Wikipedia talk:Verifiability

  1. Home
  2. Wiki

    Scientific fraud

    [edit]

    Not sure if this is the best place for this discussion but I suppose my question is whether we should somehow include this information in some way in this policy. As some of you may have seen in news reports (eg. [1]:""If we're not prepared to deal with the fraud that's already occurring, then we're certainly not prepared to deal with what generative AI can do to scientific literature," Dr Richardson said. "We have no clue what's going to end up in the literature, what's going to be regarded as scientific fact and what's going to be used to train future AI models, which then will be used to write more papers.”) there is a "alarming" new study on the vast amounts of money spent and faster rising creation of fraudulent science, outstripping good science, including paying for getting cited, and a whole host of things etc. "The entities enabling scientific fraud at scale are large, resilient, and growing rapidly", PNAS https://www.pnas.org/doi/10.1073/pnas.2420092122 . But some of you know much more about this kind of thing than I do. -- Alanscottwalker (talk) 13:05, 5 August 2025 (UTC)[reply]

    I think we do quite well at handling predatory journals, and RetractionBot flags articles with expressions of concern, errata, or that have been retracted. I'm sure some slip through, the same way any other unreliable source would. Might be worthwhile to have an essay about it. -- LCU ActivelyDisinterested «@» °∆t° 17:25, 5 August 2025 (UTC)[reply]
    @Raladic, do I remember you doing a bunch of retraction work last year? And @Headbomb is always a good person for conversations about identifying problematic sources. WhatamIdoing (talk) 19:55, 5 August 2025 (UTC)[reply]
    I should start by saying "unfortunately" a few times. Scientific fraud and exaggeration at the individual level has been going on for a while. Once upon a time, there was a highly respected PhD fellow (let us call him Dr L...) who faked the data about how his program generated VLSI designs. He got published in 100% WP:RS type journals. A few years later no one could replicate the results, and the industry joke was: "what is the international standard of exaggeration?" Answer: a micro-L...
    Now, fraud and misinformation have gone beyond the individual level and become institutionalized. Wikipedia will be swept into the waves of misinformation like all else. Unfortunately I have no remedy to suggest. Unfortunately. Yesterday, all my dreams... (talk) 17:45, 13 September 2025 (UTC)[reply]
    Ouch! I just looked up the Wiki page for Dr L... and realized that he died not long ago. I am sorry. But I will still not mention his name. I will just note that his Wiki page is glorious and says nothing about the reason he left academia once the exaggeration was discovered. The published papers were never retracted. We can expect more of that at a larger scale level. Yesterday, all my dreams... (talk) 18:13, 13 September 2025 (UTC)[reply]
    We do actually have an article elsewhere on-wiki that discusses the unit mentioned by Yesterday, but without any hint of fraudulent behavior from L. Perhaps even that should be removed. But without reliable sources for fraud we cannot mention it in our articles. —David Eppstein (talk) 18:32, 13 September 2025 (UTC)[reply]
    The link to the exaggeration article? By the way, I clicked and the wiki page for one of his programs has a criticism section. But I will not edit those pages given that I had met him, and feel sorry. Too many people in that industry get sick too young... So do look after yourself. Yesterday, all my dreams... (talk) 19:29, 13 September 2025 (UTC)[reply]
    Search for the unit name and you will find the articles I had in mind. —David Eppstein (talk) 20:18, 13 September 2025 (UTC)[reply]

    RfC drafting at RSP for platform categorisation

    [edit]

    Information icon There is currently a discussion at Wikipedia talk:Reliable sources/Perennial sources regarding re-categorising self-published platforms, along with amending the section in this policy Reliable sources to include such platforms. Please note this is not an RfC, this a drafting process for an RfC. The thread is Drafting the RFC question about platforms. CNC (talk) 10:33, 21 August 2025 (UTC)[reply]

    @S Marshall, you might be interested in that discussion. I keep meaning to get back to it, but it may be a while. (I don't mean to distract you from your current project of fixing up this policy.) WhatamIdoing (talk) 19:40, 13 September 2025 (UTC)[reply]

    Expanding efns in Wikipedia:Verifiability § Responsibility for providing citations

    [edit]

    Hi, I noticed there are many footnotes in the Wikipedia:Verifiability § Responsibility for providing citations section. I'd like to propose removing them and moving the text they contain inside the same section because:

    • They are useful and are relevant to the section, especially efns c, e, and d. B could be explained better, but maybe parts of the footnote should remain hidden.
    • The section is pretty small, so they fit easily.

    The text would remain the same, so this this isn't a content issue. FaviFake (talk) 10:50, 22 August 2025 (UTC)[reply]

    Concur. The information in the endnotes should be in the body. Placing that much substantive content in endnotes was a terrible idea to begin with. --Coolcaesar (talk) 16:07, 22 August 2025 (UTC)[reply]
    Luckily it's not as bad as Exempli gratia, try expanding its footnote ;) FaviFake (talk) 16:13, 22 August 2025 (UTC)[reply]
    Can you work it out in a sandbox and show how you think the end result would look? A lot of the hidden text as written, to me, is quite wordy. While I agree they contain relevant information, having them off to the side (as they are presently) helps maintain focus. We just want to make sure we aren't degrading that in the process. -- GoneIn60 (talk) 16:43, 22 August 2025 (UTC)[reply]
    Yeah, footnotes are frequently unnecessarily wordy. Here's how it might look. I cut down on about 500 bytes and kept everything the footnotes said. FaviFake (talk) 17:25, 22 August 2025 (UTC)[reply]
    The tradeoff is:
    • The shorter the main text (additional material kept in footnotes), the more likely people are to read the main text, but the less likely they are to see the footnote material.
    • The longer the main text (additional material incorporated directly), the less likely people are to read all of the main text, but the more likely they are to see the previously footnoted material.
    I don't personally have a strong opinion about which answer is the right one for this section, but I think that editors should consider whether all of this material is of equal importance. WhatamIdoing (talk) 17:38, 22 August 2025 (UTC)[reply]
    Yes, that's true. I'll just add that this is a policy page; people don't usually read it in their leisure time, but only when they need practcal info on what to do. I think the policy reccomendations in the efns shouldn't be hidden to editors who aren't curious enough to hover over them. FaviFake (talk) 09:30, 23 August 2025 (UTC)[reply]

    One thing to remember is that this is a policy page (and a core one at that). So everything that is in the text of the policy is policy. Moving that large amount of sidebar content would be basically adding about 20 things to policy. IMO any such additions should be decided on individually.North8000 (talk) 14:34, 23 August 2025 (UTC)[reply]

    If "everything that is in the text of the policy is policy", then this isn't adding anything new. These footnotes are already supported by the existing consensus, and this would just allow people to read the entire policy without having to click many blue letters.
    Besides, the footnotes don't mandate anything, they're all mere suggestions. FaviFake (talk) 14:42, 23 August 2025 (UTC)[reply]
    I disagree with several aspects of that. IMO there is a higher level of scrutiny and consensus for what's in the text of the policy than the footnotes. (and the format is more careful and concise). Next, they don't currently mandate anything because they are footnotes. Moving them would convert them into mandates. This might be a good idea for many of them, but IMO they should be individually evaluated in that context. Sincerely, North8000 (talk) 15:12, 23 August 2025
    North8000 Thanks. Do you recommend I create a subsection for each? FaviFake (talk) 16:12, 23 August 2025 (UTC)[reply]
    Some of the footnotes mandate things. See, e.g.,:
    • "Once an editor has provided any source they believe, in good faith, to be sufficient, then any editor who later removes the material must articulate specific problems that would justify its exclusion from Wikipedia (e.g., why the source is unreliable; the source does not support the claim; undue emphasis; unencyclopedic content; etc.)."
    Note that word must: it's a mandate. There are two main points that we're trying to communicate here:
    1. BURDEN requires "me" to provide one (1) source that "I" think it's good. "You" don't get to send me on an endless game of WP:FETCH.
    2. Even if "I' provide an undeniably gold-plated reliable source, that still doesn't mean that "my" content needs to stay in this article.
    WhatamIdoing (talk) 18:10, 23 August 2025 (UTC)[reply]
    You're right, I must've missed that! Thanks. FaviFake (talk) 19:20, 23 August 2025 (UTC)[reply]

    Opposed setting this precedent....efns (notes) are mostly used in the same manner as Template:Supplement#History....that is to handle overly detailed information. Many supplemental essays are the result of notes getting too big. They are used for quick explanation of rationale as seen here to explain reasoning or used to give examples that would just crowd protocol text as seen here or as a link to discussions on how we came to this conclusion as seen here.Moxy🍁 15:34, 23 August 2025 (UTC)[reply]

    Thanks. Could you elaborate on which efns in this section only handle overly detailed info? To me, all but "d" (and "b", which doesn't actually originate in that section) seem very relevant. FaviFake (talk) 16:11, 23 August 2025 (UTC)[reply]
    In my view all the notes are summaries of other protocols there to help editors understand context without have to go to another page. The point being - keep focused on the points here without having to read many other pages for understanding. Wikipedia:Content forking/Internal#Policy forksMoxy🍁 16:27, 23 August 2025 (UTC)[reply]

    This section is trying to do two things. It's policy (this is how things shall be) and practice (this is how to get there).

    The text flipflops from policy to practice and back again. The footnotes seem to me like a (somewhat inconsistent) attempt to move the practice guidance to footnotes and leave the policy in the header, but over the course of many years and many edits it's all gone wrong. Needs a full rewrite imo.—S Marshall T/C 19:28, 1 September 2025 (UTC)[reply]

    I wonder if some of this (e.g., WP:Directly supports) could be addressed with a ==Terms and definitions== section. WhatamIdoing (talk) 22:21, 1 September 2025 (UTC)[reply]
    Possibly? Do any terms other than WP:Directly supports need defining?
    I'm not settled on any thoughts at the moment but last night I came up with a very rough draft:
    I've only skimmed it but it looks great!! In case you missed it, i had made a very quick draft too in this comment, if it's of an use. I basically just copypasted and removed some parts, im sure yours is better. FaviFake (talk) 17:52, 2 September 2025 (UTC)[reply]
    A definition of verifiable (as contrasted with cited) would also help a lot of editors.
    Taking the first paragraph from your draft as an example, consider:
    • Each fact or claim must be verifiable. – What's "verifiable" mean?
    • The burden to demonstrate verifiability lies with the editor who adds or restores material. – So I have to do it, but I don't know what "it" is.
    • It is satisfied by providing an inline citation to a reliable source that directly supports the material. – What's "directly" vs (I guess) "indirectly supporting"?
    • The cited source must clearly support the fact or claim as presented in the article. – How is "clearly" different from "directly"?
    Experienced editors like us, although in some ways the only people on the planet who are qualified to write this policy, are always bringing a h-u-g-e amount of background knowledge to these things. We know things that newbies can't even guess at. WhatamIdoing (talk) 00:04, 4 September 2025 (UTC)[reply]

    Well, some of that I'd already addressed in my previous draft, but let's do the rest of it.  :)

    Yes, I know, the WP:WOOKIEE shortcut is very inside baseball.—S Marshall T/C 16:28, 5 September 2025 (UTC)[reply]

    I like the clarity of this approach. I think in a few places the proposed wording goes too far in advising against removal, as compared to the current wording.
    A few specific points:
    • When dealing with facts or claims you think are dubious, a good option is to add a tag... Tagging is helpful when it is likely that a source exists and just needs to be located and added. Dubious unsourced information should be removed. When to remove vs tag is better dealt with under Considerations.
    • Do not overuse this clause. No one is likely to view their own interpretation as overuse.
    • Although policy allows this, removing facts or claims is a relatively drastic action that's best reserved for material you have good reason to doubt. This undermines the core policy in a confusing way. Removing unsourced claims is not drastic; it is routine. It should be done with judgment, but ...material you have good reason to doubt is not the right test. Plausible OR should still be removed.
    • Removing false claims is helpful, but replacing them with verifiable claims is much better. The contrast shouldn't be between false and verifiable.
    Here's a version with some suggested edits.

    --Trystan (talk) 14:04, 6 September 2025 (UTC)[reply]
    "They should not be restored without an inline citation to a reliable source." and sentences like it are problematic (although longstanding, I admit) because they seem to limit the reasons why a sentence or bit of info should be in an article (see NPOV, NOR, BLP, CVIO). We say in other areas like ONUS, that this is not true, but I don't think that is clear enough. I get why this happened, as this is the policy on verifiability, but we should still be more circumspect in such sentences. Alanscottwalker (talk) 13:14, 7 September 2025 (UTC)[reply]
    Further comments and objections are invited? I'm considering editing the policy along these lines now.—S Marshall T/C 07:22, 11 September 2025 (UTC)[reply]
    Similar to Alanscottwalker, I think the text is potentially unhelpful in its coverage of WP:BLUE. Facts can be uncontroversial but not directly relevant. Sourcing can be requested to help demonstrate the relevancy of uncontroversial statements and avoid SYNTH, rather than just because a fact is potentially 'controversial'. I also am unsure if we should keep the note on unreferenced categories, they aren't really cited inline in the same way, so that could read as confusing. CMD (talk) 07:35, 11 September 2025 (UTC)[reply]
    The first consideration you raise has already got its own section at WP:VNOT, so I wouldn't cover that in WP:BURDEN. I agree that the verifiability of categories is confusing.—S Marshall T/C 11:29, 11 September 2025 (UTC)[reply]
    Right, I am suggesting not covering it, which the proposed text above does. CMD (talk) 15:00, 11 September 2025 (UTC)[reply]
    Does it? Where?—S Marshall T/C 15:50, 11 September 2025 (UTC)[reply]
    The new paragraph mentioning the Thames and Paris. CMD (talk) 01:11, 12 September 2025 (UTC)[reply]
    The paragraph says that (wildly oversimplified) if it's BLUE and it's CHALLENGED, then cite it instead of arguing over whether to cite it. I think we all agree to this.
    Maybe uncontroversial is not the right word? "Not one of the four types of content that is required to have an inline citation" is wordier but perhaps more accurate. WhatamIdoing (talk) 01:24, 12 September 2025 (UTC)[reply]
    The wording of blue is "all material that is either challenged...must be cited", rather than suggest arguing with it is an option. Suggesting material "should never need citations", especially when there are other reasons for citations (like those mentioned by Alanscottwalker) is unnecessary, and if the desire is to not cover that in WP:BURDEN the easiest thing to do is not cover it. If there is a desire to cover it, use more direct wording such as "If material has been challenged, a citation should be provided." CMD (talk) 01:31, 12 September 2025 (UTC)[reply]
    If I write, "The Thames is a river in England" without bothering to provide a source, and I'm writing it in River Thames, then someone who uses WP:BURDEN on that is a bloody fool who needs to find a better hobby because they aren't competent to edit an encyclopaedia.
    If I write, "The Thames is a river in England" without bothering to provide a source, and I'm writing it in Mississippi, then someone who uses WP:BURDEN on that is still a bloody fool. Yes, they can legitimately say "That's true, but maybe it doesn't belong in this article," and they're likely right. But that's not about verifiability and we don't need to bend the verifiability policy out of shape to address it. An easily-sourceable statement that's in the wrong place is still a verifiable statement.—S Marshall T/C 14:19, 12 September 2025 (UTC)[reply]
    We do have a certain number of fools...[2] WhatamIdoing (talk) 22:35, 12 September 2025 (UTC)[reply]
    I think that particular instance is more vandalism than foolishness, but yup, we certainly have. It did make me laugh.—S Marshall T/C 01:15, 13 September 2025 (UTC)[reply]
    Why assume they are using BURDEN? That's not a scenario I've seen, but I have seen people respond "it doesn't need to be cited" or similar when the "likely right" issues are raised. CMD (talk) 15:59, 14 September 2025 (UTC)[reply]

    It looks good to me. I do like the "avoid overuse" aspect. The most common overuse is nit-picking the source when sky-is-blue "I don't like it" material is restored. I'd suggest a very thorough (re) review by a thorough expert like S Marshall  :-) be made before putting it in. North8000 (talk) 14:17, 11 September 2025 (UTC)[reply]

    As @Trystan says, I'm never going to view my use of this clause as "overuse". However, I think it's important for the rest of the community to have something like this, so you can all tell me "No, really: stop being WP:POINTY or WP:DISRUPTIVE or whatever already. It says so right there in the policy that restraint and judgment are needed".
    It's unfortunate that NOTBURO ("Do not follow an overly strict interpretation of the letter of policies without considering their principles") isn't more widely respected these days, but "Don't overuse this" is WP:V's version of NOTBURO. WhatamIdoing (talk) 01:20, 12 September 2025 (UTC)[reply]
    Well, it will be, if I can get consensus to add it. I'm very much hoping that we can coalesce around wording as strong as "do not overuse this clause".
    There are people whose content work needs large-scale removal (not to mention Lugnuts by name... Oh, oops!) but that shouldn't be done under BURDEN. It needs a prior consensus.
    BURDEN is the right tool for small numbers of disputed sentences or paragraphs. It can be, and has been, misused to follow someone around challenging their content work on a very large scale. This is not a pleasant experience for the victim. It feels like griefing.—S Marshall T/C 08:05, 12 September 2025 (UTC)[reply]
    As a watcher of this page who rarely pitches in, I'd like to say that I strongly support a "do not overuse" wording. People aren't run by computer programmes, and such extra words help to make sure we don't forget "common sense". I also disagree with the relativism argument, if I may call it that. People will tend to have highly overlapping understandings about what these words are envisaging. If editors work in good faith then such a wordings do help in practice.--Andrew Lancaster (talk) 10:44, 12 September 2025 (UTC)[reply]
    There are policies to stop POINTY or DISRUPTIVE behaviour, in behaviour policies rather than a content one. -- LCU ActivelyDisinterested «@» °∆t° 14:23, 12 September 2025 (UTC)[reply]
    There are. Editors who're strictly following the letter of a policy page won't know they're being POINTY or DISRUPTIVE, though. We have to tell them.—S Marshall T/C 21:41, 12 September 2025 (UTC)[reply]
    Then we'll need to cover the behaviour issue caused by editors saying they don't have to reference their slop because the other editors are over using policy. -- LCU ActivelyDisinterested «@» °∆t° 22:36, 12 September 2025 (UTC)[reply]
    I agree that this policy, which has been abused in this way by a small number of editors, needs to have a brief statement about this. I don't think that anyone complaining that "Don't overuse this" means "Don't use this on anything I add" is going to be successful. We're looking for a balance:
    • You need to cite four types of content – but you don't need to cite other types of content.
    • You don't need to cite other types of content – but if someone demands a non-required citation, it's faster to cite it "unnecessarily" than to argue about whether it's one of the four types of content.
    • You need to cite what you add – but even if you cite it, it can be removed (e.g., bad content, bad sources, wrong article...)
    • If you didn't cite it, someone else is allowed to remove uncited content – but it's not usually the best thing to do.
    • If you didn't cite it, someone else is allowed to remove uncited content – but large-scale removals are usually the wrong thing to do.
    We have a problem that sounds like this:
    • A: I hereby CHALLENGE all uncited content from here to infinity! I shall blank it all!
    • B: Um, that's POINTY.
    • A: My WP:V policy trumps your little behavioral guideline. I am allowed by the One True™ Policy to blank it all, and you can't stop me!
    That editor needs a note in this policy to rein him in. WhatamIdoing (talk) 22:46, 12 September 2025 (UTC)[reply]
    No that editor should be taken to an appropriate forum for disruptive behaviour, rather than giving one side in an argument a get out of jail card. Why should someone misusing "Don't overuse this" be handled any differently than those who are over using it, and if one is less likely to be censored by the community why should policy be changed in against that community. -- LCU ActivelyDisinterested «@» °∆t° 22:52, 12 September 2025 (UTC)[reply]
    Have you seen any of those discussions? You can only get newbies blocked for blanking large swathes of uncited material. Anyone who knows how to play the game will spend a couple of days at the drama board and leave with nothing worse than a "reminder" that this is supposed to be a collegial place.
    It's not a get out of jail free card for "one side". The point behind the balance is that we're requiring something from both sides (add sources for four types of content; use judgment and restraint), and giving something to both sides (don't argue about whether you 'should' have to cite it; use judgment and restraint). WhatamIdoing (talk) 19:17, 13 September 2025 (UTC)[reply]
    We have a much larger problem of unsourced content that verges on OR, if it isn't actually OR. And many editors who when pushed to show sources get upset they their content has been challenged. -- LCU ActivelyDisinterested «@» °∆t° 22:59, 12 September 2025 (UTC)[reply]
    Technically I guess that's more what NOR is about, but in any case I don't think that including "don't overuse" is in anyway going to distract away from the core messages of these policies. I think no-one is proposing any kind of watering down of those. I think it is wrong to use the term "get of jail" because by all accounts no one is proposing any kind of exception to any rule, or any change at all really? It is just a reminder to be reasonable.--Andrew Lancaster (talk) 14:00, 13 September 2025 (UTC)[reply]
    @ActivelyDisinterested, I'd like you to tell me more about problem of unsourced content that verges on OR. Do you mean that if the article says "Alice Expert is an author", then that verges on OR, but if it says exactly the same thing, but with a little blue clicky number after it, then it doesn't? WhatamIdoing (talk) 19:19, 13 September 2025 (UTC)[reply]
    No I'm mean that there is a lot of content without without sources that appears to be OR. -- LCU ActivelyDisinterested «@» °∆t° 20:15, 13 September 2025 (UTC)[reply]
    Yes there is. I'm not saying "do not use this clause". I'm only saying "do not overuse this clause".
    What I intend is that when you find one or two things in an article that aren't verifiable, you get to remove them. But if you find seven things in an article that aren't verifiable, you either (a) head to the talk page to start a discussion, and while you're doing that, you leave them in; or (b) you head to AfD to start a discussion, and the disputed material stays up until that discussion is closed; or (c) you head to WP:AN/I and start drama about the people responsible for all this unverifiable content, depending on the situation. What wording do you think would have that effect?—S Marshall T/C 21:38, 13 September 2025 (UTC)[reply]
    (a) doesn't seem like a desirable outcome, and (b) only if "seven" means "enough to think the article is a hoax". Nikkimaria (talk) 21:50, 13 September 2025 (UTC)[reply]
    Unfortunately there are entrenched editors who write articles as if they were essays based on their person knowledge, and who dislike "fools" questioning their content. "do not overuse this clause" is going to make matters worse, because it will be misused. Trying to stem one issue by creating another, when both are handled already by behaviour policies, isn't constructive. -- LCU ActivelyDisinterested «@» °∆t° 21:58, 13 September 2025 (UTC)[reply]
    Well, I guess we'll see if anyone agrees with you.—S Marshall T/C 22:52, 13 September 2025 (UTC)[reply]
    I do agree with ActivelyDisinterested. And I think the discussion is getting to be longer than necessary. Yesterday, all my dreams... (talk) 00:41, 14 September 2025 (UTC)[reply]
    Well I thought it better to bring these issues up now, before any full proposal to change long standing policy was put forward. -- LCU ActivelyDisinterested «@» °∆t° 07:18, 14 September 2025 (UTC)[reply]
    Yes, and I think your comment "there is a lot of content without sources that appears to be OR" is correct, and is in fact an understatement. The situation is serious in key articles. Yesterday, all my dreams... (talk) 09:50, 14 September 2025 (UTC)[reply]
    Guidance not to overuse this policy needs to be more concrete than "don't overuse it". Specific guidance on exercising restraint is already in the above drafts, and boils down to: Removing verifiable material just because it doesn't have an inline citation right now is unhelpful. How bold an editor should be in removing uncited material depends on how confident they are it is unverifiable (or subject to other reasons for removal). An editor familiar with the subject and what the sources do and do not say can be very bold about removing material, even large amounts of it. That is a service to Wikipedia, not a disruption. A reader new to a topic should be much more cautious. I don't think a top-level policy can be more specific than suggesting that Whether or how quickly unsourced content should be removed depends on the material and the overall state of the article.--Trystan (talk) 13:56, 14 September 2025 (UTC)[reply]
    I think that some editors struggle with the concept of uncited content being verifiABLE. For them, "verifiable, uncited material" is an oxymoron, because they're thinking verifiable means "supported by the cited source". WhatamIdoing (talk) 18:38, 20 September 2025 (UTC)[reply]

    IMO "over use" is gentle code for "Don't use it for other than it's intended purpose",("intended purpose" = the general intention of wp:ver). This is a common problem. Maybe saying it more directly would be a good thing but either a direct or indirect nudge on that would be a good thing. North8000 (talk) 15:53, 14 September 2025 (UTC)[reply]

    Axiomatic systems

    [edit]

    S Marshall, I should address your comment above that "if you find seven things in an article that aren't verifiable, you either (a) head to the talk page to start a discussion, and while you're doing that, you leave them in; or (b) you head to AfD". Who says there is anyone who bothers to read the talk page? There are now so many, many, many articles that some pages get no attention at all. And in technical cases 99% of Wiki editors know nothing about the subject. I kid you not. I should probably give you an example. Take a look at the talk page and edit history for Axiomatic system. The ONLY user who seemed to know anything was CBM, who is now gone, it seems. The rest are clueless. And most of the content was/is not just OR but reflects a basic lack of knowledge. I have cleaned if up a little, but do not have time to write a book on the subject. The best I can do is delete the nonsense. And this is a "key subject" in logic. There are many more like this. Yesterday, all my dreams... (talk) 10:09, 14 September 2025 (UTC)[reply]

    • With axiomatic system we're lucky, because as well as CBM, it's also been edited by Charles Matthews who, I am quite confident, is far from clueless about axiomatic systems. If we could clone Dr Matthews a few hundred times our maths articles would be a lot better.
      I've long said that articles about maths and logic need special treatment in WP:V because so many of them are, or can be, self-verifying. Look for example at Banach space; my position is that footnote #4 does fully verify the claim it supports. One day, when I'm Ruling Tyrant of Wikipedia and I get to write the policies, articles within the scope of WikiProject Mathematics will have additional considerations and special carve-outs that allow this.
      I fully acknowledge that our maths articles are one of several topic areas where very large numbers of important articles are maintained by an extremely small corpus of tired and overworked people. I don't know how to fix Wikipedia's problems with recruitment and retention.
      I do know that I'm writing policy to help us resolve conflict. We need policy for articles about post-1932 US politics, medical claims, the India-Pakistan conflict, etc. etc. etc. In cases like axiomatic system we don't need policy, because you're the only person editing it, so your editorial judgement prevails and we just have to hope we can trust you.—S Marshall T/C 12:14, 14 September 2025 (UTC)[reply]
      That is so funny. Just funny. Please take a look at the talk page for Formal system and my comment there on June 13, 2025. I pinged Charles Matthews and also suggested that someone should talk him into fixing that page which also needs help from FEMA. Result: nothing doing. As for trusting me on logic, modesty forbids a reply. Let me just note that Matthews did his PhD on number theory, so he probably knows more about that topic than yours truly. But I probably know as much about logic than any current Wiki editor. But no way to verify that, of course. And modesty forbids further... Yesterday, all my dreams... (talk) 12:47, 14 September 2025 (UTC)[reply]
      Non-gender-specific dude — either say who you are or please stop talking about it. Right now it sounds like you'd like people to be intimidated to disagree with you, but you won't specify why we should. Of course, no one should be intimidated to disagree with even Terry Tao; that isn't how Wikipedia works, in theory. But in practice, stop trying to have the best of both worlds. --Trovatore (talk) 21:34, 15 September 2025 (UTC)[reply]
      Yes: there's an awful lot to do, and only one of him, and he only has so much volunteering time. The Augean Stables are going to take forever to clean by hand.—S Marshall T/C 13:03, 14 September 2025 (UTC)[reply]
      The longer and short of it is that ActivelyDisinterested was correct above. Yesterday, all my dreams... (talk) 13:53, 14 September 2025 (UTC)[reply]
      For the use cases you're envisaging, he is. Which tells me that neither AD nor you have ever seen your watchlist get lit up by a griefer who's gone through targeting everything you've ever written with WP:CHALLENGES. You probably don't think that happens; or that when it happens, it's easy to deal with.—S Marshall T/C 14:51, 14 September 2025 (UTC)[reply]
      we will jump off that bridge when we get to it. And I have spent too much time on this, so will move on. Yesterday, all my dreams... (talk) 15:54, 14 September 2025 (UTC)[reply]
      Of course observer bias is unavoidable, I can only assume you haven't seen the difficulties caused by the other issue. Maybe there is an option that covers both without encouraging one bad behaviour in an attempt to discourage another. -- LCU ActivelyDisinterested «@» °∆t° 19:11, 14 September 2025 (UTC)[reply]
      Sigh. That version says nothing that encourages any bad behaviour at all.—S Marshall T/C 21:22, 14 September 2025 (UTC)[reply]
      Sigh. I'm afraid it does. Well when the changes to long standing policy are put forwards for consensus everyone can have a say. -- LCU ActivelyDisinterested «@» °∆t° 22:56, 14 September 2025 (UTC)[reply]

    I wasn't aware of having been pinged, can't say why. A message on my talk page should work.

    I was working on formal system in 2003, which is two decades ago now. Things here have changed greatly. The page now looks like the top level article in a bunch of articles related by summary style (WP:SUMMARY), and I would say that is how it should look. In other words, various sections that deal summarily but usefully with major topics that have their own articles. Possible criticisms are that the summaries are too concise, are too inaccurate, or possibly are inadequately referenced (though being picky about that isn't too helpful, if good referencing is available in the detailed article). So that is how verifiability fits in. Equally important, really, is that the overall organisation of the material serves the purposes of the encyclopedia.

    I was working on axiomatic system in 2004. It does seem to be in worse shape than formal system. Neither does it look much like the top level in articles organised under summary style, nor does it have a clear "logical flow" (ironically enough). Probably the latter criticism is more actionable. Systems of axioms are set up for reasons. Probably the article needs major surgery.

    Personally, I think everything written here before 2007 needs to be rewritten. Charles Matthews (talk) 21:32, 14 September 2025 (UTC)[reply]

    Yes, most of those articles need to be rewritten. And world hunger needs to end. But we both know that neither is going to happen in the next decade. The fact that the Wikipedia model has so far failed with respect to technical articles is not supposed to be mentioned here. But it is a fact. So let us not pretend that discussions on verifiability here will affect that. Those discussions assume that plenty of people care. In the technical articles that assumption fails, obviously. All that the few of us can do is to remove the most obvious errors. In the articles on logic the most glaring errors are/were based on the fact that the amateurs who wrote the material did not understand the difference between provability and satisfiabilty. I have removed some of those but there are plenty more. Yesterday, all my dreams... (talk) 09:17, 15 September 2025 (UTC)[reply]
    Given the incremental and ameliorative nature of Wikipedia editing, I don't see that the rhetoric is called for. Charles Matthews (talk) 06:44, 16 September 2025 (UTC)[reply]

    Next iteration

    [edit]
    Unpacked and rewritten
    Responsibility for providing citations Basics
    All content must be verifiable. The burden to demonstrate verifiability lies with the editor who adds or restores material, and it is satisfied by providing an inline citation to a reliable source that directly supports (A source "directly supports" a given piece of material if the information is present explicitly in the source, so that using this source to support the material is not a violation of Wikipedia:No original research. The location of any citation—including whether one is present in the article at all—is unrelated to whether a source directly supports the material. For questions about where and how to place citations, see Wikipedia:Citing sources, Wikipedia:Manual of Style/Lead section § Citations, etc.) the contribution. Once an editor has provided any source they believe, in good faith, to be sufficient, then any editor who later removes the material must articulate specific problems that would justify its exclusion from Wikipedia (e.g., why the source is unreliable; the source does not support the claim; undue emphasis; unencyclopedic content; etc.). If necessary, all editors are then expected to help achieve consensus, and any problems with the text or sourcing should be fixed before the material is added back.

    The cited source must clearly support the material as presented in the article. Cite the source clearly, ideally giving page number(s)—though sometimes a section, chapter, or other division may be appropriate instead; see Wikipedia:Citing sources for details of how to do this.

    Facts or claims without an inline citation to a reliable source that directly supports them may be removed. They should not be restored without an inline citation to a reliable source.

    Whether or how quickly material should be removed for lacking an inline citation to a reliable source depends on the material and the overall state of the article. Consider adding a citation needed tag as an interim step to removing unsourced material, to allow references to be added. When an article contains a lot of uncited information, it may be impractical to add specific {{citation needed}} tags. Consider then tagging a section with {{unreferenced section}}, or the article with the applicable of either {{unreferenced}} or {{more citations needed}}. For a disputed category, you may use {{unreferenced category}}. For a disambiguation page, consider asking for a citation on the talk page. When tagging or removing material for lacking an inline citation, state your concern that it may not be possible to find a published reliable source, and the material therefore may not be verifiable. When tagging or removing such material, please communicate your reasons why. Some editors object to others making frequent and large-scale deletions of unsourced information, especially if unaccompanied by other efforts to improve the material. Do not concentrate only on material of a particular point of view, as that may appear to be a contravention of Wikipedia:Neutral point of view. Also, check to see whether the material is sourced to a citation elsewhere on the page. For all these reasons, it is advisable to clearly communicate that you have a considered reason to believe the material in question cannot be verified. If you think the material is verifiable, you are encouraged to provide an inline citation yourself before removing or tagging it.

    Do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living people or existing groups, and do not move it to the talk page. You should also be aware of how Wikipedia:Biographies of living persons also applies to groups.

    Policy

    Each fact or claim must be verifiable. The burden to prove verifiability lies with the editor who adds or restores material.

    Facts or claims without an inline citation to a reliable source that directly supports them may be removed. They should not be restored without an inline citation to a reliable source.

    Do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living people, and do not move it to the talk page. You should also be aware of how Wikipedia:Biographies of living persons also applies to groups.

    Practice

    A fact or claim is "verifiable" if a reliable source that supports it exists. Facts and claims can be verifiable even if there is no citation for them in the article at the moment.

    The burden to prove verifiability is satisfied by providing an inline citation to a reliable source that directly supports the material. The cited source must clearly support the fact or claim as presented in the article.

    A source "directly supports" a fact or claim if that fact or claim is explicitly written in the source. For questions about where and how to place citations, see Wikipedia:Citing sources and Wikipedia:Manual of Style/Lead section § Citations.

    Once an editor has cited a source they believe, in good faith, is sufficient, any editor who later removes the material must explain. State specific problems that would justify its removal. If necessary, all editors are then expected to help achieve consensus. Problems should be fixed before the material is added back.

    Removing text is sometimes mistaken for vandalism. To avoid this, when using WP:CHALLENGE, best practice is to say in an edit summary and/or talk page note that you have removed the text because you don't think it's verifiable unless you are removing text that makes an unsourced, negative claim about a living person, in which case best practice is to remove it and say in an edit summary or talk page note that you have removed it using WP:BLP.

    {{policy shortcut}} WP:WOOKIEE

    Some claims are so uncontroversial that they should never need citations. ("The Thames is a river in England". "Paris is the capital of France.") But if good faith dispute exists about whether a claim is uncontroversial, it is better to add a citation than to argue about whether a citation is needed.

    Considerations

    Removing verifiable material just because it doesn't have an inline citation right now is unhelpful. Removing unverifiable claims is helpful, but replacing them with verifiable claims is much better.

    Before removing text, always check to see whether the disputed material is sourced to a citation elsewhere on the page.

    Where you feel the need to remove large quantities of material, it's best to start a discussion first.

    Instead of removing text, consider adding a tag, which allows other editors time to find references. When an article contains a lot of uncited facts or claims, it may be impractical to add specific {{citation needed}} tags. Consider tagging a section with {{unreferenced section}}, or the article with {{unreferenced}} or {{more citations needed}}. For a disputed category, you may use {{unreferenced category}}. For a disambiguation page, consider asking for a citation on the talk page.

    Misuse

    Do not use WP:CHALLENGE to selectively remove material to favour a particular point of view. Always make sure the article is neutral.

    Do not follow an editor around using WP:CHALLENGE on things they've written, and particularly, do not follow an editor from article to article doing this. Always be mindful of harassment and hounding.

    Do not use WP:CHALLENGE to gut an article that you have already nominated for deletion. In some circumstances you might use WP:BLP to do this.

    This variant tries to do an end run around the stonewalling about a "do not overuse this" clause by being more specific about what constitutes misuse.—S Marshall T/C 12:17, 17 September 2025 (UTC)[reply]

    I wonder whether the "burden" language could be removed. "If you want to add or restore material that requires an inline citation, you need to cite a source yourself, instead of expecting others to do that work for you"?
    The word exists, as in "A fact or claim is "verifiable" if a reliable source that supports it exists", seems to confuse people. Some people (correctly) say "A source exists – in my library!" Others say "No source exists – see, there are no little blue clicky numbers in this whole article!" Perhaps this would be clearer: "A fact or claim is "verifiable" if any reliable source has been published that says the same thing." WhatamIdoing (talk) 19:18, 20 September 2025 (UTC)[reply]

    We might be stuck with the "burden" terminology because WP:BURDEN has become such a meme? I agree that "exists" is likely to be misunderstood and this paragraph needs changing.

    Before Revised
    A fact or claim is "verifiable" if a reliable source that supports it exists. Facts and claims can be verifiable even if there is no citation for them in the article at the moment. A fact or claim is "verifiable" if a reliable source that supports it could be cited, whether or not there is a citation for it in the article at the moment.

    And also

    Before Revised
    {{policy shortcut}} WP:WOOKIEE

    Some claims are so uncontroversial that they should never need citations. ("The Thames is a river in England". "Paris is the capital of France.") But if good faith dispute exists about whether a claim is uncontroversial, it is better to add a citation than to argue about whether a citation is needed.

    		{{policy shortcut}} WP:WOOKIEE

    Some claims are so uncontroversial that they should never need citations. ("The Thames is a river in England". "Paris is the capital of France.") But if there is good faith dispute about whether a claim is uncontroversial, it is better to add a citation than to argue about whether a citation is needed.

    How about this?—S Marshall T/C 21:16, 20 September 2025 (UTC)[reply]

    "Challenge", as you have framed it, as only in contra to the person's good faith belief, is a poor indicator because there are many good reasons to challenge content beyond verifiable and even within the verifiability rubric, these are suggested in the footnote c: "the source is unreliable; the source does not support the claim; undue emphasis; unencyclopedic content; etc." Alanscottwalker (talk) 14:53, 21 September 2025 (UTC)[reply]
    No, the proposed text allows WP:CHALLENGE for all those reasons. The constraint is that the challenger must "state specific problems that would justify [the disputed content's] removal." I haven't tried to enumerate them and I suggest we don't.—S Marshall T/C 15:00, 21 September 2025 (UTC)[reply]
    That's the problem I pointed to, the way you have framed it is limiting. 'Burden', as a catchword is more narrowly focused on 'provide your source', but the challenges don't end there. Burden is minimally met (and its yardstick is personal perception), the challenge is multifaceted and its operation is communal. Alanscottwalker (talk) 15:21, 21 September 2025 (UTC)[reply]
    That's not a change. It's how it works now.—S Marshall T/C 19:40, 21 September 2025 (UTC)[reply]
    You appear to be trying to not listen. Your proposal changes the words in the direction that I find objectionable. You may disagree with my objection, but don't pretend your proposal is not trying to change how this reads. Or perhaps you are unaware, that your use of CHALLENGE completely merges it with BURDEN, indeed makes the section cite itself to explain itself (just follow the links), but I'm saying they can't merge and they should not, as your proposal would have it. Alanscottwalker (talk) 20:49, 21 September 2025 (UTC)[reply]
    I'm trying to listen, and to overlook the levels of aggression and hostility that you're displaying, but I don't find your objection intelligible.
    Proposed text Policy shortcut
    Each fact or claim must be verifiable. The burden to prove verifiability lies with the editor who adds or restores material. This is WP:BURDEN.
    Facts or claims without an inline citation to a reliable source that directly supports them may be removed. They should not be restored without an inline citation to a reliable source. This is WP:CHALLENGE
    They aren't merged.—S Marshall T/C 21:48, 21 September 2025 (UTC)[reply]
    Aggression? What? Is it that you somehow did not understand when I said your use of challenge limits challenge? What you just now quote limits challenge. -- Alanscottwalker (talk) 21:54, 21 September 2025 (UTC)[reply]
    Can you explain exactly how, please. Explain it like I'm ten.—S Marshall T/C 22:10, 21 September 2025 (UTC)[reply]
    Let's start off socratically. Are burden and challenge two different things? Are they raised by different users? Is answering a challenge done solely by fulfilling burden?
    My answers are yes, yes, and, no, which I explained above as 'Burden', as a catchword is more narrowly focused on 'provide your source', but the challenges don't end there. Burden is minimally met (and its yardstick is personal perception), the challenge is multifaceted and its operation is communal. And, there are many good reasons to challenge content beyond verifiable and even within the verifiability rubric, these are suggested in the footnote c of the present policy: "the source is unreliable; the source does not support the claim; undue emphasis; unencyclopedic content; etc.
    Your text makes challenge the mirror of burden, which is fulfilled by meeting the limited burden, thus they merge. But they should be distinct. Alanscottwalker (talk) 22:24, 21 September 2025 (UTC)[reply]
    WP:CHALLENGE should probably redirect to a different section. I see WP:BURDEN as a subset of CHALLENGE. Blueboar (talk) 23:08, 21 September 2025 (UTC)[reply]
    On my screen, challenge and burden are shortcuts that point to the same section of the same policy. The plain meaning of that policy section is that you can remove text that doesn't have a citation and nobody can put it back in without a citation. If someone puts it back in with a citation, they've met the CHALLENGE and satisfied the BURDEN so you don't get to take it out again, but if you're not happy with their citation, you take it to the talk page and everyone tries to achieve consensus. That's all it says, so that's all that my revision says.
    It does not say anything else.
    You can remove text for other reasons. You can remove it for being a copyvio or for being an unsourced negative statement about a living person or violating NOT or being a SYNTH or for many other reasons, but those are not CHALLENGE situations, so supplying a citation does not counter them. We do not want to make them CHALLENGE situations because we do not want to make supplying a citation into a counter for them.—S Marshall T/C 23:43, 21 September 2025 (UTC)[reply]
    CHALLENGE is currently pointing to the same section as BURDEN… but I don’t think it should (since you can challenge material for other reasons than just Verifiability). It’s easy to point a redirect elsewhere. Blueboar (talk) 00:38, 22 September 2025 (UTC)[reply]
    Well no, CHALLENGE is not on the page, at all, although it does secretly point to that section. Nor should we now readily introduce a word in policy that takes on a narrow peculiar specialized meaning when capitalized. And no, to fulfill BURDEN, all you need to do is cite a source that you personally believe in good faith is reliable and supports. The thrust of BURDEN is to get a source the community can examine and judge, so we can decide what to do next. Giving the source, is the 'to check' reason for verifiability, and thus sources are the irreducible control for almost all our content policy operations. You can't even judge cvio without a source. Alanscottwalker (talk) 00:51, 22 September 2025 (UTC)[reply]

    Folks, this policy is phrased in a pretty obscure way and I'm pleased that my attempt to rewrite it in plain English is shining a light on all the ways it is currently being interpreted. That's a good thing.

    When we're done writing what it says in plain English, there will be scope to discuss what it should say. But let's get the plain English version of what it says right now sorted first?

    I want to be very clear that my rewrite is not changing the rather narrow meaning of WP:CHALLENGE or WP:BURDEN as currently written, and these objections are seeking to add desired changes to the rewrite.—S Marshall T/C 07:36, 22 September 2025 (UTC)[reply]

    1) Well, your use of challenge is not plain language, as we have said.
    2) Nor is the current challenge as narrow as your proposal, as I have pointed out, the current version suggests multiple reasons for challenge (which is plain language meaning for challenge). Alanscottwalker (talk) 11:39, 22 September 2025 (UTC)[reply]
    S Marshall - I accept that your intent is not to change meanings… but I have to agree with Alan that your proposed language does in fact change meanings. Blueboar (talk) 12:12, 22 September 2025 (UTC)[reply]
    In what way, specifically?
    What I need from you two is either (preferably) a suggested way to phrase my rewrite, or (failing that) a clear explanation of what's wrong with what I've written that will lead to something actionable. What I'm getting is a recondite disquisition on the philosophical, socratic nature of verifiability, which while obviously fascinating doesn't do anything to improve the policy wording.—S Marshall T/C 12:40, 22 September 2025 (UTC)[reply]
    I'm having trouble understanding this (fair warning: I'm tired and hungry):
    @Alanscottwalker, a WP:CHALLENGE, in its simplest form, means I removed uncited text because it's uncited (and suspected of being WP:Glossary#unverifiable). After all, as the lead says, we're talking about "material whose verifiability has been challenged", not "material whose presence in the article has been disputed for any of a wide variety of reasons, most of which have nothing to do with verifiability".
    All that stuff in footnote c ("the source is unreliable; the source does not support the claim; undue emphasis; unencyclopedic content...") is about removing content for non-WP:CHALLENGE reasons. I wrote that footnote because of the people who were playing endless games of WP:FETCH, and who thought that if they couldn't CHALLENGE already-cited (including poorly cited) content, then nothing could ever be removed from any Wikipedia article again.
    This is a WP:CHALLENGE:
    • Alice adds something she thinks doesn't need an inline citation.
    • Bob reverts it with an edit summary of WP:CHALLENGED.
    • If Alice (or anyone else) wants to revert Bob, they have to add an inline citation when they restore the material.
    This is not a WP:CHALLENGE:
    • Alice adds something she thinks doesn't need an inline citation.
    • Bob reverts it, claiming that it's off topic.
    This starts as a WP:CHALLENGE but ends up being something different:
    • Alice adds something she thinks doesn't need an inline citation.
    • Bob reverts it with an edit summary of WP:CHALLENGED.
    • Alice reverts Bob and adds an inline citation.
    • Bob reverts it again, claiming that Alice's addition is bad grammar. Or a copyvio. Or anything at all, except a (false) claim it's still uncited.
    WhatamIdoing (talk) 01:39, 23 September 2025 (UTC)[reply]
    Well, if I recall correctly I wrote part of note c, and it was in part because the reasons for removal are many and in part because the standard given for fulling burden is internal (what they believe in good faith) to the one providing the source, but the reasoning in response needs to be external from another user.
    Regardless, in its simplest form, things are challenged by removal (which does not mean there are not more involved ways to challenge like taking it to the talk page, etc.). Again, CHALLENGE is not in overt usage in the current policy, although it does in secret anchor "challenge" to the BURDEN section, which in the current policy gives multiple reasons for removal. Which only makes sense because any challenge is to how an article reads or a topic is presented (for which there can be many good reasons). So, no, I can't agree with any of this fussiness about CHALLENGE, or more correctly, challenge, being so limited.
    I can't think of any good reasons, we would want to promote arguments about, 'NO!, you did not challenge, according to some peculiar limited meaning!' or 'NO! you did not use the magic words or the correct fomulairy for a challenge!' etc. If you want to say, 'no, you had no good reasons to challenge', simply say that, and that will cover it perfectly, and does not require a limited meaning. -- Alanscottwalker (talk) 05:39, 23 September 2025 (UTC)[reply]
    Oh, right, so we've got "challenge" in the WAID sense and "challenge" in the ASW sense and you two have meant different things by the word for all this time. That's actually simple, then. I see how I can finesse this. Rewrite to follow later.—S Marshall T/C 09:05, 23 September 2025 (UTC)[reply]
    • Let’s cut through the Wikilawyering… there are many reasons why someone might challenge a statement: it could be that the challenger thinks the statement is not Neutral. It could be that the challenger thinks the statement is Original Research. It could be that the challenger thinks it isn’t verifiable. Etc. etc.

    Unfortunately, challengers are not always clear as to why they are challenging (even when they leave an edit summary or a comment on the talk page). However, clear or not, the first step in resolving any challenge is for those who want the statement to remain to add an inline citation. That may or may not end the challenge, but it gives everyone a base for further discussion. Blueboar (talk) 12:22, 23 September 2025 (UTC)[reply]

    Iteration #3

    [edit]
    Unpacked and rewritten
    Responsibility for providing citations Basics
    All content must be verifiable. The burden to demonstrate verifiability lies with the editor who adds or restores material, and it is satisfied by providing an inline citation to a reliable source that directly supports (A source "directly supports" a given piece of material if the information is present explicitly in the source, so that using this source to support the material is not a violation of Wikipedia:No original research. The location of any citation—including whether one is present in the article at all—is unrelated to whether a source directly supports the material. For questions about where and how to place citations, see Wikipedia:Citing sources, Wikipedia:Manual of Style/Lead section § Citations, etc.) the contribution. Once an editor has provided any source they believe, in good faith, to be sufficient, then any editor who later removes the material must articulate specific problems that would justify its exclusion from Wikipedia (e.g., why the source is unreliable; the source does not support the claim; undue emphasis; unencyclopedic content; etc.). If necessary, all editors are then expected to help achieve consensus, and any problems with the text or sourcing should be fixed before the material is added back.

    The cited source must clearly support the material as presented in the article. Cite the source clearly, ideally giving page number(s)—though sometimes a section, chapter, or other division may be appropriate instead; see Wikipedia:Citing sources for details of how to do this.

    Facts or claims without an inline citation to a reliable source that directly supports them may be removed. They should not be restored without an inline citation to a reliable source.

    Whether or how quickly material should be removed for lacking an inline citation to a reliable source depends on the material and the overall state of the article. Consider adding a citation needed tag as an interim step to removing unsourced material, to allow references to be added. When an article contains a lot of uncited information, it may be impractical to add specific {{citation needed}} tags. Consider then tagging a section with {{unreferenced section}}, or the article with the applicable of either {{unreferenced}} or {{more citations needed}}. For a disputed category, you may use {{unreferenced category}}. For a disambiguation page, consider asking for a citation on the talk page. When tagging or removing material for lacking an inline citation, state your concern that it may not be possible to find a published reliable source, and the material therefore may not be verifiable. When tagging or removing such material, please communicate your reasons why. Some editors object to others making frequent and large-scale deletions of unsourced information, especially if unaccompanied by other efforts to improve the material. Do not concentrate only on material of a particular point of view, as that may appear to be a contravention of Wikipedia:Neutral point of view. Also, check to see whether the material is sourced to a citation elsewhere on the page. For all these reasons, it is advisable to clearly communicate that you have a considered reason to believe the material in question cannot be verified. If you think the material is verifiable, you are encouraged to provide an inline citation yourself before removing or tagging it.

    Do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living people or existing groups, and do not move it to the talk page. You should also be aware of how Wikipedia:Biographies of living persons also applies to groups.

    Policy

    Each fact or claim must be verifiable. The burden to prove verifiability lies with the editor who adds or restores material.

    Facts or claims without an inline citation to a reliable source that directly supports them may be removed. They should not be restored without an inline citation to a reliable source.

    Do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living people, and do not move it to the talk page. You should also be aware of how Wikipedia:Biographies of living persons also applies to groups.

    Practice

    A fact or claim is "verifiable" if a reliable source that supports it could be cited, whether or not there is a citation for it in the article at the moment.

    The burden to prove verifiability is satisfied by providing an inline citation to a reliable source that directly supports the material. The cited source must clearly support the fact or claim as presented in the article.

    A source "directly supports" a fact or claim if that fact or claim is explicitly written in the source. For questions about where and how to place citations, see Wikipedia:Citing sources and Wikipedia:Manual of Style/Lead section § Citations.

    Once an editor has cited a source they believe, in good faith, is sufficient, any editor who later removes the material must explain. State specific problems that would justify its removal. If necessary, all editors are then expected to help achieve consensus. Problems should be fixed before the material is added back.

    Removing text is sometimes mistaken for vandalism. To avoid this, when using WP:BURDEN, best practice is to say in an edit summary and/or talk page note that you have removed the text because you don't think it's verifiable.

    Some text violates more than one policy. In that case, when removing it, best practice is to add a talk page note giving all the reasons.

    {{policy shortcut}} WP:WOOKIEE

    Some claims are so uncontroversial that they should never need citations. ("The Thames is a river in England". "Paris is the capital of France.") But if there is good faith dispute about whether a claim is uncontroversial, it is better to add a citation than to argue about whether a citation is needed.

    Considerations

    Removing verifiable material just because it doesn't have an inline citation right now is unhelpful. Removing unverifiable claims is helpful, but replacing them with verifiable claims is much better.

    Before removing text, always check to see whether the disputed material is sourced to a citation elsewhere on the page.

    Where you feel the need to remove large quantities of material, it's best to start a discussion first.

    Instead of removing text, consider adding a tag, which allows other editors time to find references. When an article contains a lot of uncited facts or claims, it may be impractical to add specific {{citation needed}} tags. Consider tagging a section with {{unreferenced section}}, or the article with {{unreferenced}} or {{more citations needed}}. For a disputed category, you may use {{unreferenced category}}. For a disambiguation page, consider asking for a citation on the talk page.

    Misuse

    Do not use WP:BURDEN to selectively remove material to favour a particular point of view. Always make sure the article is neutral.

    Do not follow an editor around using WP:BURDEN on things they've written, and particularly, do not follow an editor from article to article doing this. Be mindful of harassment and hounding.

    Do not use WP:BURDEN to gut an article that you have already nominated for deletion. In some circumstances you might use other policies (particularly WP:BLP) to do this.

    578 words 597 words (NB: too long)

    The shortcut WP:CHALLENGE would still point to this section, as it does at the moment, until we get consensus otherwise.—S Marshall T/C 12:35, 23 September 2025 (UTC)[reply]

    Can I ask, that we change:
    "Removing unverifiable claims is helpful, but replacing them with verifiable claims is much better."
    With
    "Removing unverifiable claims is helpful, but depending on the context, replacing them with policy compliant claims is much better."
    Verifiability is not the only reason you should include something, and many times, trimmed articles are better, and sometimes 'less is more'. Alanscottwalker (talk) 11:26, 24 September 2025 (UTC)[reply]
    To save wordcount, as an alternative, can we move WP:VNOT including WP:ONUS to directly beneath WP:BURDEN?—S Marshall T/C 12:59, 24 September 2025 (UTC)[reply]

    RfC on BLPCRIME

    [edit]

    There is currently an RfC on WP:BLPCRIME at Wikipedia talk:Biographies of living persons § RFC: Amount of coverage in reliable primary news sources. voorts (talk/contributions) 00:13, 23 September 2025 (UTC)[reply]

    Discussion at Wikipedia talk:Citing sources § WMF seeking feedback on Reference Check

    [edit]

     You are invited to join the discussion at Wikipedia talk:Citing sources § WMF seeking feedback on Reference Check. Sdkb-WMF talk 18:09, 25 September 2025 (UTC)[reply]

    Addition

    [edit]

    Hello,

    This is regarding a PhD Dissertation on the origin of Sephardic-Portuguese surnames, particularly the 'Pimentel' surname, and the injustices that were visited against the Portuguese 'New Christians,' throughout centuries of the Holy Inquisition of Portugal, which officially began there in 1536 a.d.

    This dissertation thoroughly investigates, clearly and carefully explains this compelling & thought provoking historic event, against the backdrop of the global monopoly of the 'Spice & Sugar Trade' by the Portuguese Old & New Christian merchants & traders, at the apex of the Portuguese Empire. The sources are abundantly cited!

    Could a dissertation like this be worthy of inclusion in this section?

    Thank you. I appreciate your time and attention, and I look forward to your reply.

    Best Regards,

    George 2607:FEA8:121D:6500:6909:E7B6:E4D2:4F20 (talk) 18:04, 26 September 2025 (UTC)[reply]

    It wouldn't be appropriate to include it on this page, as this is the Wikipedia policy page for verifiability. You should ask you question on the talk page of whatever article you want to change. In general published dissertation are considered generally reliable, see WP:SCHOLARSHIP. That's not to say that they will always be reliable, or that just because they are reliable they must be included. Those are questions better answered in context on an articles talk page. -- LCU ActivelyDisinterested «@» °∆t° 18:13, 26 September 2025 (UTC)[reply]