeting wp:ver is used as a way to force in bad material. "Our" side feels that the latter is a much more common problem, and also that the wording is more directly relevant to and used in the latter. Respectfully North8000 (talk) 16:39, 25 April 2017 (UTC)

Well, my analogy wasn't really meant as a completely accurate description of the situation, with all the hyperbole over aliens and galactic empires :-) It was more an emotional appeal as a reply to your vivid "Nobel prizes are for humans" image.

I admit that I hadn't though about the difference between sourceable and sourced content; but now that you point it out, I still don't think it's really relevant. Verifiable content is one for which is known that reliable sources exist, even if they are not included in the article right now. I agree that content with a "presumption of verifiability", which is the standard for WP:Notability, wouldn't be enough; but when WP:ONUS is invoked as a reason for inclusion by someone arguing "the content is verifiable (and I don't see reasons for exclusion)", I assume we are the point at which at least one RS is known. At this point, the "talent" of being noticed by a valid reference has been met.

WP:ONUS already defaults to excluding content when there is no consensus, so "your side" is at an advantage; recognizing (or at least not rejecting) verifiability as a (possible) reason towards inclusion (but not a guarantee) should entice the editors wanting to exclude the content to provide themselves a good reason why such content, reported by a reliable source, is nevertheless not a good fit for the article. Right now, they could merely reject it with a poor argument like "verifiability is not a reason for inclusion, and you have not achieved consensus for it" (i.e. the current wording of WP:ONUS). Diego (talk) 16:15, 26 April 2017 (UTC)

I don't have much of a direct response except to point out that the crux of the argument in your last sentence is wp:onus, the "not a reason" part is just a sidebar. FYI probably the most common of the types of situations that I'm thinking about are when there is fast "hit and run" type editing in order to POV an article. And there an edit summary of "undo removal of sourced material" is considered to be a strong argument to force inclusion and sufficient to undo a removal. The others are where a single source was used on a statement (in the voice of Wikipedia) that was either highly biased or clearly factually in error, where that particular statement tilted the article in an area that was under dispute. And in those cases meeting wp:ver is given as a force for inclusion. North8000 (talk) 17:57, 26 April 2017 (UTC)

Diego: Your "WP:ONUS already defaults to excluding content when there is no consensus" is ambiguous in not recognizing an important caveat. Note that consensus might be consdered to have three states: consensus FOR inclusion, consensus AGAINST, and the indeterminate state of NO consensus. WP:ONUS is really about the last state. But the implication of such a limited view is that no material can included without an explicit consensus for inclusion. This would be unworkable. (And possibly this is the essence of your objection?) Resolution of this can be had by recognizing that WP:ONUS applies where inclusion/exclusion has been contested.

Content is typically added (or removed) Boldly, without discussion (or even comments) on specific items, without any reason given. (Though "editorial discretion" is implied.) And sometimes such content is equally boldly Reverted, per WP:BRD. Up to that point (and subject to the recognized restrictions) I say that editors are pretty much free to add or delete as they consider fit, and no reasons need be given. It is only when an edit is questioned that we get into discussion, argument, and consensus, with WP:ONUS favoring exclusion where consensus is not reached.

I suspect that where "verifiability is not a reason for inclusion" (alone) is given is a reason for exclusion it is likely because someone argued that "verifiability IS a reason for inclusion". The intent of the current language is to clarify that verifiability is not "a reason", whether for inclusion or exclusion, but only a minimal requirement. Any argument that cannot come up with better reasons than "verifiability" is a poor argument. ~ J. Johnson (JJ) (talk) 23:21, 26 April 2017 (UTC)

I understand the intent of the current language ("verifiability is merely a minimum requirement"), and I would agree with it if that was all it did. The problem with the new wording is that it also interacts with WP:DUE WEIGHT, by undermining the following argument (for which "verifiability is a reason for inclusion" is merely a shorthand):

"This fact has been noted by a reliable source, connected to the topic of the article."
"When a reliable source notices some fact, it provides some support for it having due weight."
"Facts regarding a topic should be covered in an article with prominence relative to their due weight."
"Therefore, that this fact was noted by a reliable source provides some support for covering it in the article."

You are right that this is an argument from WEIGHT more than VERIFIABILITY; however, it can be legitimately made every time some fact is verifiable by a reliable source. Thus, the new wording is having effects on a different policy, which I think we all agreed is not a good idea.

If we could find a wording without this side effect, I wouldn't oppose this position of defining verifiability in this page merely as a binary requirement; as long as it didn't interfere with the idea that "verifiability on its own isn't a reason for inclusion, but it provides some due weight, which might be a reason itself". Diego (talk) 09:38, 27 April 2017 (UTC)

Possibly the answer lies in what you just wrote. You'll note that those arguments for inclusion weren't for something that met wp:ver (=sourcABLE). they were that several aspects of the sourcING were also factors on the side of inclusion. One idea would be for wp:ver to remain silent on that topic. Another might be along the lines of "the nature and contents of the sourcing may be very relevant to any inclusion discussion" North8000 (talk) 11:45, 27 April 2017 (UTC)

Yes, those both seem very valid options. Do you think the wordings proposed in the Inclusion and exclusion section below can be shaped into something that follows them, and is agreeable to "your side"? Diego (talk) 12:02, 27 April 2017 (UTC)

Its worth a try.North8000 (talk) 13:07, 27 April 2017 (UTC)

I think that the one that is garnering some support fulfills the "remain silent" approach? North8000 (talk) 17:30, 27 April 2017 (UTC)

I think you're right about "verifiability is a reason for inclusion" ("v-is-arfi"? as opposed to "v-not-arfi"?) as being shorthand for this WP:WEIGHT argument. But that is actually the problem: "verifiability" is being used for purposes other than WP:V. I think the argument you point to can be quite valid (assuming the premises are correct), and I have no issue with that. The issue is a "rectification of names" matter, in calling some issue a matter of verificiability when it is really about due weight or notability or such. Saying that this argument does not belong in WP:V does not mean that the argument is no longer valid, only that it should be called by its rightful name, which is not "verifiability". The problem is not that the current wording affects another policy, but that another policy is getting sucked into WP:V. Distinguishing them does not alter any other policy. ~ J. Johnson (JJ) (talk) 21:36, 27 April 2017 (UTC)

@J. Johnson:I think that your logic is the soundest of all. My only quibble is not relevant here which is that there are (rightfully) more things in play than policies and guidelines in determining what gets into an article. My discussion is more of a pragmatic one. The previous wording did harm in the area, if only by inference. The current wording makes positive statements to fix the problem. I was exploring possibilities for a compromise which merely does no harm in this area. The "remaining silent" would be the simplest variant of this. Wording that acknowledges that other factors relating to sourcing could carry weight in the inclusion/exclusion discussion is IMO accurate but out of place in this policy, but does does no harm and could be a compromise to settle this. North8000 (talk) 23:45, 27 April 2017 (UTC)

I'm not quite clear on what you are saying here. I think it is not out of place to expressly mention that other (and I do emphasize other) "things", factors, considerations, and policies are pertinent to inclusion/exclusion discussions, and that there is no need to misuse this policy. To the extent we are denying a certain usage some editors feel is essential we really need to indicate the alternatives. To be silent on this leads to the current situation of ambiguity, where we have a clash of interpretations. ~ J. Johnson (JJ) (talk) 23:28, 28 April 2017 (UTC)

Inclusion and exclusion

I am wondering if the problem we are having reaching consensus stems from the fact that we mention the term "inclusion", without mentioning its flip side: "Exclusion". Perhaps doing so might help us to find common ground. So... I offer the following for thought, and as a (potential) alternative...

While a lack of Verifiability (on its own) is enough to justify the exclusion of material, the opposite is not always true. Verifiablitity (on its own) may not be enough to justify inclusion. This is because Wikipedia has other policies and guidelines that also play a role in determining whether some bit of information is included or excluded.

The exact wording may need additional tweaking, but I think something similar to this (mentioning both inclusion and exclusion) might clarify what the policy is actually trying to say. Thoughts? Blueboar (talk) 14:45, 25 April 2017 (UTC)

I like the effort. Here's with a tweak: "While a lack of Verifiability (on its own) is enough to justify the exclusion of material, the opposite is not always true. Verifiablitity (on its own) may not be enough to justify inclusion. This is because Wikipedia has other policies, guidelines and article development practices that also play a role in determining whether some bit of information is included or excluded. North8000 (talk) 14:55, 25 April 2017 (UTC)

"Verifiablitity (on its own) may not be enough to justify inclusion" has the same problems as "Verifiability is not a reason for inclusion". I agree with Blueboar that focusing the message in "there may be other reasons to exclude the content" is more likely to get us a consensus than "verifiability can't be used" to justify/support inclusion. I would summarize it removing that part: "While a lack of Verifiability (on its own) is enough to justify the exclusion of material, the opposite is not always true. Wikipedia has other policies, guidelines and article development practices that also play a role in determining whether some bit of information is included or excluded." Diego (talk) 15:44, 25 April 2017 (UTC)

I support this wording. It respects the two unspoken realities I most often observe: 1. The verifiability standard is not a reason for inclusion itself, rather, the fact that an editor has inserted a datum meeting the verifiability standard -- before or after being reverted -- is a reason for inclusion because it is an implicit assertion of the editor's assessment that it is relevant. 2. An insertion is reverted citing lack of verifiability often because that's the easiest way to get edits deemed violative or trivial prima facie out of the article summarily. If the factoid can't be verified, often that's the end of it -- as intended. Implicitly, however, the takeaway for the serious editor, especially if a newbie, is that proven verifiability will constitute a ground for retention, otherwise the revert would have been accompanied by a different justification. FactStraight (talk) 00:04, 26 April 2017 (UTC)

I support Diego's wording. Sławomir Biały (talk) 00:17, 26 April 2017 (UTC)

I think the sentence "Verifiablitity (on its own) may not be enough to justify inclusion." is important. I think it is the heart of the warning that we want to give editors (especially new editors). WP:V is usually the first policy a new editor encounters. A new editor quickly learns that he/she needs to add sources to support additions... but then new editors often feel blindsided and bitten when that verifiable information continues to be challenged (even though they have added a source). I think it vital to state upfront (and in blunt language) that Verifiability isn't always enough. Blueboar (talk) 11:55, 26 April 2017 (UTC)

I like Diego's better than Blueboar's(/mine). I think that that sentence does more harm than good.North8000 (talk) 13:02, 26 April 2017 (UTC)

I think that the current wording is better but could support Diego's as a final compromise to settle this. If this goes any further, we should be more precise about it. What would come out from exactly where and what would go in exactly where. North8000 (talk) 10:52, 26 April 2017 (UTC)

FWIW, Diego's looks good to me. (Another drive-by comment from – Philosopher ^{Let us reason together.} 14:43, 26 April 2017 (UTC))

Remove the words in parentheses. Spell out what's "the opposite". Replace "play a role" with "may determine". Remove "some bit of" and "While a". Change "is included or excluded" to "should be included or excluded". Use "information" because "material" appears to be contested (although I prefer "material"). And keep the last sentence of WP:ONUS of course: Lack of verifiability is enough to justify the exclusion of information, but verifiability is not sufficient to justify inclusion. Other Wikipedia policies, guidelines and practices may determine whether information should be included or excluded. The onus to achieve consensus for inclusion is on those seeking to include disputed content. Bright☀ 15:32, 26 April 2017 (UTC)

I like the shortened parts; but I'm with North8000 that it's a mistake spelling out an absolute "is not sufficient". If the goal is to warn newcomers that sometimes verifiable content will be further challenged, then such "sometimes" should be added to the wording.

Also, I'd like to add a Wikilink to WP:Consensus, given that it's a relevant part of the final sentence in ONUS. I think it would even make sense to include the initial sentence at Determining consensus, which clarifies how policies and guidelines apply to that consensus-building.

So this is how I'd have it: Lack of verifiability is enough to justify the exclusion of information, but verifiability may not always be sufficient to justify inclusion. Other Wikipedia policies, guidelines and practices may determine whether information should be included or excluded. The onus to achieve consensus for inclusion is on those seeking to include disputed content. Consensus is ascertained by the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy". Diego (talk) 15:55, 26 April 2017 (UTC)

I like this as a way forward. I'm concerned that the proposed phrasing may be inaccessible to some editors, particularly if English isn't their first language. I suggest using simple words and focusing on what editors can do or should do. Here's my draft:

Original	Reduces to
Lack of verifiability is enough to justify the exclusion of information	You may remove information that isn't verifiable.
but verifiability may not be sufficient to justify inclusion.	Even if information is verifiable, it can still be removed
Other Wikipedia policies, guidelines and practices may determine whether information should be included or excluded.	because of another rule, such as WP:NOT or WP:NOR.
The onus to achieve consensus for inclusion is on those seeking to include disputed content . Consensus is ascertained by the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy.	If there is a dispute, the disputed information should be left out unless there is "consensus" to keep it in. Wikipedia has special rules about consensus which are here.

Hope this helps—S Marshall T/C 16:30, 26 April 2017 (UTC)

I think that "practices" or even better "article development practices" should stay in. North8000 (talk) 16:37, 26 April 2017 (UTC)

The more accurate version is better. There are no "rules", there are policies and guidelines, and we shouldn't use second person. The sentence explaining consensus is detrimental, consensus should be explained on WP:CONSENSUS so there's no conflict between policies. Bright☀ 16:41, 26 April 2017 (UTC)

Yes, my more accurate version is better. ;) Seriously -- Wikipedia is full of rules. Some of them are called policies and some of them are called guidelines, and some of the guidelines (such as WP:N) are treated like policies, and some of the policies (such as WP:PRESERVE) are treated like guidelines; and there are occasional essays such as WP:ATA that are actually bright-line rules. Wikipedia is overflowing with rules, and it's okay to say so. Using the second person is certainly appropriate -- whyever would you think it isn't? Your position makes no sense to me at all. Since we're trying to be clear, it is in fact a great deal more appropriate to write in simple declarative sentences than needlessly wordy hortative circumlocutions such as "editors may". This is guidance for editors trying to work their way through difficult judgment calls, not a set of rules put up on an office wall. Someone reading it needs to know what they can and can't do, what they should and shouldn't do, and what they need to think about. (It wouldn't hurt us to say "please", either. These are volunteers and they're entitled to respect.) You and J Johnson will fight this, of course, because it's me saying it, but there are others I might just convince. :)—S Marshall T/C 20:38, 26 April 2017 (UTC)

S Marshall is it a coincidence that when (not) working towards consensus you use the wording that's given as examples of what not to do? WP:BEANS... Bright☀ 08:31, 27 April 2017 (UTC)

The sentence explaining consensus is copied verbatim from Determining consensus (the "Original" version at least). Diego (talk) 17:17, 26 April 2017 (UTC)

I think it better to refer to other policies / guidelines than to copy them to here. North8000 (talk) 17:30, 26 April 2017 (UTC)

True for whole sections, but I think copying a snippet or sentence here and there may clarify what part of the other policy is the most relevant, more than a single unqualified link. Diego (talk) 17:37, 26 April 2017 (UTC)

I agree that simplifying the language may be a good idea. Taking into account the above feedback, we have:

Editors may remove information that isn't verifiable. Even if information is verifiable, it can still be removed because of other Wikipedia policies, guidelines and article development practices.

The onus to achieve consensus for inclusion is on those seeking to include disputed content. Consensus is ascertained by the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy.

I like that "Even if information is verifiable, it can still be removed" is different from "verifiability is not sufficient to justify inclusion" and completely avoids "not a reason". Diego (talk) 17:32, 26 April 2017 (UTC)

I support this without the line break and without the last sentence. Put a wikilink for consensus in the previous sentence instead. Bright☀ 08:31, 27 April 2017 (UTC)

Support, with BrightR's fix. Also, IMO "isn't" -> "is not". I like that the focus is shifted to already written article text rather than to "preemptive hypothesizing" on what to include and what not. Writing and article is akin to brainstorming. A person finds a piece of info they think worth including. Some other thinks this is a worthless bit. Clearly a case of WP:CONSENSUS to work. IMO nothing shall prevent editors from reasonable, in their opinion, additions. Staszek Lem (talk) 17:21, 27 April 2017 (UTC)

@Staszek Lem: DO NOT EDIT MY COMMENTS. Bright☀ 13:36, 28 April 2017 (UTC)

@BrightR: DO NOT PANIC AND SHOUT and AGF. It is customary to highlight !ivotes, for clarity. Didn't you notice everybody else is doing so? If you think my edit distorted your intentions, I apologize. I will keep in mind that in the future. Staszek Lem (talk) 18:14, 28 April 2017 (UTC)

Tired of all these editors who should know better editing my comment to what I "really" mean. Don't edit other people's comments ever, with these exceptions. Bright☀ 18:25, 28 April 2017 (UTC)

OK. Sorry again. Staszek Lem (talk) 18:57, 28 April 2017 (UTC)

Agree with both BrighteR and Staszek Lem's fixes. A question about "article development practices", though: Is this a new term of art on Wikipedia, similar to "policy" and "guideline"? It sounds clunky and I would rephrase it unless it's widely used in other policies and guidelines. – Philosopher ^{Let us reason together.} 20:32, 27 April 2017 (UTC)

I made it up, it was an attempt to generally refer to the normal editorial processes, normal editorial decision-making processes...basically everything else that goes into building articles aside from / in addition to the constraints placed by policies and guidelines. North8000 (talk) 21:21, 27 April 2017 (UTC)

Support, for "Even if information is verifiable, it can still be removed because of other Wikipedia policies, guidelines and article development practices. The onus to achieve consensus for inclusion is on those seeking to include disputed content." I consider that "article development practices" include the to-and-fro process whereby specific facts, word choices, grammar and style are agreed upon, i.e. the collaborative editing which resolves disagreements not addressed by or too trivial to invoke a policy or guideline, but about which concurrence is necessary to stabilize an article's content. FactStraight (talk) 01:42, 28 April 2017 (UTC)

Support Without the last sentence. North8000 (talk) 11:37, 28 April 2017 (UTC)

Needs a fix of the issue raised by TransporterMan. North8000 (talk) 14:28, 28 April 2017 (UTC)

Weak support - Not really happy that this is being phrased as a permission focused on removers ("Editors may remove...") rather than as a warning to adders that removal might happen. But, I can live with it, if it will move us to consensus. Blueboar (talk) 12:04, 28 April 2017 (UTC)
Support my proposal, removing the last sentence as long as the wikilink to "achieve consensus" points to the section Determining consensus of WP:CONSENSUS (which focuses on explaining how to achieve consensus, rather than what is the concept of consensus). Diego (talk) 12:07, 28 April 2017 (UTC)
Oppose This can be read, because of the absolutism of "is not", to reverse the standard set in BURDEN that information can be removed simply because it is unsourced, with the only requirement (if that) being that you express a "concern" that it may not be verifiable. May not, not is not. I acknowledge that the link to BURDEN ought to fix that, but the words will be cited to the opposite effect and a complex argument will have to be made to refute it. This cures a minor drama by creating a bigger one. Regards, TransporterMan (TALK) 13:17, 28 April 2017 (UTC)

What if the first sentence is changed to "Editors may remove information if they think that it ~~may~~(P.S: might?) not be verifiable"? (or "if they have the concern that...", to keep it closest to the wording of CHALLENGE). Diego (talk) 14:12, 28 April 2017 (UTC)

Good point TransporterMan, that could really cause a problem.North8000 (talk) 14:28, 28 April 2017 (UTC)

perhaps "might" is a better word than "may"?... just a thought. Blueboar (talk) 14:31, 28 April 2017 (UTC)

Would that make a significant difference? The text at WP:CHALLENGE uses "may"... Diego (talk) 14:48, 28 April 2017 (UTC)

The problem is that it can be read to place a higher requirement for removal of material based on wp:ver grounds....putting the burden on the remover to show that it isn't verifiable. One idea "verifiability is a requirement for inclusion" ...I know this is retro but I think not in any disputed area? North8000 (talk) 15:03, 28 April 2017 (UTC)

I'm afraid that "Editors may remove information if they think that it may/might not be verifiable", while accurate, may result in an rebellion from the preservationist point of view here. The current hard-fought language in BURDEN strikes a subtle balance supporting the removalist point of view, but the suggested language can be read to tilt that balance even more in that direction. While I, as a removalist, am down with that, I even more do not want to argue about it here — again — for another six months. On another point, this language also suggests that existing sourced material in which the source is clearly not reliable or in which the source is challenged as being unreliable can simply be removed in a single step, rather than removing the bad or challenged source as a first step and removing the material only if a new reliable source is not provided. We've talked here about whether that should be a one-step or two-step process without, as far as I can recall, coming to a firm consensus. Again, as a removalist I'm down with a one-step and this language would support that being the base standard (but having said that I also agree that two-step is the better practice, though it should be as it is now optional, not required). Once again, however, I don't think that the preservationists are going to care much for that hardening of position. I've stayed out of this discussion, while watching it, because I thought we could live with either the pre-change language or the post-change language (though my choice would be the old title with the new text), but we need to be careful not to mess other stuff up. Best regards, TransporterMan (TALK) 16:01, 28 April 2017 (UTC)

I'm guessing not, but is "Editors may remove information they believe is not verifiable" any better? DonIago (talk) 16:59, 28 April 2017 (UTC)

Maybe we are getting off track. I think we're talking about an area that is not under debate, a sentence or phrase that is just a preface to the wording that is under debate. If so, perhaps we should just the old or current version preface sentence, e.g. "verifiability is a requirement for inclusion". North8000 (talk) 17:24, 28 April 2017 (UTC)

I'm with DonIago. "You may remove information if you don't believe it is verifiable." Or, if you have some fundamental objection to writing a clear procedure, I'd agree to "Editors may remove information which they don't believe is verifiable" as a less-good alternative.—S Marshall T/C 00:41, 29 April 2017 (UTC)

Now we're starting to discuss wording in an area that is not even in dispute. And the overall RFC being malformed (with no explicit resolution proposed) is not going to navigate us out of this. And so I think we're really getting mired down. But, on the positive side, I think that 90% or more of us are evolving towards something that might fly in the core areas of disagreement. I'm going to create a new section that that tries to gel something to move forward. North8000 (talk) 22:04, 29 April 2017 (UTC)

I thougt (a month ago) that we all understood the issue here to be solely about the inclusion (or exclusion) of verifiable material. As to unverifiable material: the lede says it may be removed. And in the case of contentious BLP material it asks for immediate removal. If an editor has a question of whether material is verifiable it can be tagged, and perhaps this should be mentioned even in the lede. But that's not what we are here for. The issue here is about the addition or removal of verifiable material, and please let's not get off-track. ~ J. Johnson (JJ) (talk) 22:03, 29 April 2017 (UTC)

Looks to me like the "off-tracking" can be resolved by distinguishing between potentially and actually sourced edits thus, ""Even if information is verified, it can still be removed because of other Wikipedia policies, guidelines and article development practices. The onus to achieve consensus for inclusion is on those seeking to include disputed content." FactStraight (talk) 14:46, 1 May 2017 (UTC)

I incorporated suggestions expressed so far in #Diego's version II. Staszek Lem (talk) 18:50, 1 May 2017 (UTC)

Moved below

Diego's version: Take II

Information that is not verifiable may be contested and eventually removed by editors. Even if information is verified, it can still be removed because of other Wikipedia policies, guidelines and article development practices. The onus to achieve consensus for inclusion is on those seeking to include the disputed content.

I tried to incorporate all fixes and objections expressed. Aside of style issues, my phrasing "may be contested and eventually removed by editors" is an attempt to address the objection by user:TransporterMan about the tilt towards 'removalism'. Staszek Lem (talk) 17:45, 1 May 2017 (UTC)

Looks like viable compromise. If you don't mind I'll copy it to below. North8000 (talk) 18:44, 1 May 2017 (UTC)

Try to gel and organize a compromise and move forward

First, on an organizational side, let's say that step one is trying to gel the "main statement" in the body of the policy. Then, once that is accomplished, we could discuss the details on the three places (lead, subheading and policy-body statement) involved. Second, if we could get an OK from 90% of the active participants, I think we could go slightly-bold and edit it.

How's this?:

Verifiability is a requirement for inclusion but not a reason for inclusion. Inclusion is subject to other policies, core guidelines and other article development processes. However, in inclusion/exclusion discussions, particulars related to the sourcing of material may be considerations or reasons for inclusion.

Sincerely, North8000 (talk) 00:07, 30 April 2017 (UTC)

You have an extra "for" in the first sentence, it appears. I also don't really care for the last sentence, though I'd be willing to defer to others on that one. DonIago (talk) 02:29, 30 April 2017 (UTC)

Thanks, I fixed it.North8000 (talk) 12:51, 30 April 2017 (UTC)

Nope. This is a step backwards, as it returns the word "reason". Multiple comments have objected to that word, saiding that verifiability can be "a reason". We are not going to find consensus if we use that word... so try again. Blueboar (talk) 01:28, 1 May 2017 (UTC)

I think the problem is not so much with the use of word itself, but with the differing views on whether verifiability is, or is not, a "reason". Avoiding the word leaves the differences hidden, and any supposed consensus is illusory if it only papers-over the differences. ~ J. Johnson (JJ) (talk) 23:44, 1 May 2017 (UTC)

No. It was noticed here the the term "inclusion" does not have a unique meaning, and people may be not aware during a discussion that they are speaking in same terms. Same with "reason". Staszek Lem (talk) 02:59, 1 May 2017 (UTC)

I suspect you mean "not aware ... that they are not speaking in same terms". In that case, yes, that is a good part of the problem, and why any discussion on this needs explicit qualification of terms. ~ J. Johnson (JJ) (talk) 23:50, 1 May 2017 (UTC)

I'm trying to wear a second hat,removing my "one side" hat and just trying to facilitate something. To save time and words I'm withdrawing the above. Based on one exchange above I mistakenly thought that having wording that says that sourcing can provide reasons for inclusion would assuage the concerns of person who don't like the current wording, but as elucidated by Blueboar and Staszek Lem I was wrong. Blueboar and Staszek Lem, I was wondering if you could give your opinion on the "Sorting out a few fundamantals" section above? This would help clarify if there is a an underlying fundamental disagreement or whether its mostly a matter of wording or other considerations. North8000 (talk) 11:56, 1 May 2017 (UTC)

I believe that #Diego's version is on the right track. However' judging from "oppose" arguments there (and the "opposes" in onther places) people want to squeeze all our major content policies into a single phrase. (Which is probably good to have, but quite hard). Staszek Lem (talk) 16:09, 1 May 2017 (UTC)

I also like Diego's version. I thought that the last sentence was a little awkward and possibly unnecessary, but had no specific objection to it. Also TransporterMan pointed out a specific problem which I missed, and which probably conflicts with wp:ver and so that will probably need to be fixed. But then if we try to write a more accurate of "may remove" we'll end up having write half of the whole policy in here. My draft above was a (botched) attempt to follow the principles of Diego's version. I also thought about about total removal of those type statements as a (compromise) possibility. I'll noodle on it. North8000 (talk) 17:21, 1 May 2017 (UTC)

OK so here is Staszek Lem's version (copied from above and then tweaked by them) which they evolved from Diego's version:

Information that is not verifiable may be contested and eventually removed. Even if information is verified, it can still be removed or moved to a different article because of other Wikipedia policies, guidelines and article development practices. The onus to achieve consensus for inclusion is on those seeking to include the disputed content. North8000 (talk) 18:48, 1 May 2017 (UTC)

current text

While content must be verifiable in order to be included in an article, verifiability alone is not a reason for inclusion, and does not guarantee that any content must be included in an article. Consensus may determine that certain information does not improve an article, and that it should be omitted or presented instead in a different article. The onus to achieve consensus for inclusion is on those seeking to include disputed content.

Support as a compromise I think that the current version is better but would support this as a compromise if that would settle it. North8000 (talk) 18:51, 1 May 2017 (UTC)
Support as the editor of modifications. Staszek Lem (talk) 19:09, 1 May 2017 (UTC)
Oppose due to inclusion of word "eventually", which implies that unverifiable information may not be immediately removed, which it can be (though there are better, though optional, practices). Best regards, TransporterMan (TALK) 19:41, 2 May 2017 (UTC)
- Our policies are to describe better practices, not what is actually and sloppily and aggressively and stupidly done. A better practice is to discuss text rather than slash and cut, unless it is outright nonsense. We don't have to squeeze all our guidelines about challenging the content into this small paragraph, which focus is role of WP:V. Staszek Lem (talk) 21:21, 2 May 2017 (UTC)
- P.S. Originally I did think of writing "contested or outright deleted", but I realized this this will be immediately objected by "preservationists", who will interpret as a direct invitation to delete everything not referenced. Staszek Lem (talk) 21:53, 2 May 2017 (UTC)
  - I concur with your PS, but over repeated, hard-fought discussions here we've created a series of acceptable and better practices and "eventually" can be easily argued to change that consensus. I continue to object. Best regards, TransporterMan (TALK) 15:05, 3 May 2017 (UTC)

Maybe our problem is that we're unwittingly trying to write a a policy summary instead of dealing with the core issue. The pre-February wording hinted that meeting wp:ver is a force for inclusion. The current wording states clearly that it is not. Maybe a simple middle ground compromise is to have wp:ver be silent on that topic. And so completely remove the disputed material, (plus one related 5th FAQ) and don't put anything in it's place. North8000 (talk) 11:30, 3 May 2017 (UTC)

Just want to repeat that the intent of the pre-February wording was definitely not to make a policy statement about inclusion (or exclusion)... the intent was simply to alert editors (especially new editors) to the fact that there is more to the whole mishmash of determining inclusion/exclusion than just WP:V. I have a feeling that some of us who have been involved in this debate (on both sides) have been over-interpreting the section (treating it, intentionally or unintentionally, as if it were some sort of WP:INCLUSION policy statement). Blueboar (talk) 12:21, 3 May 2017 (UTC)

I'm kind of worn out. Most of the attempts at compromise fatally waded into unintended other more complex areas. Of course all versions include that verifiability is a requirement for inclusion. The question is whether or not meeting verfiability carries additional weight for inclusion. The changes in February to the current version basically boil down to changing "not a guarantee" (which does not say but sort of implies that meeting wp:ver carries some additional weight) to "not a reason" which clearly says that it doesn't. Note that saying what it is "not" implies that it might be everything less "extreme" (lower on the below list) than that. This range could be (in order):

Guarantee
Force, strong argument for
Reason

Two areas of possible "middle of the road" compromise are

Use "force" (or something equivalent at that level) instead of "reason"
Keep "reason" but give it less prominence. Maybe take it out of the subsection title (but don't go back to the the other "not a guarantee" extreme) and/or take the new summary completely out of the lead.

If this looks to be of interest, I could ask the question in a way conducive to getting organized feedback. Or we could just take a rest and sit back and see what happens. With no action items proposed in the RFC (and thus none discussed) , my guess is that it will end up staying as the current version, and then on to a new RFC and another 100,00 words. Maybe that is a motivation for a compromise :-) :-) Sincerely North8000 (talk) 19:47, 4 May 2017 (UTC)

I don't know what kind of concessions could be made either way to make a compromise. The current state of this issue seems to be akin to which side of the road to drive on, and "down the middle" is not a viable option.

I think the best result we could hope for is that someone (with strong analytical skills, and lot of spare time) works out some statements of the underlying concerns, the philosophical differences, and examples showing actual problems, all in language that we all interpret in the same way. Then we would have a basis for starting a discussion. ~ J. Johnson (JJ) (talk) 23:35, 5 May 2017 (UTC)

Support current article language. The old "WP:V Classic" language can be reasonably misconstrued even by a seasoned admin. Many conscientious editors have understandably interpreted the old formulation to mean that providing an RS citation for a claim warrants that it "bears repeating in an encyclopedic article".

This misconception may then force the editor to discount the WP:UCS essay recommendation that you add (or remove) a claim only if you thereby "better inform your readers".

Illustration: A (formally) reliable source says that Syria dropped 12,960 barrel bombs last year, killing 635 civilians. You add that claim per WP:V Classic. You're thereby leading unwary readers to think that it takes 20 barrel bombs to kill 1 civilian; yet you reasonably believe you're just observing the policy. --Dervorguilla (talk) 07:26, 6 May 2017 (UTC)

I fear your example is somewhat spoiled in that the inference (calculation?) of the number of barrel bombs per civilian is likely covered by other policies, such as OR. A better example might be the addition of "Syria dropped 12,960 barrel bombs last year" (reliably sourced material!!) into, say, an article on Sex determination factors of some exotic butterfly. Lacking some showing of relevance (perhaps those bombs leave a residue in the main wintering grounds of that butterfly?) I would say verifiability does not warrant inclusion. Perhaps that is more to your point? ~ J. Johnson (JJ) (talk) 23:40, 6 May 2017 (UTC)

It doesn't even have to be on the wrong subject. We could probably find a reliable source for the location of certain highly public figures on every day. But does any article need a complete list of which days Donald Trump spent in which cities? Is any article going to be improved with a sentence that says "Donald Trump was in the White House last Tuesday?" WhatamIdoing (talk) 16:48, 17 May 2017 (UTC)

Disputed section header

I looked into the current version once again and I see we forgot the highly contested change of section header.

Now:
- Verifiability is not a reason for inclusion
Suggested:
- Verifiability does not guarantee inclusion

Staszek Lem (talk) 19:05, 1 May 2017 (UTC)

I opened with suggesting just getting the main sentences agreed on and then moving onto specific changes in three areas. I just realized that the above main body text is already ready-to-be specific, we just need to say "replace the body of the xxx sub-section with:". But that still leaves the sub-section title and the summary in the lead. Maybe it would be best to leave that go? I would oppose the above. While the body compromise wording sort of goes to the middle ground, this title IMHO doesn't, it goes back to implying that it is a force for inclusion. . Maybe a neutral one something like "Inclusion and exclusion" or "Verifiabiity, inclusion and exclusion"?North8000 (talk) 20:41, 1 May 2017 (UTC)

Call it "Removing verifiable information".—S Marshall T/C 23:33, 1 May 2017 (UTC)

It has been discussed previously that "guarantee" is a stronger word than "reason". And that the lack of a guranteee is implicit, as the alternative – that verifiability guarantees inclusion – is just what some editors have vigorously insisted no one is arguing for. Denying that verifiability is guaranteed (a strawman?) still leaves open its use as a reason, which is the essence of the issue. ~ J. Johnson (JJ) (talk) 00:06, 2 May 2017 (UTC)

It's not necessary to split that hair. A compromise is when we use language all sides can agree on. We won't agree on "reason" or "guarantee".—S Marshall T/C 19:38, 3 May 2017 (UTC)

It's not a "hair" when a certain editor (you) has called it a "drastic change" 22:02, 17 Apr, and "a novel interpretation that can't remain in this policy unless there's a community consensus" 18:10, 12 Apr. And it is already split, as reflected in the strong division of opinion on this specific point. As to using "language all sides can agree on": exactly. But we don't agree on "guarantee" either, so by your reasoning that is also out.

What you seem to not understand is that no words will suffice where we have fundamental differences in the concepts they are expected to express, or where we have different interpretations of the words. But possibly you were in a serial stream-of-consciousness editing mode and hadn't seen my prior comment just below here. ~ J. Johnson (JJ) (talk) 21:25, 3 May 2017 (UTC)

It seems that one side objects to the word "reason", and the other side objects to the word "guarantee"... may I suggest that the only way to find consensus is to find another word. One which neither side objects to. Blueboar (talk) 00:37, 2 May 2017 (UTC)

Since we're just talking about the section header here- "Verifiability as a basis for inclusion"? It makes no claims as to whether V is or is not actually a basis for inclusion, just lays the groundwork for the detailed information to follow...at least, that's how I read it. DonIago (talk) 05:44, 2 May 2017 (UTC)

That's getting close to a neutral ground. I think making it a question (if that is allowed in a title) would be even better. "Is verifiability a basis for inclusion?"North8000 (talk) 11:51, 2 May 2017 (UTC)

The words represent the different views (interpretations), which need to be resolved to a point of shared understanding, not avoided. If the policy is interpreted differently by different editors we have, effectively, different policies. And (quite importantly) the resolution has to be in (or at least mappable to) the language and terms each "side" uses, and address the issues raised. Without that, whatever words are used have no connection to the issues. ~ J. Johnson (JJ) (talk) 22:59, 2 May 2017 (UTC)

The header must succinctly represent the content. If "guarantee" is bad word then "basis" is just the same. A "non-prescriptive" variant would be:

"Verifiability of text and its inclusion".

(BTW, we habitually use this "and" trick in article titles to make them NPOV: e.g. Islam and violence") Staszek Lem (talk) 19:17, 2 May 2017 (UTC)

Looks OK with me. North8000 (talk) 19:31, 2 May 2017 (UTC)

Call it what it's about: "Removing verifiable information".—S Marshall T/C 19:38, 3 May 2017 (UTC)

Alternately: "Non-removal of verifiable information. "

Stuart, you are looking at only one facet of the issue. Which was explained to you back in February or so, but apparently to little effect. ~ J. Johnson (JJ) (talk) 21:29, 3 May 2017 (UTC)

Just so everyone knows... "Verifiability does not guarantee inclusion" was just quoted at RSN. Blueboar (talk) 18:16, 5 May 2017 (UTC)

I couldn't find it there. Do you mean someone saying "Reliability does not guarantee inclusion"? North8000 (talk) 18:51, 5 May 2017 (UTC)

See the last paragraph of the section on "Conflict between Primary and secondary sources" Blueboar (talk) 21:12, 5 May 2017 (UTC)

Comment by Kautilya3 (11:48, 5 May) at WP:RSN#Conflict between secondary and primary sources: "Not everything that appears in RS needs to be included in a Wikipedia article." ~ J. Johnson (JJ) (talk) 22:56, 5 May 2017 (UTC)

How to best close this discussion?

I discussed closing this discussion at User talk:S Marshall#Who to close the RFC discussion at Wikipedia talk:Verifiability and then requested teamwork closure at the WP:AN. Winged Blades of Godric said he'll volunteer as part of teamwork closure, so I'm awaiting one or two more closers. --George Ho (talk) 15:30, 3 May 2017 (UTC)

I think that our best hope is to try to work out a compromise. As I indicated above, the RFC just asks for an overall opinion on collectively several different changes made in February, without asking about or proposing any specific actions on any of them. If anyone is thinking that it is asking about any particular action, it is malformed with respect to that. And so it's hard to see any resolution come out of a close of the RFC as worded. North8000 (talk) 15:56, 3 May 2017 (UTC)

The fact that I didn't specify any actions in the question certainly does not prevent any actions in the close. It's for the closers to determine the actions that arise from this RfC. If there's a consensus, they will determine the appropriate actions based on the consensus they find; and if there's no consensus, they will decide what the status quo ante is, and restore it.—S Marshall T/C 17:04, 3 May 2017 (UTC)

My point is that they are unlikely going to extrapolate to actions from the RFC when it did not specify any actions for discussion. An so it would be a good idea to keep trying on a compromise. Sincerely North8000 (talk) 17:53, 3 May 2017 (UTC)

The problem with taking any kind of action that was not proposed in the RfC is the lack of proper consideration. That a possible action was raised in any subsequent discussion is inadequate, as many respondents will comment on the RfC as stated, and then move on, with no regard to any subsequent proposals.

Certain editors have argued that the consensus reached for the existing language was inadequate for something as important as WP:V because it had only eight or so editors. If that is so, then having only three admins vote on something that possibly wasn't on the menu would be even less adequate. The underlying issue needs resolution, but this RfC doesn't do that, doesn't even provide a basis for doing that. ~ J. Johnson (JJ) (talk) 21:36, 3 May 2017 (UTC)

Rather than pre-emptively challenging what the closers might or (might not) say... Let's wait, and see what the closers actually say. Blueboar (talk) 11:27, 5 May 2017 (UTC)

I wasn't meaning to pre-emptively challenge, only to show that any closure done on the basis of (say) three admins would be subject to the same "inadequate basis" objection raised against the current language, and thus a basis for challenge which ever way the result might go. ~ J. Johnson (JJ) (talk) 22:54, 5 May 2017 (UTC)

J. Johnson, the above is too much to read, but I'm having difficulty seeing what the substantive difference is between the versions. Both mean that verifiability is a necessary but not a sufficient condition for inclusion. What is the difference between the versions such that it has sparked so much discussion? SarahSV ^(talk) 21:40, 16 May 2017 (UTC)

Yes. On all accounts. With the key differences being deeper than the words.

It is difficult to see anything in this RfC. This is due to a number of reasons, starting with poor formulation and total lack of definition of key concepts and terms. Discussion has gone into the ditch several times because of different interpretations of various terms, failure to identify key caveats and assumptions, failure to observe those terms (etc.) that have been identified, and editorial malfeasance. In the discussion last February the intent was not to change the policy, but to clarify it. But it seems that unclear language has allowed ambiguous interpretations; clarifying it has exposed those fundamental differences. It's not so much the words themselves, but the strong feelings as to what they mean.

There is also a deep divide in fundamental philosophy, such as whether certain material should be "guaranteed" a place "somewhere in the encyclopedia". As it is, a bunch of editors have been arguing at cross-purposes, and about supposed cases that don't seem to exist. Without resolving – which would start by identifying – all these differences, I think the only compromise possible is one where the language (whether the older form, or something new) is again so ambiguous everyone think it supports their interpretation. Which only puts the bickering off to individual cases where the same language is invoked for different ends. ~ J. Johnson (JJ) (talk) 23:59, 16 May 2017 (UTC)

Several responses, in both the yes and no columns, could apply to either version. SarahSV ^(talk) 00:48, 17 May 2017 (UTC)

If the whole discussion is very difficult to summarize and close, is closing separate subsections, including the original discussion (at the very top), possible? If so, which others besides the very top one need separate closures? George Ho (talk) 03:07, 17 May 2017 (UTC); struck. 19:28, 17 May 2017 (UTC)

It's not in any way difficult to summarise or close. It's bloated but not complicated at all. Despite J Johnson's continuing refusal to understand this point, nobody at all thinks verifiability should guarantee inclusion and nobody at all has said so. During the discussion, I said that material that's genuinely attributable to a truly reliable source almost always belongs somewhere in the encyclopaedia; but the language I and several others advocate in the policy is that verifiability is not a guarantee of inclusion.

I, and certain editors who agree with me, object to the phrase "verifiability is not a reason for inclusion", mainly because we feel that verifiable information is more includable than information that isn't verifiable. The other side disagree.

Nothing about this discussion is hard to understand or complicated to close. The reason for requesting a triumvirate close is to put the conclusion beyond doubt and enable this discussion to come to an end.—S Marshall T/C 18:35, 17 May 2017 (UTC)

Got it. Thanks. ...Um, I already know that the original question is simple, but thanks anyway. :) George Ho (talk) 19:28, 17 May 2017 (UTC)

What is bloated is this discussion. As to ending it, that is very simple: just stop. As for a "conclusion beyond doubt" – hah. There is no basis for that, as there has been no resolution of anything. As I noted just a few paragraphs above, ANY result here regarding any wording will be challengable. And the strong and substantial objections ON BOTH SIDES shows there is no consensus, either way.

Marshall, I am not "refusing to understand" any point. At nearly every point where our "understandings" seem to differ I have tried to sort them out. If my understanding is in any way incorrect or deficient, it is likely attributable to your failure to explain. For the benefit of any newcomers I will note that in the discussion leading up to this RfC (now archived) you refused to clarify your ambiguous statements. Particularly, at 23:57, 25 Mar you said: "I decided not to get sidetracked into a semi-relevant discussion and didn't answer some questions."

That you say "[n]othing [here] is hard to understand or complicated to close" doesn't make it true. Of course, it may seem that way if you understand only one point of view. If you think it would be "not difficult" to summarize all that has been argued here, please do so. But it would be no basis for closing if it is incomplete or one-sided. ~ J. Johnson (JJ) (talk) 23:19, 17 May 2017 (UTC)

I think the (unstated, but obvious) question that lies at the heart of this RFC is: Should the policy a) go back to the pre-February language, b) continue with the post-February language, or c) use some other (yet to be determined) language? That is essentially what we have been debating all this time. If the closers can give us an answer to that question, then we will know how to move forward. Blueboar (talk) 00:16, 18 May 2017 (UTC)

The challenge is "unstated", plus only the non-specific is "obvious". Sincerely, North8000 (talk) 13:03, 18 May 2017 (UTC)

Yet a consensus of "no" to the RfC question would clearly reject option b), the text after the post-February change. Diego (talk) 13:21, 18 May 2017 (UTC)

Sorry, no, but "what we have been debating all this time" is NOT "obvious". Sure, we could even all agree that "it's obvious", but are we even talking about the same thing? Marshall certainly wants to revert the previous language, but that is not what he asked. He asked (regarding the change): "Q: Was this a good idea?" Whether it "was a good idea" and whether it "should be reverted" are two different questions. And it is quite conceivable that one might think "it was a good idea, but should be reverted", or that "it was not a good idea, but should not be reverted". To ask one question, then apply the answers to a different question, is your basic switch-and-bait tactic. It's saying one thing, but meaning something else. Yet another reason why the past three months of discussion here is worthless. ~ J. Johnson (JJ) (talk) 08:47, 19 May 2017 (UTC)

You're right that, "the change is a bad idea" is a different question than "the change should be reverted". But it is also obvious that if the change was a bad idea, it can't work as the guideline and can't stand there, as it wouldn't represent the community consensus. This leaves us with only options a) and c) above, either reverting or changing it to something else. Diego (talk) 09:49, 19 May 2017 (UTC)

Yeah, the RfC question, the responses, and what to do if there is consensus are perfectly clear. The !votes are pretty evenly split, so consensus is not immediately clear, nor is it immediately clear what to do if there is no consensus (though I have my own opinion), but that's for the team of admins to decide.--Wikimedes (talk) 10:12, 19 May 2017 (UTC)

I think that J. Johnson's post goes to the core of the formation/malformation of the RFC. If some were advocating the closers dictate action on the project page, they are basically advocating a new RFC with participation limited to the three closers,(presumably with input from this RFC) not a closing of this RFC. This gets stretched even further when one notes that there is really no consensus either way in this RFC. Looking at it from another angle, the question would become whether what was developed in the February process has standing vs. being considered to be a "new" proposal. North8000 (talk) 10:47, 19 May 2017 (UTC)

The admins who have been asked to close this RFC can read the discussion and reach their own conclusions as to how to close it. It may be that the original question was malformed... but the closers are intelligent people... they can take that malformation in stride, and answer the unstated question that lies behind the stated one. To argue that they are limited to only answering the stated question is pure Wikilawyering. It is natural that each of us wants them to close this RFC in a way that supports our own opinions... and we have all repeatedly stated what those opinions are. Now, I suggest that we back off, and let the closers do their job. Blueboar (talk) 12:58, 19 May 2017 (UTC)

I don't think that anyone has argued the extreme they they are so limited, and I think that nobody has sunk to the normal meaning of wiki-lawyering, and so I would take issue with the use of that term here. And BTW my highest hope is that it ends up with none of the nice & great people who have been participating here being hurt or upset. For me the result is secondary to that. North8000 (talk) 17:16, 19 May 2017 (UTC)

(ec)Has the closure discussion among admins begun? So far I see Winged Blades of Godric volunteering at Wikipedia:Administrators'_noticeboard#Needing_more_than_one_to_close_RfC_discussion_at_WT:V, but no one else. I'm worried that the request for closure has been forgotten, or that potential closers are waiting for the RfC participants to do something while the RfC participants are waiting for the closers to do something. On a related note, although the RfC discussion is sporadically and occasionally continuing, and there are probably many wording compromises that could gain clear consensus support, I don't see any compromise wording getting enough support within this RfC - there's just too much of a wall of text for enough people to be bothered to find a newly proposed wording and express an opinion on it.--Wikimedes (talk) 17:55, 19 May 2017 (UTC)

Diego's comment reminds me (and my apologies for letting this slide earlier) of a further ambiguity: one can object to the change in respect of its substance, or its form or process. E.g., one of Marshall's objections is that the February change was not done with an RfC. That objection could be made even if no one objected to the content or effect of the change. To object to some supposed defect in the process does not disqualify the change if the defect is resolved.

Blueboar: I did not say the closers are limited in what they can do. I'm saying that the discussion is inapplicable to any question but the one presented. Otherwise it's like passing a law without an official version, and everyone has their own private version of what it means. It's bait-and-switch, and it's hardly wikilawyering to object to such nonsense. ~ J. Johnson (JJ) (talk) 23:05, 19 May 2017 (UTC)

I'll let the closers decide how they will close. Nuff said. Blueboar (talk) 23:22, 19 May 2017 (UTC)

Another news, everyone: Tazerdadog will be the second closer of the teamwork closure, teaming up with Godric. --George Ho (talk) 02:34, 6 June 2017 (UTC)

I saw confirmation that Primefac will team with Tazerdadog and Godric. --George Ho (talk) 04:08, 14 June 2017 (UTC)

New proposal

A small group of editors — S Marshall, North8000, and I, TransporterMan — representing, I believe, most if not all sides of this discussion — have been working together to try to come up with a version of the section which satisfies all of our concerns. We can all support the following and here propose it to the larger group for consideration. Alanscottwalker also participated but neither supports nor rejects the proposal.

The entire current "Verifiability is not a reason for inclusion", including the title, would be replaced with the following:

"==Verifiability as eligibility=="

"Only verifiable information may be in Wikipedia. Even if information is verifiable, this does not mean it must be included or kept. Inclusion ~~Whether information that is verifiable will be included, excluded, or removed~~ is determined by other policies and guidelines and consensus. The onus to achieve consensus for inclusion is on those seeking to include disputed content."

The current "see also's" and shortcuts would be retained

Offered for your consideration. Best regards, TransporterMan (TALK) 01:15, 20 May 2017 (UTC)

After further discussion we are modifying our proposal as indicated by the underlines and strikeouts noted above. See that discussion for the reasons for the change. Best regards, TransporterMan (TALK) 19:19, 7 June 2017 (UTC)

Support:

~~S Marshall (talk · contribs), North8000 (talk · contribs), TransporterMan (talk · contribs) — As proposers.~~
- Support as modified, not as originally proposed. S Marshall (talk · contribs), North8000 (talk · contribs), TransporterMan (talk · contribs) — As proposers. 19:19, 7 June 2017 (UTC)
This is an improvement over the current wording. Sławomir Biały (talk) 12:20, 20 May 2017 (UTC)
Do I think this suggestion is perfect? Nope. But I do think it is acceptable. More importantly, it will give us way to move forward while we continue to discus those small tweaks that might make it even better. In other words... it is good enough for me to support. Blueboar (talk) 14:12, 20 May 2017 (UTC)
While concurring that the language can still be improved, I find it summarizes the degree to which our discussion yields agreement substantial (though not unanimity). I note valuable observations in the opposition arguments about improving process, pursuing a more definitive and/or useful consensus, but not, to my mind, objections that render the proposed wording inaccurate, unuseful or lacking in apparent consensus among current discussants. I dissent from tossing out the fruit of this thoughtful, wide-ranging effort toward resolution. I'm fine with saying "This is what we substantially agree upon in this matter thus far." FactStraight (talk) 18:43, 26 May 2017 (UTC)

Neutral:

Alanscottwalker (talk · contribs)
Doniago (talk · contribs) - I wanted to be a Support, but I'm not a huge fan of "eligibility" in the header as-is...I'd personally prefer "eligibility for inclusion". I also think "should" would be a better word-choice than "will" in the third sentence. Also in the last sentence, I think "burden" might be slightly more readable than "onus". Overall, I'm alright with this, but I just think there's too much room for improvement for me to be comfortable listing myself under Support. Noble effort though!
- We're stuck with that word, because WP:ONUS and WP:BURDEN are different sections. WhatamIdoing (talk) 07:32, 28 May 2017 (UTC)

Oppose

The wording is bad. "Eligibility" is the wrong term here; just because something is eligible doesn't mean it should be on Wikipedia. Anything may be on Wikipedia, the question is whether it should be, and the current wording does a much better job explaining that consensus etc determines what should or shouldn't be in an article on Wikipedia, as opposed to what may be or what is eligible.

Additionally, saddling this proposal to the discussion above is incredibly counter-productive, instead of waiting for its (non-)conclusion and starting fresh. Splitting off the discussion to a user talk page, not letting the original discussion come to a conclusion, tacking on a proposal-within-proposal... this all reeks of WP:GAMING. Bright☀ 04:28, 20 May 2017 (UTC)

It was not planned, just an opportunity that popped out of a 2 person discussion which I lurked into. North8000 (talk) 12:14, 20 May 2017 (UTC)

We don't need both a "Reject" and an "Oppose" section above, so I'm removing the reject section. I invite BrightR to bullet-point his oppose entry to keep the formatting consistent. Regards, TransporterMan (TALK) 16:14, 20 May 2017 (UTC)

Oppose, but could support in a new RfC with a few stipulations:

1) This RfC closes and this new proposal is presented in a new RfC. The change needs wider support than is likely to happen in a new proposal within this very long RfC. Also, it’s been a long hard road for the major contributors to the discussion, and this can lead to a strong desire to reach any resolution, even a compromise that is worse than either of the two current versions – fresh eyes are needed for a sanity check.

2) The proposal includes a corresponding change to the lede. The February changes to the section header and the lede were the major problems with the changes (for me anyway).

3) Both the pre-February version and the new changes are included as options in the next RfC. Both have had considerable support in this RfC, and it wouldn’t be right to throw one or the other of them out of consideration.

I see this proposal as good groundwork for the next RfC, rather than something that can be considered as part of this RfC. Continuing the groundwork, I'd like to point out that there was a version of the lede that had widespread support among participants back in the February discussion. This reads: “However, being verifiable is not sufficient for including something.” (This is somewhat different than February change of “However, while verifiability is required for including something, it is not a reason for inclusion.”) For the section header, this could read “Verifiability is not sufficient for inclusion”. This might be a good option for the next RfC.--Wikimedes (talk) 18:53, 21 May 2017 (UTC)

That sounds like an excellent plan for achieving a no-consensus result. WhatamIdoing (talk) 07:32, 28 May 2017 (UTC)

Oppose. The proposal has some merit, but lacks a critical element: a shared (common) understanding of what it means and implies. And it suffers by any connection with this current discussion. In that regard I concurr with Bright*'s assessment ("saddling this proposal to the discussion above is incredibly counter-productive ...") and Wikimedes' stipulation ("This RfC closes and this new proposal is presented in a new RfC.") This discussion has run for three months with nothing to show but how fundamentally conflicted we are on this policy; it is no basis for any kind of result.

It is an error to polish the language before agreeing on the policy and implied practices the language describes. We need to work on reconciling our different views on what should be included, the basis for including or deleting, etc. But prior to anything else we need to:

1) Close this RfC. Give it a decent burial to avoid spoiling any subsequent discussion.

2) Take a break. A rhetorical fire break as I suggested a month ago, so that when we come back to this topic (hopefully refreshed) we will be less likely to carry over the bad feelings and bad rhetoric that has poisoned this discussion.

Then we can (and should) proceed to deal with the questions of:

3) What should the underlying philosophy of inclusion be?

4) What kinds of situations or edits are we concerned about, and what kinds of edits should be allowed or disallowed?

5) What are the terms (language) by which we express the foregoing?

Only then will we have an adequate foundation and tools for working out the policy. ~ J. Johnson (JJ) (talk) 21:12, 21 May 2017 (UTC)

I wasn't going to comment on the procedural points, but the fact is that RFC's are open until they're closed. There's no semi-closed-waiting-for-formal-closing status. And until they are, in fact, closed they're open for further discussion and proposals. IMHO, the only reasonable conclusion on the current RFC prior to our new proposal was "no consensus". Whether you agree or disagree with that evaluation, our only intent and motive was to add another proposal which might actually result in a consensus. It's disappointing to see that folks are opposing it not on the basis of its merit or lack thereof but instead on the basis of some self-defined criteria about how RFC's are supposed to work (and, indeed, I would invite any closers to take into account the degree to which any opposes here are based on procedure/bureaucracy rather than merit/what's best for the encyclopedia). But this is a wiki and people can opine on whatever basis that they care to do so, including the color of the font in which I posted the proposal. Regards, TransporterMan (TALK) 22:22, 21 May 2017 (UTC)

I believe that there's a distinction drawn between "closed" (boxed or 'hatted', if you prefer; the point is that further comments are discouraged as pointless) and "had a formal summary statement posted". It is possible to "close" or stop a discussion without (or in advance of) a "statement" being posted. WhatamIdoing (talk) 07:32, 28 May 2017 (UTC)

I just want to comment on something J. Johnson says in his five point list above. He asks us to consider "What should the underlying philosophy of inclusion be?"... this goes way beyond the original intent of the section under discussion. The original "Verifiability is not a guarantee of inclusion" section (pre Feb.) was intended to be nothing more than a brief warning to editors, alerting them to the fact that there are other policies and guidelines that have an impact on inclusion/exclusion. It was never intended to be some sort of underlying philosophical statement (as if it were our WP:INCLUSION policy). And I would object to any attempt to turn it into one. Blueboar (talk) 13:59, 22 May 2017 (UTC)

Indeed. But some of the divisiveness in the this discussion arises from different takes on what should be included (such as everything verifiable having "a virtual guarantee of inclusion somewhere..."). Resolving matters such as that (at least as far as verifiability is concerned) could reduce conflict at a higher level. Likewise for reaching a mutual understanding of key terms and caveats.

That such philosophical considerations go "way beyond the original intent of the section under discussion" is hardly a credible objection, as the discussion here goes "way beyond" the original question ("Q: Was this a good idea?").

Note that the February change was not intended to be make any change in policy (or philosophy), it was intended as a clarification. What seems to have happened is that making a more specific statement revealed differences of interpretation of the underlying policy.

TM: "what's best for the encyclopedia" would be better organized discussions that don't plop all over the rhetorical landscape to little result. Which is why we have established procedures (such as RfCs) and rules (such as WP:V). To imply (falsely!) that anyone is insisting on procedure over merit is quite ironic, as Marshall's objection to the undeniable consensus reached in February was that it did not go through the bureaucratic procedure of an RfC.

What BrightR, Wikimedes, and I are for is starting fresh. What we oppose is not so much this "new proposal", as the context of basing it on the current RfC. Let this and any other proposals have their own RfC. But we need a real break, not just a speed bump in a crappy discussion that has burgeoned well beyond any chance of summarization. ~ J. Johnson (JJ) (talk) 21:35, 22 May 2017 (UTC)

Regarding the core items under debate, yours and Bright Round's ideas are the ones that I most agree with. But the topic of inclusion in general is determined by dozens of policies and guidelines plus editorial practices in general. I think that the more specific topic is that wp:ver is often misused on this topic. Some hints in the mis-summaries that were changed in February contributed to this. So IMHO the February change changed those from "does harm" to "really fixes the problem". IMO the new proposal went to the middle ground of just "does no harm"; I supported it as a compromise to settle it, seeing this as an intractable mess at the moment. Not pushing anything, just explaining my thoughts. Sincerely, North8000 (talk) 22:24, 22 May 2017 (UTC)

The underlying problem is that different editors here have different ideas – philosophies — of what should or should not be included, and how that is to be effected. Unless someone can find a way of how to actually accomodate different policies and practices of WP:V and inclusion, any "compromise" on the wording is only being ambiguous enough that each side thinks they have it their way, only to continue to butt heads in actual practice.

One way of dealing with this is to define the issue of WP:V so narrowly that it does not reach the matter of what should be included (etc.). (And this is the desire of some of us, to remove WP:V from inclusion/exclusion debates.) This requires careful and strict specification of the scope of the issue, which the OP here rejected.

Alternatively, we can't reconcile the essential differences – i.e., compromise — without dealing with where they come from. I am not saying we must deal with every policy and guideline that touches on inclusion. But as long as we are are not actually on the same page there will be conflict, so we have to get to the bottom of all the differences.

And all of that would be better done in a new RfC. ~ J. Johnson (JJ) (talk) 23:42, 23 May 2017 (UTC)

Comments: When a (the) break is over, or a new RFC is started, I would like to state that one word should not be negotiable as indicated in bold. As I read; "Only verifiable information may be in Wikipedia. Even if information is verifiable, this does not mean it must be included or kept. Whether information that is verifiable will be included, excluded, or removed is determined by other policies and guidelines and consensus. The onus to achieve consensus for inclusion is on those seeking to include disputed content.". It does not matter if an agreement is made between editors, or with some admin ruling, consensus is the actual determination of inclusion or exclusion. Exclusion can be the result regardless of the source. All the rhetoric about well sourced content belonging "somewhere" is irrelevant. In this such case, as the changing of policies or guidelines, according to current consensus that includes BRD, as soon as any changes were contested the content could have been reverted. But! that is another story. Otr500 (talk) 01:12, 28 May 2017 (UTC)

BRD is not policy, and it begins with warnings that it should not always been used. WhatamIdoing (talk) 07:32, 28 May 2017 (UTC)

Oppose. The onus is on the editor who wants to exclude relevant content because they want to whitewash the article. Any editor who whitewashes any article should not be editing or recommend any changes to any policy page. QuackGuru (talk) 19:36, 7 June 2017 (UTC)

You do realize, don't you, that the onus sentence is already in the policy — see WP:ONUS — and opposing this won't do anything about that? We just carried it forward in our proposal. Regards, TransporterMan (TALK) 21:21, 7 June 2017 (UTC)
- You do realise there are editors who are seeking to exclude content by disagreeing for no logical reason. I noticed they like to point to WP:CON and/or WP:ONUS. I noticed WP:CON has directly led to original research and content that failed verification. Its start off bad: "Only verifiable information may be in Wikipedia." If you want to get things going then we can start a new policy. Admins do not consistently enforce policy. We can change that. When an uninvolved admin supports content that failed verification they must be blocked. How about something like that? QuackGuru (talk) 00:18, 8 June 2017 (UTC)

re: "You do realise there are editors who are seeking to exclude content by disagreeing for no logical reason."... by who's definition of "logical"? I am sure that an exclusion based on WP:CON and WP:ONUS will be perfectly logical to most editors. If you disagree, then it's up to you to explain why a given exclusion isn't logical (or at least attempt to explain). Blueboar (talk) 14:55, 24 June 2017 (UTC)

WP:CON and WP:ONUS lead to things that are almost impossible to believe. There is now hoax content. It says when an editor clicks on the article: "Encyclopedic content must be verifiable." No, the WMF does not enforce that notice. What a joke. I am not obligated to fix it. QuackGuru (talk) 15:51, 24 June 2017 (UTC)

How about some diffs to prove that "WP:CON and WP:ONUS lead to things that are almost impossible to believe. There is now hoax content."? If that assertion is true, then the diffs will show that the application of those policies resulted, after a consensus discussion, in hoax content which has persisted here despite being substantively challenged. — TransporterMan (TALK) 17:32, 24 June 2017 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Proposal to draft an RFC re verifiability as reason for inclusion

My bad. I read that as no more than a fortnight. We'll come back to this in a nonce. - TransporterMan (TALK) 03:00, 2 July 2017 (UTC)

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

This is a proposal for an RFC to try again to resolve the verifiability-as-reason-for-inclusion issue, as recommended by the closers of the "Recent changes to policy about verifiability as a reason for inclusion" discussion, above. This is not an RFC but a discussion to draft an RFC. Please feel free to suggest other methodologies or to add additional options to the options set out below. Please do not remove or modify any of the options set out below (if you wish to propose a variation on one of those options, make it a new option, don't modify the existing option), but I encourage you to feel free to add additional options and add an origin description for them in the introduction if you care to do so (and I hereby waive the talk page guidelines regarding modification of others' talk page postings to the extent of addition of additional options and description, but I do not waive them for removal or modification without my consent).

The draft of the proposed body of the RFC is as follows:

Please choose one option from each part and feel free to present ranked choices. For example "Support 1B, 2B (first choice), 2A (second choice), 2C (third choice), and 3C" is a valid !vote.

The "A" options below are the current state of the policy (and being the same as the state of the policy prior to the February 27, 2017, changes which set off this discussion but which were subsequently reverted ). The "B" options are the state of the policy made by the February 27, 2017, changes. The "C" options in Part 2 and Part 3 are the changes proposed by S Marshall, North8000, and TransporterMan in the "New Proposal" section of the closed discussion (they made no specific proposal in regard to the lede, but Option 1B was in the lede at that time and they made no proposal to remove or change it).

Part 1 — Lede

Option 1A - Nothing in the lede about verifiability being a reason for inclusion or not
Option 1B - Inclusion in lede: "However, while verifiability is required for including something, it is not a reason for inclusion."

Part 2 — Section header

Option 2A - "Verifiability does not guarantee inclusion"
Option 2B - "Verifiability is not a reason for inclusion"
Option 2C - "Verifiability as eligibility"

Part 3 — Section body

Option 3A - "While information must be verifiable in order to be included in an article, this does not mean that all verifiable information must be included in an article. Consensus may determine that certain information does not improve an article, and that it should be omitted or presented instead in a different article. The onus to achieve consensus for inclusion is on those seeking to include disputed content."
Option 3B - "While content must be verifiable in order to be included in an article, verifiability alone is not a reason for inclusion, and does not guarantee that any content must be included in an article. Consensus may determine that certain information does not improve an article, and that it should be omitted or presented instead in a different article. The onus to achieve consensus for inclusion is on those seeking to include disputed content."
Option 3C - "Only verifiable information may be in Wikipedia. Even if information is verifiable, this does not mean it must be included or kept. Inclusion is determined by other policies and guidelines and consensus. The onus to achieve consensus for inclusion is on those seeking to include disputed content."

Best regards, TransporterMan (TALK) 22:13, 1 July 2017 (UTC)

This is confusing and does not explain what is in the policy and what is not in the policy. QuackGuru (talk) 22:22, 1 July 2017 (UTC
- Clarified. If this is still too dense it might be well to include a detailed history. - TransporterMan (TALK) 23:25, 1 July 2017 (UTC)
  - There are no links to specific sections. If it is not easier to understand then I can't comment on any proposals without doing more digging. QuackGuru (talk) 01:37, 2 July 2017 (UTC)

Too soon... the close tells us to wait at least two weeks before discussing this again. Blueboar (talk) 00:18, 2 July 2017 (UTC)

I read the closing comments as no sooner than two weeks before starting a new RfC. That doesn't mean, to me, that discussions regarding the form of said RfC couldn't occur. Though my gut feeling is that this is too soon. DonIago (talk) 01:49, 2 July 2017 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Considerations and proposals at Wikipedia talk:Recent years

The following considerations and proposals are discussed:

--George Ho (talk) 02:54, 3 July 2017 (UTC)

Locked again

I've fully locked the policy again, this time for two weeks, because apparently editors cannot agree if any consensus was obtained on particular material. If this recommences after expiration of the protection expires, I will take it to WP:AN and let other administrators deal with determining how best to go forward, including whether editors should be sanctioned (sanctions don't necessarily mean blocks - it could mean warnings, for example).--Bbb23 (talk) 20:11, 2 July 2017 (UTC)

There is a small group who continue to propose changes and continue to fail to gain consensus. They don't know when to move on. QuackGuru (talk) 20:14, 2 July 2017 (UTC)

I accept the closing and will not challenge it, but the proper procedure for challenging it is set out in the "Challenging other closures" section of CLOSECHALLENGE, not by reverting it. If anyone does choose to challenge it, including raising the matter with the closing editors on their individual talk pages, notice of each step in that process should be given to the rest of us here on this page. Best regards, TransporterMan (TALK) 15:38, 3 July 2017 (UTC)

What do you mean by "not by reverting it"? QuackGuru (talk) 16:17, 3 July 2017 (UTC)

Challenging the close by reverting the edit to the policy made by the closer to implement the close. — TransporterMan (TALK) 16:39, 3 July 2017 (UTC)

Who is "Challenging the close by reverting"? QuackGuru (talk) 16:44, 3 July 2017 (UTC)

I didn't say that anyone was, maybe they have, maybe they haven't, but I'm looking forward, not back. — TransporterMan (TALK) 19:08, 3 July 2017 (UTC)

But you did previously say there was reverting. That was incorrect? QuackGuru (talk) 16:24, 5 July 2017 (UTC)

Was it incorrect that there was reverting? You're an experienced editor, you don't need my help to look at the edit history and figure that out. If you've got a point to make here, make it and stop the back and forth. — TransporterMan (TALK) 18:35, 5 July 2017 (UTC)

Which one is correct? This or this? The point is you can't answer a simple question and then you want to continue to try to change complex policy (without explaining what is the current problem)? QuackGuru (talk) 18:40, 5 July 2017 (UTC)

I'm not going to respond to that personal attack. Answer this simple question: Why engage in this inquiry in the first place? What's your point? — TransporterMan (TALK) 19:03, 5 July 2017 (UTC)

Protected edit request on 9 July 2017

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

Update link/avoid redirect in hatnote which clarifies ambiguity (change this line for the next):

Redirect|WP:PROVEIT|the editing tool|User:ProveIt GT
Redirect|WP:PROVEIT|the editing tool|Wikipedia:ProveIt

Widefox; talk 22:29, 9 July 2017 (UTC)

Done Callanecc (talk • contribs • logs) 12:29, 11 July 2017 (UTC)

"Verifiable" vs. "attributable" in the nutshell

I support "verifiable" but I did not edit it back to that because such is a change. North8000 (talk) 20:36, 28 May 2017 (UTC)

"attributable to reliable, published sources" is the definition of verifiable. Its redundant to say something is verifiable if it can be verified and redundancy tends to not give us much in the way of information.(Littleolive oil (talk) 18:33, 29 May 2017 (UTC))

Perhaps "confirmably attributable"? If I claim that a reliable published source once existed, but all copies have been destroyed, it's still attributable but no longer verifiable. – SJ + 20:22, 29 May 2017 (UTC)

Do I see you actually saying "verifiably attributable" ? :-) - This discussion's nutshell (nuts hell). Staszek Lem (talk) 21:46, 30 May 2017 (UTC)

:D The issue remains: how do we confirm that a cite accurately attributes a claim? (different from trying to confirm that the claim is 'correct'.) – SJ + 16:06, 5 June 2017 (UTC)

Duh. Slap {{Verify quote}} on it. :-) Suppose I tell you the rules how to confirm that a cite accurately attributes. Now, how to confirm that you accurately applied the rules to confirm that a cite accurately attributes? The issue remains: sooner or later everything boils down to the aggregate soft skills of the wikipedia community, involving domain expertise, comprehension skills, communication skills, etc. Staszek Lem (talk) 00:40, 6 June 2017 (UTC)

I support "attributable" because it better covers how we should phrase statements of opinion. Attribution is crucial in any controversial topic. It is important for editors both readers and editors to understand the distinction between a statement of fact, and a statement of opinion. When we cite a statement of opinion, we are not required to verify that the opinion itself is accurate... we are required to verify that the opinion exists, and that Wikipedia has summarized the opinion accurately. For that we need attribution. Blueboar (talk) 11:49, 31 May 2017 (UTC)

For purposes of citation and verification, is it really important to distinguish statements according to whether they assert opinions, or supposed facts? I agree that verification does not extend to determining the objective accuracy (or truth) of a statement (not necessarily limited to statements of opinion), but only to determining that a statement's use on WP is accurate. Given that, is it fair to say that attribution is saying who has asserted material (statements, etc.) and where it is found, while verification is the confirmation that the use of said material on WP is accurate? In that sense we could say that "all" (with the usual caveats) content must be attributed, and should be verifiable. ~ J. Johnson (JJ) (talk) 20:09, 6 June 2017 (UTC)

I can't agree with User:J. Johnson's statement 'In that sense we could say that "all" (with the usual caveats) content must be attributed, and should be verifiable.'

Since Johnson in the above paragraph mixes "verifiable" in the sense of the existence of a reliable source (the meaning used in this policy) with the sense of the statement being absolutely true, ' should be verifiable ' defies interpretation.

'"All" (with the usual caveats) content must be attributed' is wrong because it implies that most content must be attributed, with a few exceptions covered by the usual caveats. But the usual caveats (challenged, likely to be challenged, direct quotes) covers a minority of the text, so the use of the word "all" is so misleading as to make the statement false. Jc3s5h (talk) 20:23, 6 June 2017 (UTC)

You have seriously mis-taken my position. I said (bolding added): "verification does not extend to determining the objective accuracy (or truth) of a statement ...." In that regard I follow WP:NOTTRUTH that, for purposes of citation and verification, it is irrelevant whether a statement is "absolutely true". I have NOT mixed that with verifiability; I expressly excluded it. Your interpretation of this is simply wrong.

Your assertion that only "a minority of the text" must be attributed is rather troubling. ALL content is potentially challengable, even "sky is blue" stuff (see WP:NOTBLUE). While your "usual caveats (challenged, likely to be challenged, direct quotes)" closely follows the WP:V nutshell, you blithely skipped over the preceding sentence: "all material must be attributable to reliable, published sources." (Whether it should be so is a different discussion.) ~ J. Johnson (JJ) (talk) 22:43, 7 June 2017 (UTC)

JJ, I've been surprised to see you say this, because of all people, I expect you to read a text closely. This policy has never cared whether it's merely possible to challenge material. It only cares whether that has already happened ("challenged") or has a greater-than-50% change of it happening ("WP:LIKELY to be challenged"). Someone could make a WP:POINT of challenging anything and everything, but if (using your very best editorial judgment) you believe that it is unlikely to actually happen, then WP:V doesn't actually require you to supply an inline citation. WhatamIdoing (talk) 21:57, 5 July 2017 (UTC)

W: I have carefully read the text – in this case, the WP:V nutshell — and I believe I have a grasp of its meaning. But even after careful study of your comments it seems to me that you (and Jc3s5h) might be reading the text too closely, as in too narrowly. E.g., Jc3s5h objects to the statement "all content must be attributed", taking issue not with the implications of "ed" vs. "able", but with "all", arguing that the last sentence in the nutshell – "Additionally, quotations and any material challenged or likely to be challenged ..." – limits the scope of "all". To which I pointed out that the preceding sentence (the second sentence in the nutshell) clearly states that "all [emphasis added] material must be attributable ....".

Jc further argues that "the usual caveats (challenged, likely to be challenged, direct quotes) covers a minority of the text, so the use of the word "all" is so misleading as to make the statement false." To which I respond: ALL content is potentially challengeable. And now you assert that the policy is not about the mere possibility of challenge, but only "whether that has already happened ("challenged") or has a greater-than-50% change of it happening". Which is essentially the "challenged or likely to be challenged" of the nutshell, aside from your qualification of "has already happened" and your quantification of "likely" (an emendation not found in WP:V).

You say that this policy "has never cared whether it's merely possible to challenge material", as if the possibility is not a factor. but only "only whether that has already happened". That is a subtle mis-read, and misses my point: ALL content is potentially challengeable. You suggest it would be POINTy to challenge "everything", but please note: I am not suggesting that everything should be challenged, only that anything can be challenged. The sense I get from your comments is that some material is not challengeable, suggesting that if an editor can slip something in before a challenge "has already happened", it is permanently protected from challenge. Which is the implicit argument some editors have already made.

In summary: where you say that (in certain cases) "WP:V doesn't actually require you to supply an inline citation", I say: yes — until it does require it. ~ J. Johnson (JJ) (talk) 21:19, 7 July 2017 (UTC)

You know... sometimes I think all this would be clearer if we switched the two parts of the provision...

from: "challenged, or likely to be challenged"

to:"...likely to be challenged, or actually challenged."

The change in emphasis might change understanding. Something to think about. Blueboar (talk) 22:33, 5 July 2017 (UTC)

That's a good idea, Blueboar. Maybe "...any material whose verifiability seems likely to be challenged in the future, or whose verifiability has already been challenged"? WhatamIdoing (talk) 06:22, 14 July 2017 (UTC)

See Wikipedia talk:Verifiability#Attributability for the main discussion. QuackGuru (talk) 18:45, 1 June 2017 (UTC)

Protected edit request on 16 July 2017

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

https://sites.google.com/site/gyanjyotinainpur/ Dajharia (talk) 12:36, 16 July 2017 (UTC)

Not done: it's not clear what changes you want to be made. Please mention the specific changes in a "change X to Y" format. DonIago (talk) 16:15, 16 July 2017 (UTC)

Mistake in "reliable sources" section

The Reliable sources section of the project page currently states: "The word 'source' in Wikipedia has three meanings ... The type of the work (some examples include a document, an article, or a book) ...". This seems mistaken: a source is not a type of work, but the work itself. The latter part of the quote should be, for instance: A work (such as an article or book) ....

It's worth considering the text that immediately follows the full definition—and also that a prior version substituted 'aspects' for the word 'meanings'. Thus, before any mistake is corrected, it'd seem that some sort of decision has to be made about the aim of the opening paragraph—is it to define what a source is or is it to list the attributes of a source that can affect its reliability? Presently, it's a conflation of the two. Pololei (talk) 23:08, 4 July 2017 (UTC)

The definition given here seems to have diverged from its original... which was in line with WP:RS. I agree that we should revert to that older language so the two pages present the same definition. Blueboar (talk) 23:39, 4 July 2017 (UTC)

I agree: the original wording should be restored. Sławomir Biały (talk) 11:56, 5 July 2017 (UTC)

OK... since the page is currently protected due to edit warring... we will have to have an admin make the change.

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

The proposed change (below) is taken from the definition currently at WP:RS... and it is proposed that this language be restored here in WP:V#Reliable sources. Thank you Blueboar (talk) 12:49, 5 July 2017 (UTC)

The word "source" when citing sources on Wikipedia has three related meanings:

The piece of work itself (the article, book)
The creator of the work (the writer, journalist)
The publisher of the work (for example, Random House or Cambridge University Press)

I... I don't exactly disagree, but I think I might understand where the newer/wrong-er version came from. In some cases, the type of work (a message posted on social media vs. a scholarly publication) really does matter. The specific, individual work might matter less in some cases than whether it's a desirable "type" of work. WhatamIdoing (talk) 22:09, 5 July 2017 (UTC)

Done Callanecc (talk • contribs • logs) 12:34, 11 July 2017 (UTC)

Thanks to all for contributing. One belated response to the point made by WhatamIdoing is: the type of a work is implicit to the work itself. (That is, the proposed wording covers the type of work, just not explicitly.) Pololei (talk) 19:39, 17 July 2017 (UTC)

What is the goal?

Looking back at the previous RFC, I have to wonder if the reason we could not reach consensus is that we did not even agree on what the purpose of the section in question was. It seems we each had very different ideas on what the policy is trying to convey. I get the impression that everyone involved (and I do include myself) came to the discussion with a different goal in mind. We each may have had underlying agendas and concerns that influenced our preference when it came to language... and because we had different underlying agendas and concerns, we kept talking past each other in our discussions. We were not even trying to achieve the same goal. (For example, some may have had the agenda/goal of preventing the removal of verifiable information, others may have had the agenda/goal of allowing the removal of irrelevant information, and yet others may have had a different goal/agenda entirely).

In other words.. before we can even begin to re-discuss language, I think we need to come to some consensus as to what the goal (or the underlying meta-agenda) behind the section is. So... what do you see as being the goal of the section in question? WHY are we telling editors that verifiability is "not a guarantee" or "not a reason" (etc.)? Blueboar (talk) 13:59, 14 July 2017 (UTC)

I'm too worn out to engage in another debate or contest on this, but would like to engage in a friendly conversation to try to sort this out. This would be my answer / summary: It's starts with the overall objective which is to develop good articles that are in line with the Wikipedia constraints and objectives. There are editors who have and act on objectives which conflict with those, generally exhibited as efforts to POV an article. These can be via low level "wearing down" of the persons following the Wiki objectives, (e.g. maybe an edit accompanied by a wikilawyering statement where nobody wants to invest the time threshold to overcome that and fix it) up through a large intense debate and battle. An objective is to have the intent of Wikipedia policies, guidelines and objectives "win out" in these situations. Wikipedia policy sets verifiability as a requirement for inclusion. Nowhere does Wikipedia policy set verifiability as a force or reason for inclusion. Yet, there is a widespread, widely invoked misconception that such is the case. A common example of this is an edit summary of "undo removal of sourced material". This distorts the result of all of the above described situations, rather then using actual policies, guidelines and objectives to guide the result. IMHO a statement here at wp:ver, in the most relevant place like "verifiability is a requirement for inclusion, not a reason for inclusion" would help get rid of that urban legend / misconception and help all of those problems. Sincerely, North8000 (talk) 17:08, 14 July 2017 (UTC)

I'd like to fine-tune what Blueboar has proposed just a bit. The current purpose of the section is to refute the frequently-made argument that because something is verifiable (or, in this context, more often verified to a reliable source, but sometimes merely verifiable) that it is the obligation of those opposed to it being in the article to come up with a reason to exclude it rather than the other way around. (Here's a recent example of just that argument.) The real issue isn't, therefore, which way it should work, but whether we should change the way it works. Let me note that this would be a change that goes far beyond this one policy (see e.g. WP:CONSENSUS#No consensus) and would introduce a concept which is the reverse of how Wikipedia has worked up until now. I can't cite a source, but I believe that the philosophy has been that it is better for useful (or true) information to be excluded than to chance harmful (or false) information to be included. Putting the burden on opponents to justify exclusion of information creates an unreasonable risk of the latter. Why? For one thing, reversing the presumption says that you do not have to obtain consensus to include material. The mere fact that it is verified/able means that it goes in and that it is up to the opponent to achieve consensus for his/her reasoning that it shouldn't be included. The fact is that most editors who work here aren't policy wonks like those who regularly chime in here. When you create that presumption, you allow the inclusionist-who-knows-a-little-about-policy to force in material against the vast majority of editors who know little about policy. The dialog would go something like this, "That information isn't useful and shouldn't be in the article." "But it's verified to a unquestionably reliable source and is relevant to the article, X section of the Verifiability policy says that it must be included; here's what it says, "[quoting policy]]". "Oh." [Crickets; information goes in even though it has problems with NPOV, BLP, or whatever.] To my best knowledge we have never had a policy which requires particular information to be included in an article without consensus or anything more. To change this would create such a policy and would be a very, very bad idea indeed. Regards, TransporterMan (TALK) 17:14, 14 July 2017 (UTC)

There must be an obligation of those opposed sourced content being in the article to come up with a reason to exclude it rather than restoring unsourced content. Verifiability is being used to exclude sourced content and restore unsourced content because "consensus" must be obtained for the sourced content. QuackGuru (talk) 17:44, 14 July 2017 (UTC)

If that's happening, it shouldn't be. BURDEN makes clear that unsourced information should never be restored without a source. Whether the sourced material should or should not be included or not is an independent question from whether the unsourced material should be restored. If the sourced material is removed in that situation (correctly or incorrectly), it should not result in the restoration of the unsourced material. What we're talking about here isn't about that and the current policy handles that situation just fine. Regards, TransporterMan (TALK) 17:51, 14 July 2017 (UTC)

This is not about "unsourced information" without using a citation. This is about information that passes verification is being deleted while information that fails verification is being restored. Many editors add citations that do not verify the claim or rewrite the content from sourced content, to content that failed verification. QuackGuru (talk) 18:12, 14 July 2017 (UTC)

You are bundling two completely different actions (removal of material and addition of material) together as if they were a single action. May I suggest discussing them separately? North8000 (talk) 18:29, 14 July 2017 (UTC)

It is a single action when sourced content is being reverted while the content that failed verification is being restored. It is done by a single revert. QuackGuru (talk) 18:31, 14 July 2017 (UTC)

But what does that have to do with what we're discussing here? Regards, TransporterMan (TALK) 18:38, 14 July 2017 (UTC)

Verifiability says "Consensus may determine that certain information does not improve an article, and that it should be omitted or presented instead in a different article. The onus to achieve consensus for inclusion is on those seeking to include disputed content."

Verifiability is being used to restore content that fails verification. QuackGuru (talk) 18:46, 14 July 2017 (UTC)

But what you just quoted doesn't say anything about restoring material. I still don't understand how this has anything to do with what we're talking about here. Regards, TransporterMan (TALK) 18:51, 14 July 2017 (UTC)

Wait! I think I understand. Give me a minute to write a more complete response. — TransporterMan (TALK) 18:53, 14 July 2017 (UTC)

Is this the situation you're talking about? There's material in an article which says "widgets are grey" and cites a source. Editor A looks at the source and determines that it does not support the material. In a single edit, Editor A replaces the material with "widgets are red" and cites a clearly-reliable source. Editor B sees the edit and reverts it, citing ONUS. Is that what you're talking about? Regards, TransporterMan (TALK) 18:57, 14 July 2017 (UTC)

The name of the section is "Verifiability does not guarantee inclusion". But content that failed verification is often guaranteed inclusion. It says "The onus to achieve consensus for inclusion is on those seeking to include disputed content." Sourced content is being disputed, while the content that failed verification remains.

What is the situation I'm referring to? There's material in an article which says "the color of the grass is black" and cites a source. Editor A looks at the source and determines that it does not support the material. In a single edit, Editor A replaces the material with "the color of the grass is green" and cites the same source. Editor B sees the edit and reverts it, citing ONUS. Editors continue to replace sourced content with original research. It seems the lede is where this occurs most often. QuackGuru (talk) 19:59, 14 July 2017 (UTC)

HOLD ON... let's not get ahead of ourselves... above, TransporterMan states: "The current purpose of the section is to refute the frequently-made argument that because something is verifiable (or, in this context, more often verified to a reliable source, but sometimes merely verifiable) that it is the obligation of those opposed to it being in the article to come up with a reason to exclude it rather than the other way around." Are we agreed that this actually is (or should be) the purpose of the section?... I ask because this certainly wasn't the original purpose of the section (the original purpose was simply to inform editors that other policies besides WP:V can affect inclusion/exclusion... and to point them to those other polices.) Blueboar (talk) 20:01, 14 July 2017 (UTC)

Maybe we can add "There are other policies that affect inclusion/exclusion, including...." QuackGuru (talk) 20:06, 14 July 2017 (UTC)

Before we talk about adding or changing stuff... we need to be in agreement on why we are adding or changing it. I don't think we are (yet). Blueboar (talk) 20:18, 14 July 2017 (UTC)

Agreed. And that is one of the underlying issues. But a different one that I am trying to address. Would anyone mind terribly if I pop this section up a level so it can be its own thread? ~ J. Johnson (JJ) (talk) 20:28, 14 July 2017 (UTC)

Fine by me. Let's try to dissect this to keep it simple. Trying to get into overall processes for overall inclusion / exclusion would hopelessly mire this. North8000 (talk) 20:42, 14 July 2017 (UTC)

Blueboar, I might agree that you are right about the limited purpose of the section were it not for this edit on August 13, 2014, with the edit summary "Anything else is a POV-pusher's charter." That edit added the last sentence — the onus sentence — to the section and expanded the purpose to what I said above, that "it is the obligation of those opposed to it being in the article to come up with a reason to exclude it rather than the other way around." The edit summary clearly indicates that the addition was for that very purpose. That addition (which as far as I can determine was not discussed on the talk page) was not contested (there was an objection-via-revert, but the objecting editor immediately changed his/her mind and self-reverted his/her reversion). Regards, TransporterMan (TALK) 20:50, 14 July 2017 (UTC)

Although it may seem related, I think that that is a totally different topic. It basically puts an extra finger on the scale in inclusion/exclusion discussions. I'd say take it out, but that is a different topic. North8000 (talk) 21:17, 14 July 2017 (UTC)

The question asked by Blueboar was, "I think we need to come to some consensus as to what the goal (or the underlying meta-agenda) behind the section is" (emphasis added). So long as the onus sentence is part of the section it affects the meaning and purpose of the section. Regards, TransporterMan (TALK) 22:19, 14 July 2017 (UTC)

I feel that the goal of the section is to enable editors to reorganise, or in some cases to remove, sourced information. My concern is that there are a lot of reasons why editors might want to remove sourced information from the encyclopaedia. Some editors have a COI or want to advance a position, which is why I'm so anxious that this policy doesn't unambiguously empower editors to remove well-sourced criticism or controversies from articles about their favourite company, product, nation, religion or ethnicity. I think that removing sourced content should be subject to scrutiny, review and reversion, and if an editor does revert a removal of sourced content, then the policy should require talk page consensus to re-remove it.—S Marshall T/C 23:45, 14 July 2017 (UTC)

Well, that is most definitely a different view of the goal (then again, this is exactly why I asked) ... all I will say at this point is: if that is indeed the consensus goal of the section, it is a significant departure from its original intent. Not saying it is a good or bad departure... just noting that it is one. Blueboar (talk) 01:04, 15 July 2017 (UTC)

Intriguing. You wrote Verifiability does not guarantee inclusion, but you didn't think it was basically about removing verifiable information from an article?—S Marshall T/C 01:19, 15 July 2017 (UTC)

Yes and no... as originally intended, the section did relate to the removal of verifiable information (or more accurately the potential for removal), but the goal was not to "enable" or "allow" editors to do so. Here is the original text that I wrote:

While information must be verifiable in order to be included in an article, this does not mean that all verifiable information must be included in an article. Verifiable information that is deemed irrelevant or trivial to the topic of the article may be omitted. Whether specific information is irrelevant or trivial to the topic (or not) should be determined by consensus.

Note that it does not talk about "removal"... but about "omission"... there is a subtle difference in these two words. Also note that it highlights the need for consensus to omit. My intent was actually to separate the issue of inclusion/exclusion from the issue of verifiability. To alert editors to the fact that there were other issues, and other policies to consider besides verifiability.

That original language was soon clarified to:

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

I agreed with this change, because it better meshed what we said with WP:PRESERVE... and again the intent was to point the reader to inclusion/exclusion criteria in policies other than verifiability (in fact, it was at this point that the section started to included a "see also" note pointing to WP:UNDUE).

To put this another way... the long standing language was intended as a warning (or caution) that verifiable information might be omitted (due to policies other than verifiability)... it was never intended to be a justification for the removal of verifiable information, nor a counter to the "but it is verifiable" argument. That interpretation came in later. Blueboar (talk) 12:33, 15 July 2017 (UTC)

Okay. This is an example of the kind of problem that needs addressing in V. (I say "addressing", but I feel the "addressing" could be quite brief -- I would advocate one single paragraph of text headed "verifiability does not guarantee inclusion" accompanied by a prominent pointer to NPOV.)
Personally, my position is that removing "sourced trivia" is legitimate but removing sourced information that actually affects the bias of an article is something that should not be done without discussion. Defining "sourced trivia" will be the challenging part, as I would want a definition that permits removal of, for example, most of the pop culture references in sexuality in Star Trek and most of the statistics in electronic cigarette that have no thesis and reach no conclusion, provided the article remains NPOV when the editing is finished; but removing even one (1) sourced word is not permitted without discussion if it affects the article's balance.—S Marshall T/C 13:21, 15 July 2017 (UTC)

Um... the example you give has nothing to do with WP:V... the remover apparently focused on BLP as a reason to remove the information, not anything WP:V said. Now... the BLP argument may or may not have been valid (the complainer certainly thinks it wasn't... but I have no opinion on that, since I don't know the specific topic). That, however, is not something we can determine here at this policy... it needs to be discussed on the article talk page so that a consensus can determine the issue. I don't think there is anything we could say here in this policy that would alter that situation one bit. There is no way that this policy can resolve every inclusion/exclusion dispute (nor should it try to)... because there are so many other policies and guidelines that have to be considered. What we can do is point editors to those other policies, so they separate the issue of inclusion/exclusion from the issue of verifiability. Blueboar (talk) 14:00, 15 July 2017 (UTC)

We differ on this point, then. I feel that whitewashing by removing attributed information is something V should briefly discuss.—S Marshall T/C 15:26, 15 July 2017 (UTC)

There is no need for WP:V to discuss "whitewashing"... as it is already addressed by WP:NPOV. Again, this policy can not (and should not) try to resolve every inclusion/exclusion debate... because so many of those debates hinge on other policies and guidelines. Blueboar (talk) 14:12, 17 July 2017 (UTC)

If we're going to strengthen the paragraph headed "verifiability is not a guarantee of inclusion" then I do feel that whitewashing should be mentioned, along with the prominent pointer to NPOV that I advocated above.—S Marshall T/C 17:07, 17 July 2017 (UTC)

I am not at all sure we should "strengthen" that paragraph. If we change anything, I would prefer something closer to the original intent (a brief note, essentially telling editors that "inclusion/exclusion involves more than just verifiability... see NPOV, NOR, CONSENSUS and a host of other policies, guidelines and essays"). Blueboar (talk) 22:10, 17 July 2017 (UTC)

Stuart, are you now taking the position that something more than what we included in our "new proposal" is needed? Best regards, TransporterMan (TALK) 22:29, 17 July 2017 (UTC)

Oh, I'm sorry, I've been unclear. I like the "new proposal" and it would be lovely if editors could agree on it. That proposal wouldn't need the pointer to NPOV. But I suspect that in practice, the reason the new proposal failed to gain consensus because it's too inclusionist for the other side in this discussion. Therefore, if the discussion is ever going to end, then we'll need to agree on some form of wording that enables easier removal of sourced data ---- which is what I meant when I said If we're gong to strengthen.... "Strengthen" = "make it easier to remove attributed information" in my mind. I'm wanting a "whitewashing" clause in any beefed up "you can remove sourced information" paragraph. But if the paragraph isn't beefed up ---- if we go back to the new proposal and it gains consensus ---- then I'm quite content to leave out whitewashing.—S Marshall T/C 23:57, 17 July 2017 (UTC)

Thanks for the clarification, that helps. Best regards, TransporterMan (TALK) 16:37, 18 July 2017 (UTC)

Marshall: you might note the discussion above (at #Example #03) that "whitewashing" is always a POV issue. It might be that verifiability is raised in order to "whitewash" something, but if content cannot be verified when challenged it does not belong, regardless of the effect on POV. ~ J. Johnson (JJ) (talk) 23:51, 17 July 2017 (UTC)

On the other hand, editors "delete" sourced content without even giving a rationale explanation. This is done in different ways. If the can't flat out delete the content they rewrite it to be more vague or change the meaning from sourced content, to content that is unverifiable. I know that replacing sourced content with content that fails verification is not the subject of scrutiny. I think it should be. QuackGuru (talk) 00:54, 15 July 2017 (UTC)

I think that people agree that what you just described does deserve scrutiny, as two different actions. North8000 (talk) 18:16, 15 July 2017 (UTC)

Cases? ~ J. Johnson (JJ) (talk) 21:41, 15 July 2017 (UTC)

QG: where do "editors "delete" sourced content without even giving a rationale explanation"? You keep alleging stuff, but you have shown nothing. Put up or shut up. ~ J. Johnson (JJ) (talk) 23:58, 17 July 2017 (UTC)

Permitting use of established experts as references, if posted on social media?

Hi all,

A few months ago it was established here that "If social media are reliably associated with official spokespersons for the company, then see WP:SELFPUB" (Archive 65, Staszek Lem).

The ruling required that the relevant social media account could be verified as an official source of the claimed group; for example by checking earlier posts over an extended time frame to see if they matched in general style.

I'm proposing that this be extended to include "If social media are reliably associated with recognised experts who have previously had reliable works published in the relevant field, then see WP:SELFPUB and WP:POV".

The purpose is to cover expert work in between publications, i.e. say editions of a book are expected to be a few years apart, the work by that expert might be published on their facebook or twitter account in the meantime. We already permit post-publication works by experts - "Self-published expert sources may be considered reliable when produced by an established expert on the subject matter, whose work in the relevant field has previously been published by reliable third-party publications" (WP:V#Self-published sources) - so this ruling would clarify that such self-published works are permitted regardless of medium.

Thoughts? Anothersignalman (talk) 08:41, 3 July 2017 (UTC)

Already covered by the existing language... the need for "previous publication" is to establish that the scholar is indeed an expert. Once that requirement is met, then the scholar's self-published works (regardless of the medium) are also considered reliable - provided that the expert is talking about something within his field of expertise. This would include statements relating to his/her expertise posted on Facebook etc. Blueboar (talk) 11:08, 3 July 2017 (UTC)
I've had people delete my references to Facebook posts by previously-established experts, claiming that Facebook is not a reliable source regardless of author/content. Do you have examples I could use as case studies of FB references retained? (To be fair, some of those were from a locked group, but I don't see that as a problem; ref. WP:V#Access_to_sources "If you have trouble accessing a source, others may be able to do so on your behalf (see WikiProject Resource Exchange)"; that the vast majority of readers of the page would have or could gain access to the group; and that the references used relate to minor claims, not overarching statements underpinning the whole page's content). Anothersignalman (talk) 13:35, 3 July 2017 (UTC)
The fact that people have deleted references to Facebook does not surprise me at all... normally Facebook posts aren't considered reliable, and citations to them should be removed. The "expert exemption" is fairly rare.

My guess is that the editors you refer to either have not read the policy and thus did not know that we exempt experts from this restriction, or are cluelessly misapplying the policy because they are unaware that the expert in question actually qualifies as an expert. Those sorts of deletions will occur no matter what the policy says. Of course, it also could be that there is doubt that the sources actually qualify as experts (that I can not comment on).

The solution to this isn't to amend the policy (which already covers it)... the solution is to explain the existing policy on the article talk page, and establish that the expert in question qualifies... it also helps to gain a written consensus to return the citation (so you have something to point to if the citation is removed yet again). Blueboar (talk) 14:20, 3 July 2017 (UTC)
Thanks Blueboar. I'll ask for comments by Beachbo, Chrissymad, Huon, Trainsofvictoria and/or Voidxor, all of whom have been involved in discussions with me in the past about validity of references, before I restore the relevant sections. Anothersignalman (talk) 01:20, 4 July 2017 (UTC)
The specifics of the sources in question should be discussed at WP:RS/N; that's not about updating this policy and thus not relevant for this page. Huon (talk) 06:54, 4 July 2017 (UTC)
@Huon: Does that mean you agree with Blueboar's interpretation of the rules (that no new ruling is required because existing covers the issue), pending a post for each individual questionable source on WP:RS/N? Anothersignalman (talk) 23:45, 4 July 2017 (UTC)
I agree the proposed change in wording is unnecessary and redundant. That doesn't mean I agree your proposed sources are appropriate. Huon (talk) 23:54, 4 July 2017 (UTC)
Fair enough. At some point (probably today) I'll post examples of the references to WP:RS/N and see what happens from there. Anothersignalman (talk) 00:30, 5 July 2017 (UTC)

@Blueboar: I would remind you that calling other editors "clueless" is a violation of WP:NPA. See, we all overlook policies from time to time, even you. There are far too many nuances to Wikipedia policy to reasonably expect every editor to be familiar with every facet at all times. Rather than insulting editors acting in good faith, the correct thing to do would be to teach them the relevant policy or guideline. – voidxor 17:25, 4 July 2017 (UTC)

I did not call any editor clueless... I said that the editors in question may have cluelessly misapplied the policy (due to lack of familiarity with the qualifications of the source's author) ... huge difference. That isn't a personal attack - just a comment on a common error that any of us (yes, even me) could fall prey to. Blueboar (talk) 18:03, 4 July 2017 (UTC)

Already covered in Wikipedia:Identifying reliable sources#Exceptions. Bright☀ 12:34, 4 July 2017 (UTC)

Don't really see how that's directly relevant to social media presence - it talks mainly about sites wholly controlled by the expert in question, not social media posts by those people on external sites. Still, the principles match above, saying that an author who has previously had their works published can have future works considered "reliable", and that's useful in showing that my understanding of the relevant ruling isn't a one-off, but a common principle. Anothersignalman (talk) 23:18, 4 July 2017 (UTC)

The "generally accepted standard" WP:Identifying reliable sources, a.k.a. WP:RS, starts with: "Wikipedia articles should be based on reliable, published sources ..." (bolding in the original). What that guideline is a little weak on is the critical importance of publication. In particular, reliable derives not so much from the expertise of an author as from the process of the publisher: being subjected for various requirements (like citations), peer review, etc. Just because your expert is a Nobel lauareate, or has published extensive and well-regarded reviews of the literature, does not make him "reliable" for some pet theory. It's like waffles: waffle batter is not a waffle, not until it is baked, and then only if it's baked on a waffle iron.

As has been said: social media posts are accounted reliable in very limited cases, and mainly only as self-utterances. As anticipation of a forthcoming peer-reviewed (e.g.) publication: only to the anticipation. A pre-print that has not yet completed (say) peer review is not a reliable source, nor does a blog post have the same authority as a peer-reviewed article, even if it is an identical statement by the same author. ~ J. Johnson (JJ) (talk) 22:31, 5 July 2017 (UTC)

What you're calling "publication" (the process) is what Wikipedia's advice pages usually call "editorial oversight" or "editorial control". This policy defines what it means for a source to be published very explicitly: "Source material must have been published, the definition of which for our purposes is "made available to the public in some form". I don't know why this definition hasn't been copied into the WP:RS that purports to define publication (restraint on the point of avoiding redundancy is obviously not a significant consideration, given how many other parts of WP:V have been duplicated wholesale in the guideline over the years), but the definition still applies there. WhatamIdoing (talk) 06:34, 14 July 2017 (UTC)

Merely "made available to the public in some form" is the weakest from of publication, which includes everything posted on any blog, no matter how disreputable. "Publication" as used in the sciences (I am less familiar with other scholarly fields), particularly in regard of journals, is implicitly the result of an extensive process. Which is, indeed under "editorial control", but to reduce that process to just "editorial control" (or "oversight") is a disservice: it feeds the suspicion of some that science (or certain scientific fields) is "just the opinions" of a certain entrenched editorial elite with their own biases and agenda, no better than the opinion of anyone else "with eyes to see". The control is not for forwarding the editors' own opinions, but for ensuring adherence to an established process with a track record of leading to "true" results more often than not. As I said before, the WP guideline is a little weak on explaining this. ~ J. Johnson (JJ) (talk) 21:25, 17 July 2017 (UTC)

It may be "the weakest form of publication", but it's the relevant one for that statement. This statement exists to stop people from citing personal correspondence and other private documents. Some editors doubtless think it's sad that we have to bother saying that, but this policy has to cover everything, from "seriously, Karp may have told you that it was np-complete in the elevator, but your personal memory of what Karp said is not a reliable source" through "sure, sometimes you can cite Twitter posts and roadside signage, but we prefer organized publications, with authors and skilled editors and professional publishers and maybe even the occasional lawyer involved". WhatamIdoing (talk) 20:14, 5 August 2017 (UTC)

Info

Hi I have added affiliation of university with ref link for each from govt and authorize website et ne know why my information rejected from Wikipedia page Simons47 (talk) 06:04, 26 July 2017 (UTC)

You need to discuss this on the talk page of the relevant page. This talk page is for general discussion of the policies. --Muhandes (talk) 11:19, 26 July 2017 (UTC)

It looks like there are suspicions of socking. The "evidence" seems to be "someone is editing an article about a university in India", which seems a bit thin. WhatamIdoing (talk) 20:21, 5 August 2017 (UTC)

Proposal to resolve the underlying issues before anymore RfCs

Yay! The Rfc was closed (and hopefully buried). As I suggested back in May, that is the first step to making any progress. The second step is to take a break. I see the RfC closers suggested a new RfC "in no less than two weeks from now; (two weeks should be sufficient time for everyone to approach the wordsmithing with fresh eyes.)". Unfortunately, two weeks is not enough for new attitudes, or to do the fundamental work needed (which is not a mere matter of "wordsmithing") before we have any basis of a useful discussion; months still seem necessary.

Previous discussions on this topic have repeatedly come to grief because we are using terms in different ways, we have different ideas of what the problems are and what results are desired, and we don't even have shared examples of alleged problems. Any discussion that doesn't resolve these problems will be Yet Another Waste Of Time – YAWOT. Before diving into another RfC we need to (as I said before) resolve the following matters:

3) What should the underlying philosophy of inclusion be?

4) What kinds of situations or edits are we concerned about, and what kinds of edits should be allowed or disallowed?

5) What are the terms (language) by which we express the foregoing?

Perhaps we should start with #4: everyone list actual examples of what they see as a problem. We could then go over them to see if those are actually WP:V problems, or something else. ~ J. Johnson (JJ) (talk) 23:54, 1 July 2017 (UTC)

Too soon... wait at least two weeks Blueboar (talk) 00:20, 2 July 2017 (UTC)

You misunderstand. I am NOT talking about starting an RfC. I am proposing to start on the preliminary work necessary for starting an RfC. Lacking such preparation, the appropriate wait-time for an RfC with any reasonable chance of success is arguably infinite. ~ J. Johnson (JJ) (talk) 20:41, 2 July 2017 (UTC)

No, I do understand... but I think it is too soon even for that. We need to take a complete break from the issue... We need time to simply think about the issue and reflect. Then we can prep for the new RFC. Blueboar (talk) 22:33, 2 July 2017 (UTC)

Unless there is an issue with the current version there is no need to continue with more RfCs. QuackGuru (talk) 22:38, 2 July 2017 (UTC)

The next RfC? I don't contemplate any particular item, only the general prospect of future RfC's. And while I'm sure we will all enjoy a break, the problem is not a simple lack of reflection (etc.), but inconsistent reflection.

QG: There is definitely a need; there just isn't any point to further RfCs when we don't have agreement as to whether there are issues, when it seems we are not even using the same language. ~ J. Johnson (JJ) (talk) 23:14, 2 July 2017 (UTC)

I think that there is a more immediate question and, as step one to the process, propose a simple rfc that can't get bogged down; it shoul dbe carefully worded. That is, which version should be in there until such time that another version is agreed on? The January 2017 version (as put back in yesterday) or the February 2017 version? I could draft this. As it stands one person made this decision ( a non-admin close where it was discussed by the "other" side that even a one admin close was not enough and needed a three admin close). I think that the "no consensus" analysis was correct but there was no analysis and no drawing from the RFC on which version should go in. More specifically, which version should be treated as the "proposed change". Of course, valid arguments could be made in both directions on this question. Sincerely, North8000 (talk) 11:22, 3 July 2017 (UTC)

Struck my post. I mis-read. North8000 (talk) 16:44, 5 July 2017 (UTC)

Please... STOP... give it a rest. If you can not wait a mere two weeks (now just a week and a half), your behavior is bordering on obsession, and that is disruptive. Blueboar (talk) 12:11, 3 July 2017 (UTC)

Please don't impugn me like that. My post is about a potential RFC which for the interim / short term. And you are claiming that following the particulars of the close (the same one which put in the January version which I know that you prefer) is being obsessive / disruptive. . North8000 (talk) 14:31, 3 July 2017 (UTC)

Struck my post. I misread. North8000 (talk) 16:44, 5 July 2017 (UTC)

No issues have been pointed out. If there are no issues then there must not be any more RfCs. QuackGuru (talk) 14:50, 3 July 2017 (UTC)

Oh there are lots of issues... see the RFC above. However, the "no consensus" close of that RFC says we need to take a break before we start to discuss these issues yet again. Blueboar (talk) 15:30, 3 July 2017 (UTC)

How about a two-week ban on this topic for everyone? Leave a note that anyone broaching this gets a four-week personal topic ban. And once discussion resumes, a further ban on proposing any RfCs for at least a month? ~ J. Johnson (JJ) (talk) 20:26, 3 July 2017 (UTC)

Unnecessary. I apologize to everyone for mistakenly getting this off to a false start and I can certainly wait now that I know that: Verifiability as a reason for inclusion can be discussed again has started. (refresh)

. — TransporterMan (TALK) 16:58, 4 July 2017 (UTC)

What is the current problems with policy? The underlying problems must be fully explained before anymore RfCs. QuackGuru (talk) 19:10, 5 July 2017 (UTC)

I find this a rather ironic statement coming from an editor who claimed that citation underkill was a widespread problem yet proved extremely reluctant to provide diffs to substantiate their claim. If you want to know what points people regarded as concerns, why not review the closed RfC and then ask for clarification with any points on which you're still confused? DonIago (talk) 19:20, 5 July 2017 (UTC)

Making proposals for changes is very different than providing well reasoned arguments for changes. QuackGuru (talk) 19:25, 5 July 2017 (UTC)

Both benefit from diffs illustrating that they're a legitimate concern, and "well reasoned" can be a matter of opinion. DonIago (talk) 19:27, 5 July 2017 (UTC)

Re-start

Before we start thrashing about on some new RfCs, and most certainly before anyone starts arguing that some RfC will (for better or worse) be the end of Western Civilization, could we start identifying what problems or scenarious, either extant or prospective, will be alleviated or exacerbated by making any changes?

E.g.: QuackGuru, you have previously expressed a concern about "whitewashing" of articles. Without getting into whether such a thing should be allowed or not, can you describe a scenario where that happens? Or does not happen, depending on the interpretation of WP:V? Or even point to examples? ~ J. Johnson (JJ) (talk) 20:39, 13 July 2017 (UTC)

One thing that might help to clarify things would be if those insisting on verifiability not being a reason for inclusion to attempt to list what they believe are "reasons for inclusion" under existing policy. Then we can discuss why verifiability is not included in that list. Sławomir Biały (talk) 21:26, 13 July 2017 (UTC)

I think everybody should list the examples they might want to bring to the argument. But note: NO ARGUMENT at this point, nor discussion of why any thing is, or is not. We're just collecting snowballs now, and agreeing on whether they are regulation size; the fight comes later. :-) J. Johnson (JJ) (talk)

This seems to be placing the burden incorrectly on those who feel there is nothing wrong with the current policy. The onus is on those wishing to change established policy to provide examples of why they think policy must be changed. So, please provide your examples and, if you wish to continue to pursue your earlier proposed change to the policy, also list examples of "reasons for inclusion". Sławomir Biały (talk) 11:17, 14 July 2017 (UTC)

Did I not say "everybody"? Or are you thinking that one side should be exempt? (I would be pleased to set an example, but have been rather busy with other work.) ~ J. Johnson (JJ) (talk) 19:57, 14 July 2017 (UTC)

I take it you want examples of WP:V not being broken? Like this edit, for example, where relevant information is added to an article, supported by a source. Did this edit go through a complicated calculus of WP:NPOV before being made? Or was it verifiability plus relevance to the topic at hand that were the reasons for its implementation? There are millions more examples of the present wording of WP:V not being broken, but I too am rather busy in real life to present these here. In any case, I oppose the entire premise of this inquest, which appears to be that you and several others are claiming WP:V is broken, and are therefore asking us leading questions to support that conclusion. In any case, I am happy to supply examples of non-broken WP:V, but I'm afraid that they are all rather boring. Sławomir Biały (talk) 22:09, 15 July 2017 (UTC)

It sounds like what you are talking about is something broader than just WP:V ... it is essentially the ground work needed for the creation of a WP:Inclusion and exclusion policy. I'm not necessarily against creating such a policy... but we should probably move the discussion to a benue where we could get a broader consensus, such as the Village Pump. Blueboar (talk) 22:22, 13 July 2017 (UTC)

Eventually it might get there, but at this infant stage I oppose any broader venue on the basis of my variation on Brook's Law: Adding more people just makes it harder. Don't forget that back in Feb. 11 people could reach consensus in a single month, but afterwards some 20 people couldn't get close even after four months. Let's work on the concept of prepping our examples before trying for Big Policy, of collecting nails before building a house. ~ J. Johnson (JJ) (talk) 06:00, 14 July 2017 (UTC)

Well the proposed change to this policy seems rather hard to fathom if there is no such thing as "a reason for inclusion" to begin with. For my part, if I were listing such things, verifiability would be on that list. But I am not trying to push this unknown "reason for inclusion" verbiage into the policy, and it seems to be that the onus should be on those wishing to make the change to clarify what the think "reason for inclusion" means, and in particular to supply examples. If (as I strongly suspect) there is no such concept as "reason for inclusion", then I cannot see how saying Verifiability is not one of them is remotely clarifying. Sławomir Biały (talk) 22:35, 13 July 2017 (UTC)

Please: no argument on any issue in this section! The idea is to collect data, we'll argue with it later. ~ J. Johnson (JJ) (talk) 06:05, 14 July 2017 (UTC)

Note: I have split up the following so each candidate example can have its own discussion. ~ J. Johnson (JJ) (talk) 20:23, 14 July 2017 (UTC)

Example #01

Ok, here's an example that I used above. Someone adds the vapor point of mercury, with a source, to the article mercury (element). Someone else removes it, saying that merely being a verifiable fact relevant to the article is not a reason for including facts in articles (citing this guideline with its previous wording). We want to avoid this situation. Sławomir Biały (talk) 11:05, 14 July 2017 (UTC)

That's a good start. Thank you. ~ J. Johnson (JJ) (talk) 20:05, 14 July 2017 (UTC)

hmm... not sure if that is a realistic scenario, but accepting it for the sake of discussion... that begs the question: Why do we want to avoid this situation?

My first reaction to such a scenario would be to assume that the remover is objecting to the material for a reason other than verifiability... but is having difficulty expressing what the issue is.

When something like this occurs, we want policies that discourage wiki lawyering and edit warring... and encourage talk page discussion designed to find out what that non-verifiability issue actually is. Blueboar (talk) 11:40, 14 July 2017 (UTC)

But referring to verifiability nor being a reason for inclusion in no way clarifies the nature of the objection. A verifiable, relevant fact was added to an article. Verifiability was one of exactly two reasons for adding that contwnt, the other being relevance to the subject at hand. There can also be reasons for excluding that content, but "Verifiability is not a reason for inclusion" isn't one of them. Sławomir Biały (talk) 12:11, 14 July 2017 (UTC)

It remains that something like 99.9% of our articles should exclude (say some 5,500,000 articles) the vapor point of mercury despite the fact that it is verifiable. The logic and construction of our policies (see WP:5P) renders V an ancillary necessary but wholly insufficient in and of itself for WP:NPOV. So V is not, itself, a separate reason, at all. Moreover, given that V's sole requirement is 'I have a single RS' than it is manifest V is insufficient, as reason. Alanscottwalker (talk) 13:42, 14 July 2017 (UTC)

Yes, but the mercury (metal) should include that verifiable fact; so it being verifiable means that it should be included somewhere in the encyclopedia where it's relevant (as long as it doesn't infringe any other policy, and all that). Therefore "Verifiability is not a reason for inclusion" is false. If the wording had said "verifiability is not a reason for including the fact in any one particular article", you might have had a point. Diego (talk) 14:11, 14 July 2017 (UTC)

While I would agree that the vapor point of mercury probably belongs in that article... I would at least want to know why someone removed it. Blueboar (talk) 14:23, 14 July 2017 (UTC)

No. V determines neither whether it should be in the encyclopedia, nor where it should be -- that melting point's placement and its emphasis in the encyclopedia is governed by NPOV and NOR - V merely means 'I have an RS.' Alanscottwalker (talk) 14:34, 14 July 2017 (UTC)

As I said above, there are precisely two reasons in play for including that content: one is relevance to the article, and the other is V. There may be other, unarticulated, reasons for choosing to exclude that particular fact, but excluding it because WP:V is "not a reason" is false. It is one of two reasons that, absent other policies, create necessary and sufficient conditions for inclusion. Sławomir Biały (talk) 14:50, 14 July 2017 (UTC)

Sławomir, I disagree that verifiability is (or has ever been) a reason for including information. DUE is a reason for including information. The information must be verifiable (also encyclopedic and several other things), but I don't agree that "I have RS and it's relevant" is actually a reason for inclusion. "I have RS" is a necessary pre-condition, but it's not sufficient. Quite a lot of trivia (e.g., about a television episode) is "relevant" and "verifiable", but "This information is relevant to the exact article of <name of television episode> and I have RS for it" does not mean that you actually should add, say, the names of every schoolchild present during the filming of a Sesame Street song. WhatamIdoing (talk) 15:44, 14 July 2017 (UTC)

Indeed, per policy the pedia is not here to produce endless collections of relevant verifiable things and call them articles - not only is that contrary to NPOV, it is an assurance of unreadability. One article can have five 700 page books written on that articles' subject and a 100 other RS - all verifiable relevant things in every one of the million of sources' sentences (almost all of those sentences more then one verifiable relevant thing, including in each one, 'author wrote. . . .') - we per policy do not include them all, despite the fact that they are verifiable and relevant to the article. Alanscottwalker (talk) 16:24, 14 July 2017 (UTC)

Right. We have other policies that allow us to exclude content (especially NOT, which covers the concerns you have articulated here). That is also primarily how UNDUE works, in respose to WAID. Although, yes, it too creates a "reason" for including things.

We can have multiple reasons for including and excluding things, can't we? Sławomir Biały (talk) 18:14, 14 July 2017 (UTC)

Example #02

For many articles there is "whitewashing". There are others that are downright embarrassing. For example, see Talk:Chiropractic/Archive 39#PoV hatchet job. QuackGuru (talk) 21:53, 13 July 2017 (UTC)

A broad allegation – hand waving — of "many articles" is not going to get us anywhere. Take one as an example, and then detail the specific edits that exemplify this "white washing".

Any uninvolved editor should be able to tell what is the problem with that article without me having to spell it out. There is a talk page, the archive and edit history. QuackGuru (talk) 02:15, 14 July 2017 (UTC)

That is a pretty horribly written article.North8000 (talk) 03:19, 14 July 2017 (UTC)

If you or any other person read the archive I pointed to on the talk page it will become clear that the first paragraph is utter nonsense. The first paragraph does not properly summarise the body and content in the first paragraph fails verification. Is the problem the WMF? The WMF does not enforce policy. QuackGuru (talk) 04:06, 14 July 2017 (UTC)

And how is any of that "whitewashing"?

QG: Regarding your "Any uninvolved editor should be able to tell what is the problem with that article without me having to spell it out": Not at all.

You seem to think that what is blazingly apparent in your mind is so clearly and obviously right that if the rest of us don't "get it" right off we are either idiots, or perhaps involved idiots. (But aren't you also "involved"?)

I invite you to consider another alternative: none of us are mind readers. And none of us have quite the same background, sensibilities, analytical capabilities, or point-of-view as you, so it is quite naive of you to expect the rest of us to see things exactly as you do.

As to enlightenment being a simple matter of reading: the Chiropractic article has had some 10,600 revisions by 1,663 editors, and the talk page 18,503 revisons by 629 editors spread across 39 archives. Only on your third edit have you deigned to tell us we should "read the archive I pointed to on the talk page" – that would be archive 39, which looks to be around 10,000 words – then "it will become clear that the first paragraph is utter nonsense." Huh? I went there to find out about whitewashing, which you and RexxS bandy about as a perjorative three times, but never get around to describing or defining.

That is not an example. That is a run-arouund. You don't seem to understand that in some ways argumentation is like poker: to win a hand you have show your cards. And you don't have any.

So let's try this again. Can you describe, in less than, say, 200 words, an example, real or hypothetical, of editorial whitewashing (censorship)? ~ J. Johnson (JJ) (talk) 05:32, 14 July 2017 (UTC)

The first paragraph does not follow the body and the first paragraph contains content that fails verification. Until this is resolved I will not be going into more details. QuackGuru (talk) 16:08, 14 July 2017 (UTC)

If the first paragraph (lede?) does not "follow" the body, then it could be a problem of summarization. Not of verification, nor even inclusion. And sheds no light on whatever you mean by "whitewashing", let alone illustrate any kind of "whitewashing" problem (real, or merely possible). Until you resolve that, you have nothing in your hand. And any claim of whitewashing will be properly derided as just blowing smoke.

If you have any instances of what you consider whitewashing please point to them. They don't have to be perfect; the purpose of this exercise to shape up ("perfect") any candidate examples so that they can used in argumentation without requiring any mind-reading. If you can't point to any such instances then, for all purposes of discussion or argument, they don't exist. ~ J. Johnson (JJ) (talk) 21:14, 14 July 2017 (UTC)

The first paragraph does not follow the body and does not follow Verifiability policy.

First sentence: "Chiropractic is a form of alternative medicine mostly concerned with the diagnosis and treatment of mechanical disorders of the musculoskeletal system, especially the spine.[1][2]" Second sentence: "Proponents claim that such disorders affect general health via the nervous system.[2]" The part "Proponents claim" failed verification. Third sentence "These claims are not backed by any evidence." This sentence is unverifiable and is misplaced even if it were verifiable.

The current proposal in the draft to fix the first two sentences: "Chiropractic is a discipline that emphasizes diagnosis, treatment and prevention of mechanical disorders of the musculoskeletal system, especially the spine, under the hypothesis that such disorders affect general health via the nervous system.[1]" This proposal is merging the first two sentences into one sentence. The first paragraph in the draft does summarise the body without deleting all the relevant content such as "...but may also include exercises and health and lifestyle counseling". There is a lot more that I won't go into detail here. If an editor rejects summarising the body then is it whitewashing? QuackGuru (talk) 07:57, 15 July 2017 (UTC)

No. Incorrect (or lack) of summarization is not necessarily whitewashing; it is not a WP:V problem.

You claim something failed verification. Okay, that's a start. Did you then remove that material, citing WP:V? And how did that go? Was there a problem? ~ J. Johnson (JJ) (talk) 20:39, 15 July 2017 (UTC)

I made the proposal on the talk page to fix both issues. The proposal will fix the lack of summarizing the body and fix the content that failed V. QuackGuru (talk) 20:48, 15 July 2017 (UTC)

So you identified a possible verifiability problem, and have proposed a solution for fixing it. How is this case an example of any kind of problem with WP:V? ~ J. Johnson (JJ) (talk) 20:57, 15 July 2017 (UTC)

Proposing to fix it and actually fixing it are two different things. If I tried to remove the content that failed verification and replace it with content that passed verification it will be reverted. QuackGuru (talk) 21:06, 15 July 2017 (UTC)

So you haven't told us the whole story, like perhaps you are going against conensus. (Please remember that deliberate omission of material facts is tantamount to lying.) It is not "whitewashing" simply because you cannot get your way on something. It appears you have a disagreement with some other editors, and they, having a majority, reject your contention. Since you will not provide details of this case it is not a useable example of a possible problem with the policy.

You stated that there is whitewashing for many articles. If that is the case, then surely you can find better examples. Right? ~ J. Johnson (JJ) (talk) 22:49, 17 July 2017 (UTC)

I very strongly doubt that editors agree with Quack about his belief that "Proponents claim" has {{failed verification}}. First, the cited source is not the only source that makes this claim, including multiple sources that are cited inline elsewhere in the article, so the existence of the fact is ABLE to be verified by anyone with access to a web search engine. Second, the cited source is written by proponents of chiropractic, which means that this wording is WP:INTEXT attribution for their POV.

JJ, you might find it interesting to look at some of Quack's work. I believe that he's very proud of the work he did on Heat-not-burn tobacco product to make sure that every sentence is followed by a citation that contains not only the same facts, but also similar wording. WhatamIdoing (talk) 17:11, 5 August 2017 (UTC)

The part "Proponents claim" in the chiropractic lede fails verification. If you disagree then provide verification from the source.[10] QuackGuru (talk) 19:23, 5 August 2017 (UTC)

Interesting. Now he just needs to switch to short-cites so each in-line citation can have its own in-source specification, so a verifier has to scan only a particular page, not the entire article.

QG: As we have already discussed (above): you claim a verifiability problem at Chiropractic, but the other editors there disagree. Please note: no one — neither the editors there, nor those of us here — is obligated to disprove your allegations. (Or even explain any of this.) The only problem you have shown is one of WP:NOTGETTINGIT. You wave your hand around a lot and blow smoke, and seem to think your perceptions are blazingly obvious, but you repeatedly fail, even refuse, to support your allegations. (E.g., your comment above at 02:15, 14 July, that you shouldn't have to "spell it out", or your outright refusal to get into "more details".) Your behavior here is tendentious, verging into disruptive. ~ J. Johnson (JJ) (talk) 21:51, 5 August 2017 (UTC)

That is not verifying the content. Is there a reason no editor will verify the claim? You have not shown that other editors disagree that the content fails verification. No editor will provide verification because the content failed verification. The current wording is "Proponents claim that such disorders affect general health via the nervous system.[2]" Again, the part "Proponents claim" failed verification and if you disagree then provide verification from the source.[11] If no editor can provide verification then that means the content failed verification.

But I can provide verification for the content that passed verification. Part of the proposal is "....especially the spine, under the hypothesis that such disorders affect general health via the nervous system.[1]" Read under "Before going further it is necessary to specify exactly what is meant by the chiropractic hypothesis."[12] The source says "There is a fundamental and important relationship (mediated through the nervous system) between the spine and health."[13] There's no mention of "Proponents claim". QuackGuru (talk) 22:14, 5 August 2017 (UTC)

If an editor reads the archives they will notice editors ignored the concerns that there is original research in the lede. QuackGuru (talk) 22:50, 5 August 2017 (UTC)

You have not shown the specific text you dispute, nor any sequence of edits where such text was added and deleted (or deleted and restored), nor where you made any objection of verifiability on the Talk page. Even if you showed all of that, you only demonstrate a content dispute, but at least it would be a start.

Of course, if you really feel that something has failed verification, please feel free to boldly delete it. Let me know if you need more rope. ~ J. Johnson (JJ) (talk) 22:41, 5 August 2017 (UTC)

Please read the source[14] and focus on Wikipedia:Verifiability or move on. I provided verification for part of the proposal and showed part of the current wording failed verification. I don't "feel" that something has failed verification. I showed above there is a problem with part of the current wording in the lede. User:J. Johnson, have you read any of the sources or try to verify the content? If you have not then you are wasting my time. QuackGuru (talk) 22:50, 5 August 2017 (UTC)

I have read the source. Specifically, I read the lines identifying its eight authors, all of whom are proponents of chiropractic. They made this claim; they are proponents of chiropractic; therefore, proponents of chiropractic have indeed made this claim. (And if you think that these proponents somehow aren't making this claim, then perhaps you'd like to explain exactly who put it in that paper.)

When someone writes a source that promotes a particular idea, we don't require the body of the article to say that the author supports the idea.

If a bunch of Democrats write something about politics, especially if multiple different groups of them write essentially the same thing in unrelated publications, then we accurately and verifiably write, "Democrats said this".
If a bunch of mathematicians write something, especially if multiple different groups of them write essentially the same thing in unrelated publications, then we accurately and verifiably write, "Mathematicians said this".
If a bunch of people promoting chiropractic say something, especially if multiple different groups of them write essentially the same thing in unrelated publications, then we accurately and verifiably write, "Proponents [of chiropractic] said this".

It's called "summarizing", and it's what encyclopedias are supposed to do. Summarizing differs significantly from "copying exact phrases out of single sources". If you do not want to be involved in summarizing information from multiple sources, then you should stop editing. Wikipedia is not the right place for people who don't want to do this. WhatamIdoing (talk) 04:15, 10 August 2017 (UTC)

Summarizing requires an ability to assess what people are saying, and that definitely requires an ability to HEAR. It seems to me that we have had a good showing that his competency does not extend that far. ~ J. Johnson (JJ) (talk) 06:27, 10 August 2017 (UTC)

It called misrepresenting the source. The source does not say it was what proponents claim. Read under "Before going further it is necessary to specify exactly what is meant by the chiropractic hypothesis."[15] It was under the "hypothesis" that that such disorders affect general health via the nervous system. The part "proponents claim" is not a synonym for hypothesis. Stating "proponents claim" is also unencyclopedic. The same can be said for reviews by Ernst. He is a known chiropractic critic, but we don't say a "critic said". We WP:ASSERT the content without adding POV wording, especially for non-controversial content. No serious dispute has been presented. The source does not verify it was what "proponents claim". It verifies it was under the "hypothesis". It is also not neutral to say "proponents claim". The next sentence in the lede says "These claims are not backed by any evidence." That's more misleading content. It is also unsourced. Even if it were sourced it does not belong in the first paragraph. Later on in the lede there is plenty of content about effectiveness. QuackGuru (talk) 16:21, 10 August 2017 (UTC)

When Ernst is the only source that says something, then we really should give WP:INTEXT attribution to him. To omit that is to misrepresent to the reader that everyone holds that POV, rather than the view being shared by Ernst and whichever Wikipedia editor added the sentence. WhatamIdoing (talk) 20:57, 10 August 2017 (UTC)

Not according to WP:WikiVoice and WP:ASSERT. The chiropractic page is heavily based on his reviews. See "Numerous controlled clinical studies of treatments used by chiropractors have been conducted, with conflicting results.[4] Systematic reviews of this research have not found evidence that chiropractic manipulation is effective, with the possible exception of treatment for back pain.[4]" That does not mean we attribute it to who wrote it in the text. The citation attributed it to the author(s). Attributing statements to certain authors was discussed before. For example, this edit attributed it to the author, in order to weaken the statement. In this case, it was not accurate, as the source in question is by Keating and two other authors, not by Keating alone. The same can be said for "proponents claim". It is not accurate. QuackGuru (talk) 21:38, 10 August 2017 (UTC)

Which all sure looks like a content dispute. ~ J. Johnson (JJ) (talk) 22:39, 10 August 2017 (UTC)

Content that fails verification is claimed to be a content dispute. That's the problem. Editors know they can replace sourced content with POV language or worse with content that failed verification. Why? Because they know policy is not being enforced. QuackGuru (talk) 02:25, 11 August 2017 (UTC)

If Ernst is the only source that says something, then we have no firm reason to believe that it actually is "a fact" as defined in ASSERT. Also, Ernst expresses subjective opinions, and his opinions should not be written in Wikipedia's voice as if they were facts. For example, Ernst holds the opinion that Germans' fondness for homeopathy is embarrassing. That's his personal, subjective opinion; if we included that in an article, then those policies demand that his opinion be attributed to him, and not presented as a universally acknowledged fact. WhatamIdoing (talk) 18:23, 11 August 2017 (UTC)

For evidence based content it is asserted as a fact in Wikipedia's voice if there is no serious dispute. QuackGuru (talk) 18:28, 11 August 2017 (UTC)

Example #03

Here's an example: Every now and again, we get requests to remove or play down verifiable information from articles about life-threatening diseases on the grounds that it really sucks to learn that a kid you know is probably going to die. Imagine people figuring out how to leave a talk-page message (not an easy thing) just to tell us that we're destroying the family's hope. They're probably right about us destroying hope, and I assume that the person writing to us (always, within the limits of my memory, with a remarkably polite message) has just dealt with the emotional aftermath of a family member reading the Wikipedia article. I'm sure that having a Wikipedia article say "Your grandchild is almost certainly going to die soon" doesn't improve anyone's day. And so they ask: could we please find a way to emphasize the fact that 10% of the kids survive past their first birthdays, rather than suddenly saying that 90% of them die no matter what treatment you use?

I'm not sure that this is a good example, but it's the best that I can think of right now. We include ==Prognosis== information in articles because it's appropriate to the subject (all non-trivial sources about all diseases include at least some information about causes, symptoms, treatments, and prognosis; therefore WP:MEDMOS#Sections tells editors that they should normally have an entire section on the subject). We react to these occasional requests by waving at NOTCENSORED (and, I hope, with sympathy for the horrible situation the requester is dealing with). But "I can cite it, so I should add it" has AFAICT never been an argument for including information about the prognosis. Instead, we include information about outcomes (the best we've got, which is not always ideal) because the article would be incomplete without some information on the subject. Also, it's not really "editorial" whitewashing, so it may be not be a useful example in that sense, either. But perhaps it will inspire someone else to give a different example. WhatamIdoing (talk) 07:09, 14 July 2017 (UTC)

This sounds more like an example of a POV issue - Whether to write the prognosis positively (10% survive) or negatively (90% die) - or perhaps a relevance issue - rather than an an example of a verifiability issue (I assume the prognosis is verifiable). Blueboar (talk) 11:52, 14 July 2017 (UTC)

Yes, it is a NPOV issue, if we say 90% die it is because a survey of multiple RS leads to, that is the way to say it, and also to, it must be said right there - a survey of sources and the placement of their information is NPOV, it is not V, which is merely look to a single source. -- Alanscottwalker (talk) 14:19, 14 July 2017 (UTC)

"Whitewashing" is always an NPOV issue. WhatamIdoing (talk) 15:45, 14 July 2017 (UTC)

Agreed. Alanscottwalker (talk) 15:48, 14 July 2017 (UTC)

The question is: is there a verifiability issue? If not, then it is irrelevant to this policy. Blueboar (talk) 16:05, 14 July 2017 (UTC)

QuackGuru's comments appeared to be diverging from WhatamIdoing's example so I moved them to #Example #05. ~ J. Johnson (JJ) (talk) 20:49, 15 July 2017 (UTC)

I've been trying to come up with some possible way "whitewashing" – in the sense of glossing over or covering up, or any form of censorship — that could be a verifiability issue, but no luck. At best, the verifiability of something might be questioned in order to whitewash something, but in such cases I think the verifiability issue (the "fire") should be addressed directly. One possible result here: can we find any cases where WP:V was asserted (in any form) in what was basically a POV dispute? Is there any way to search on edit summaries? ~ J. Johnson (JJ) (talk) 23:08, 17 July 2017 (UTC)

That's almost all of them.

Serious POV wars, at least in the sciences, are fought undercover, by claiming that the other guy hasn't cited his sources and that the cited sources don't support the claims.

So, for example, if you cite a journal article that says in rural Ruritania, Example Herb is used by 99% of people when they have the common cold, then I won't tell you that I'm biased against your "promotion" of alternative medicine (because a few readers will misunderstand 'is used' as meaning 'it works, so you should buy some on the internet right now'), or even that I think mentioning the crazy practices of a small country is UNDUE for the article. Instead, I'll tell you that your source isn't good enough, because it's not a systematic review from a highly reputable, mainstream/anti-altmed medical journal – or whatever it is that I need to say about the source to give you the impression that I innocently care about verifiability and that it has nothing whatsoever to do with my anti-altmed POV.

Or, to give a different example, a little while ago, I cited a review article from a biochemistry journal to say that the main commercial market for a particular vegetable oil is in cosmetics and hair care products, such as shampoo. This was reverted with the complaint that this niche journal's WP:Impact factor is "only" equivalent to the average IF for multiple medical specialties (e.g., orthopedics) and that the editor-in-chief's commercial business does not seem to be attracting a large number of investors. I was even told that the chemistry article was full of promotional wikt:woo woo (do read that definition...) for saying that bleached and permed hair gets softer if you smear a little oil in it (the chemicals involved strip out natural oils), and that a fashion magazine that publishes daily horoscopes would be a better source.

I seriously doubt that the actual problem is the source. (Just imagine that at RSN: "Hi, is it okay if I use a peer-reviewed chemistry journal article to support the claim that this oil "is used in shampoo and other hair care products"? Someone said that a fashion magazine is a better source than an academic review article.")

But why would an editor object to the source like that? Well, maybe I'm wrong, but I believe that the problem is that the editor doesn't like the POV that this oil is primarily used as a trendy shampoo ingredient, instead of being primarily the next threatening alternative cancer treatment. But POV disputes are complicated. Finding the right answer means looking at a lot of sources, and deciding to rely upon some of them. Even if you've looked at multiple sources, reasonable people might legitimately disagree. You might have to find room in the article for POVs that you don't care about, such as cosmetology or cooking or agriculture in the developing world. It's far easier to say "just a bad source, just not verifiable" over and over again than to do all of that work. WhatamIdoing (talk) 17:52, 5 August 2017 (UTC)

"But why would an editor object to the source like that?" Is it the same reason an editor disagrees with replacing content that failed verification, with content that passed verification at the chiropractic page? These are not POV disputes and they are not complicated. It is very simple. Are there editors they intentionally edit against policy? QuackGuru (talk) 19:30, 5 August 2017 (UTC)

Of course there are editors who intentionally edit against policy. See, e.g, all the vandals.

It should be equally obvious that insisting that "[musculoskeletal] disorders affect general health via the nervous system" be hedged with "Proponents claim [but all right-thinking scientifically minded people know that anything said by its proponents is nonsense]" is a question of POV pushing instead of a simple question of whether the statement could be verified in any reliable source ever published, in any language, anywhere in the world. I suspect that even you already believe that some reliable source somewhere has attributed this belief to proponents of chiropractic (and not to its detractors). WhatamIdoing (talk) 20:29, 5 August 2017 (UTC)

That's not what the source says. Read the source. See Wikipedia:Verifiability. QuackGuru (talk) 20:34, 5 August 2017 (UTC)

Even if that is not precicy what the specific source that was cited says, do you think it likely that some other reliable source says it? Blueboar (talk) 02:12, 11 August 2017 (UTC)

This is not about what other sources say. WP:FAILEDVERIFICATION says "Material that fails verification may be tagged with {{failed verification}} or removed." That does not work. We can expand on that. After the failed verification content was replaced with content that passes verification the failed verification was restored. Any suggestions? QuackGuru (talk) 02:30, 11 August 2017 (UTC)

It *is* "about what other sources say", because if other (reliable) sources say this, then the material is ABLE to be VERIFIED, and therefore it doesn't truly "fail verification". "Not stated explicitly in the exact already-cited source" is not the same as "cannot be verified in ANY reliable source". WhatamIdoing (talk) 17:42, 11 August 2017 (UTC)

When the content failed verification it still failed verification. If other sources can verify the claim is a different discussion. This is about when editors replace source content with content that failed verification. I know chiropractic is on your watchlist. QuackGuru (talk) 18:12, 11 August 2017 (UTC)

Chiropractic isn't on my watchlist. I try to respond to requests posted at WT:MED about it (as I attempt to do for most requests posted there), but I happen to have little interest in the subject, and even less interest in the endless disputes caused by the POV pushing there. WhatamIdoing (talk) 18:28, 11 August 2017 (UTC)

You mentioned "and even less interest in the endless disputes caused by the POV pushing there." Different versions of the content can't be both verifiable at chiropractic when the wording is very different. QuackGuru (talk) 18:45, 11 August 2017 (UTC)

Example #04

A birth year of an actor was proposed to be part of the article title by using user-generated websites, like IMDB and FindAGrave.com. Reliable sources should verify that birth year, but the title of an article is renamed anyway regardless with a "caveat". See Talk:Timothy Scott (actor, born 1937)#Requested move 2 July 2017. --George Ho (talk) 06:35, 15 July 2017 (UTC)

Thank you, George. In that case was there any question involving verification, or lack of verification, where any editor invoked WP:V, and that revealed any kind of problem with WP:V? ~ J. Johnson (JJ) (talk) 19:52, 15 July 2017 (UTC)

In that case, I was the only one who had concerns with verifying the exact birth year. When the birth years were proposed, the unreliable sources were used. However, the consensus said yes to birth years because it's more preferable to using death years. Verification wasn't raised there when the birth years were prosed and when the consensus approved the change. Therefore, the birth year issue is treated like a minor issue, and the change makes the verification issue... moot. --George Ho (talk) 22:17, 15 July 2017 (UTC)

Example #05

[Split off from #Example #03. -JJ]

Content that obviously fails verification often remains in an article for months or years. This is the result of "consensus". It is like Consensus and Verifiability contradict each other. QuackGuru (talk) 16:08, 14 July 2017 (UTC)

Can you point to some instances of content that "obviously fails verification" and has remained in an article for months or years? (Perhaps in a separate sub-section.) ~ J. Johnson (JJ) (talk) 19:56, 15 July 2017 (UTC)

I could point to different examples but I don't think those editors would like that. The articles are still on my watchlist. Maybe if they retire (or are topic banned) I can show you and can fix it. Consensus does not mean the content passed verification. QuackGuru (talk) 20:30, 15 July 2017 (UTC)

Why should those editors not like you pointing to any problems? It sounds more like there was consensus, but you don't agree with it. That is more in the nature of what we call "a personal problem".

I think you should review our discussion under #Example #02, as it seems you still want to claim you hold a certain hand, but won't show us your cards. Any cases of unverified material being retained despite being challenged certainly seem like a prima facie WP:V problem, which bears directly on the overall issue here. But if you can't point to those cases then, for any purpose of argumentation, the don't exist. Can you find some other cases? ~ J. Johnson (JJ) (talk) 21:36, 15 July 2017 (UTC)

If they exist, it certainly can't be a fault of this policy....it is very very strong on getting rid of challenged unverifiable material. North8000 (talk) 00:36, 16 July 2017 (UTC)

Such cases could result from this policy being incomplete, vague, or ambiguous. Or perhaps weak, so that editors don't believe they have a case strong enough for challenging. Does verifiability include disputes on the authority or reliability of sources? ~ J. Johnson (JJ) (talk) 23:36, 17 July 2017 (UTC)

Verifiability encompasses decisions about whether sources are authoritative/reliable for the statements in question (i.e., all sources, whether currently cited or not). However, to keep cited material out of an article on grounds of (only) verifiability, you have to get other editors to agree that your concerns about the sources are valid, and that seems to be the sticking point in many of Quack's examples. WhatamIdoing (talk) 19:34, 5 August 2017 (UTC)

One of the problems is that admins do not enforce the verifiability policy. If they did many editors (and admins) would be indeffed. QuackGuru (talk) 19:46, 5 August 2017 (UTC)

Enforcement of this policy means editing articles to comply with it. It does not mean blocking editors. WhatamIdoing (talk) 20:24, 5 August 2017 (UTC)

It is a problem when "It does not mean blocking editors" for violating verifiability policy. As long as editors know there are no consequences for violating verifiability policy they will continue. QuackGuru (talk) 20:32, 5 August 2017 (UTC)

You have NOT provided any examples of "[c]ontent that obviously fails verification often remains in an article for months or years." Nor any examples "that admins do not enforce the verifiability policy." You're just waving your hands around, with nothing to show other than, at best, cases where other editors do not agree with you about whether some point is verifiable or not. You have not shown any problems with the verfiability policy. Other than, that is, it does not say "always defer to QuackGuru in content disputes." ~ J. Johnson (JJ) (talk) 22:18, 5 August 2017 (UTC)

I provided examples from the chiropractic article and you are not focusing on trying to verify the content or improve the article. If you are not interested in improving the article then you don't have to comment. That comment is not helpful. QuackGuru (talk) 22:25, 5 August 2017 (UTC)

You have failed to recognize that this Talk page is not the place for improving the article, nor for for verifying any particular content. The discussion here is whether there are any points where WP:V needs improvement. You have shown nothing, other than a failure to HEAR. ~ J. Johnson (JJ) (talk) 21:50, 6 August 2017 (UTC)

I have shown Verifiability policy is not being enforced. That makes this policy not helpful until admins enforce policy and are blocked if they don't enforce policy. QuackGuru (talk) 00:31, 8 August 2017 (UTC)

Not quite... at best, what you have shown is that the policy isn't being enforced in a way that you would like it to be enforced. And that is an important distinction. Your views on how it ought to be enforced frequently are at odds with the views of the rest of the community. Blueboar (talk) 00:54, 8 August 2017 (UTC)

Please show not assert where admins enforce Verifiability policy? It is not being enforced. Consensus is at odds with Verifiability policy. This results in original research and content that failed verification. The community is committed to amateurism. Editors know Verifiability policy is not being enforced and they can add their opinion to articles and replace sourced content, with content that failed verification. Content that fails verification is being treated as a joke. The joke is that "consensus" is emboldening editors to continue to support content that failed verification because admins assert that original research is a "content dispute". What is worse is when an admin replaces sourced content with content that fails verification. An article littered with hoax content I have been watching is getting much worse. There must be some way to go up against the admins. I don't like it when an admin adds hoax content. I would like new policy to block admins who add hoax content rather than it being treated as a "content dispute". Treating content that fails verification merely as a "content dispute" is a joke. QuackGuru (talk) 01:35, 8 August 2017 (UTC)

Re: "Please show not assert where admins enforce Verifiability policy"... OK, I did a random search of recent contributions from Admins who's Usernames begin with the letter "A" ... I find: this, this, this, this, this, this, this, and this... all are edits where an Admin enforced WP:V in some way. Do you want me to do this for admins who's usernames start with other letters, or is this enough? Blueboar (talk) 17:22, 8 August 2017 (UTC)

I checked all of those examples. None of them are relevant to the type of content I am discussing. Those are unsourced content without a single citation. If you read the examples I have provided you will know there is at least one citation after each sentence. Admins do not enforce Verifiability policy where editors claim the content is sourced. There is "fake news" and then there is fake Wikipedia content masquerading as sourced content. There is still content in an article that clearly fails verification and the draft fixes those problems. Why editors avoided commenting on the draft? QuackGuru (talk) 23:00, 8 August 2017 (UTC)

Like I said... admins areenforcing WP:V... they are simply not enforcing it in the way you want them to. Blueboar (talk) 23:17, 8 August 2017 (UTC)

"Why editors avoided commenting on the draft?" Because we are not mind-readers, that we can figure out what you mean when you can't be troubled to explain what you are moaning about, because we are tired of chasing your will-o'-the-wisp through a rhetorical morass, because you're an ... oh, I forgot, you don't WP:HEAR very well, so commenting is pointless. ~ J. Johnson (JJ) (talk) 06:41, 10 August 2017 (UTC)

Blueboar, if admins are enforcing WP:V then show how they will enforce it at chiropractic. QuackGuru (talk) 18:28, 10 August 2017 (UTC)

J. Johnson, why editors avoided commenting on the draft? Because they are not mind-readers or is it because they know the content summarises body and is neutral as well as sourced. What's wrong with that? Not everyone likes neutral content for a controversial topic. The proposal is not negative and against chiropractic. If you check the edit history of alt-med articles you will notice a lot of edit warring. Neutrally written content on controversial topics is better than poorly written, negative content. QuackGuru (talk) 18:28, 10 August 2017 (UTC)

Quack... could you please come up with a different example than chiropractic. Your laser focus on that article is getting very old. Even if your assertion that Verifiability was not followed at that article was 100% correct, it could be waved off as a case consensus to apply WP:Ignore all rules at one single article. You really need to establish that the issue you are concerned about is a wide spread problem, and not just something that happened at one specific article. We never amend policy due to something that happened at one single article. Blueboar (talk) 19:46, 10 August 2017 (UTC)

No editor can apply WP:Ignore all rules for a case when the content fails verification. An article does not improve by keeping content that fails verification. It is a wide spread problem for content that fails verification. This is different than the examples you presented for unsourced content. Content that fails verification is common. Even changing single words can drastically change the meaning from sourced content to content that fails verification. If an editor does not like the wording they rewrite the content. Sometimes it is to improve the wording but there are times that is not the case. There are times when editors have rewritten content without even reading the source. If the new wording is not accurate it does not follow the source. What is the best way to amend Verifiability policy to try to slow down editors who rewrite content from source content to content that fails verifiability? Any suggestions? QuackGuru (talk) 20:11, 10 August 2017 (UTC)

Horse=Dead. Continued Use of Stick=Pointless. Unless you can show that this is in fact a widespread problem... something that is occurring beyond (potentially) one article, then I would say we don't need to amend the policy. Unless you can come up with other examples, we can only conclude that the policy is working perfectly well everywhere else... and that any issues at that may or may not be occurring at that one article are unique to that one article. Blueboar (talk) 21:48, 10 August 2017 (UTC)

One possible result here is a determination that QuackGuru's example of "Content that obviously fails verification often remains in an article for months or years" is, at best, applicable to a single article. But not well demonstrated. ~ J. Johnson (JJ) (talk) 22:47, 10 August 2017 (UTC)

You have never seen an editor replace sourced content with content that fails verification? This is a common problem. If they can't delete the content they rewrite it. The result can be content that does not match the source presented. QuackGuru (talk) 02:22, 11 August 2017 (UTC)

Nope, in eleven years of editing Wikipedia, I have never seen an editor replace sourced content with content that fails verification.

For the sake of discussion, I am willing to accept that it might have occurred a few times in all those years... but I do not accept that it is a common problem. I might change my opinion if presented with several examples. Blueboar (talk) 10:22, 11 August 2017 (UTC)

Agreed that examples are needed. A common issue since QuackGuru began interacting with this Talk page is that they make broad allegations that they fail to support with examples, even when directly asked to do so. DonIago (talk) 14:14, 11 August 2017 (UTC)

Content that failed verification is very common. QuackGuru (talk) 18:15, 11 August 2017 (UTC)

Quack, how about doing some work to build your case rather than being so lazy about it that you only provide a single link? Pick, say, three of those and write up your case for how they prove that this is a pernicious issue that requires work on the Verifiability policy? Why should any of us care to get involved in this when you, thus far, don't seem to care enough to build a detailed argument? If you tried doing this at ANI you'd be laughed off the page. DonIago (talk) 19:47, 11 August 2017 (UTC)

When I will do work to build my case it will be for topic bans and blocks. For now see Wikipedia_talk:Verifiability#Problem_with_wording. QuackGuru (talk) 20:48, 11 August 2017 (UTC)

Until you build your case your comments are premature, entirely unfounded, and tendentious. As to topic bans and blocks: who are you going to get topic banned? Yourself? ~ J. Johnson (JJ) (talk) 22:59, 11 August 2017 (UTC)

OK... I went through the edit histories for the first 10 articles listed at the template list that Quack points us to (at which point I stopped)... I looked into when the information was the tagged as being problematic, and tracked down when the information was first added. I found that they all fell int two patterns... either 1) the information was initially added without a citation to any source, challenged, the source added, and then the source was tagged for not actually verifying the information.... or 2) the information was added with the source, and then subsequently the source was tagged... Not one was an instance where sourced information had been replaced with unverified information. Blueboar (talk) 12:32, 12 August 2017 (UTC)

On controversial topics editors add inaccurate content and replace sourced content with unverifiable content, but if you are against including this simple clarification then there is absolutely no reason for me to provide examples for you to review. QuackGuru (talk) 16:51, 12 August 2017 (UTC)

Ok... We can't force you to provide examples if you refuse to give them... but don't expect us to agree that your concerns are valid when you refuse to demonstrate that they are. Further commentary on this will simply be disruptive... so I will end my participation now. Blueboar (talk) 17:31, 12 August 2017 (UTC)

Please stop

You're doing the exact same thing you did in the RfC. Please stop. This discussion and the RfC are deliberately structured in a way that prevents consensus from forming. The RfC by positing its suggestion as "right/wrong? discuss" which is specifically given as the canonically bad way to make an RfC, and now this discussion is using the... inadvisable... technique of splitting the discussion five or six ways right at its inception. If you really wanted to resolve this issue you'd follow the suggested RfC wording. So, please stop. Bright☀ 09:29, 11 August 2017 (UTC)

A point of contention

A point of contention on Wikipedia articles: Some articles are, and have been used to RE=WRITE HISTORY! My case in point: some of your facts about Cray Research are simply NOT TRUE! I've added addenda and have had my work simply erased. What better source than one who worked on the objects discussed with their own two eyes? I proved my actual employment at Cray Research by asking anyone to look up my name at www.excray.com. They have far stricter requirements than Wikipedia when it comes to members! I'm #119 there, Arthur Blackwell is my name. If anything, since I worked for Dr.Chen as a Scientist first ( I discovered the amount of time a signal went through a 5-4 Gate Array, I've done many DIRECT things that involve the transitions from designing the ORIGINAL X-MP 5-4 Gate arrays into the SECOND iteration that used 16 gate gate arrays. ECL logic was replaced by NMOS gate arrays with I/O drivers to slam everything back to ECL levels for data transmission ( Seymour Cray still used the hotter ECL logic, that was why the Cray-2 used Flourinert as a coolant. Yes, we did a trade show using florinert in a fish tank with both a live fish and a tv set running in it at the same time!

I know ALL the details about the 16 gate gate array the SECOND iteration the X-MP used. I jumped ship out of Silicon Valley when Cray Research needed someone with talents that colleges like SJSU didn't ever have coursework for ( know what the colleges specialized in then, the truth is NOT nice'! ). What happens when a person who was actually there collides with a " press handout " or a PR piece? As a Scientist, I dealt only with FACTS and Cray Research paid me to ALWAYS GIVE OUR DEVELOPMENT TEAM THE FACTS. Yet I still see that unverifiable facts message instead of my edits THAT ARE TRUE BECAUSE i SAW THEM WITH MY OWN EYES! ( this really gets strange when I became our Network Security Engineer because I showed how easy it waqs to steal log-ins and passwords from our branch of DARPAnet! ) That is why I can no longer use anything Wikipedia says without checking a second source; You are no better than Snopes when it comes to finding actual facts. If I say it, I mean it. Quite literally, in my case. I would rather see all the entries about Cray Research PULLED from Wikipedia. Most of your entries look like PR pieces anyway. Why spread dis-information?

And that is the heart of the matter. If you are wrong on Cray Research issues, are you wrong on all the other entries on Wikipedia? At that point, "I" have to make a judgement to use Wikipedia as a source or use my own skills to dig out the truth. Aeb1barfo (talk) 04:22, 14 August 2017 (UTC)

Have you reviewed Wikipedia's definition of what constitutes a reliable source? The reality is that this is the internet, we have no way of knowing that anyone posting here really is who they claim to be, and consequently we need information to have been published at some point. Otherwise I could claim to have developed critical pieces of the Windows 10 OS, for instance, and you'd have no way to refute my claims. It isn't realistic to expect any serious online encyclopedia to allow the publication of information that nobody can verify independently. DonIago (talk) 04:48, 14 August 2017 (UTC)

You should also read our WP:No original research policy. It covers why we can not accept information that has never been published before. I understand your desire to correct the record, and to present the (previously unpublished) truth... but Wikipedia is the wrong venue to do this. What you need to do is write a book about your experiences at Cray, and get it published... then other editors can add the information, citing your book. Blueboar (talk) 10:55, 14 August 2017 (UTC)
@Aeb1barfo:, I know it may be frustrating to follow Wikipedia policies, but they've been distilled to protect the encyclopedia from large swaths of misinformation.

As a compilation work, Wikipedia is only as good as the references it uses as the origin for its knowledge. If you want to correct some fact we got wrong, the only way to do that is adding it to the world knowledge, by publishing it in other reliable venue; this way we can find it and incorporate it with some guarantees. There's likely no need for it to appear in a full book as Blueboar suggested; we also include content from scientific articles, news reports or interviews published in mass media with a reputation for fact-checking. Diego (talk) 13:02, 14 August 2017 (UTC)

There are several things you could do and several things you cannot do. As others have said you cannot add your unpublished recollections. You can examine the sources cited in the Cray article and produce better reliable sources, or if the cited sources are not being followed correctly in the Wiki text, you can correct the article text to conform with the cited source, or if the Wiki-text is not cited, you can remove it, noting your concerns that it is uncited and mistaken. If you do the things you can do, you may have to work out any disputes on the talk page of the article. -- Alanscottwalker (talk) 13:18, 14 August 2017 (UTC)