Wikipedia talk:Verifiability/Archive 76

I feel like I've had too many conversations recently in which someone has selectively quoted BURDEN to claim that all uncited sentences are unable to be verified. The idea is that "other people" are entirely, hopelessly unable "to check that the information comes from a reliable source" unless a reliable source is presented to them on a silver platter.

I think it is probably time to add a sentence about the difference between cited and verifiable. NOR uses the line "verifiable, even if not verified", but "verified" could be understood as someone actually checking the source, rather than a source merely being listed in the article, so I'm inclined to suggest "verifiable, even if not cited" instead. A less stylish way might be to note that "if some material is not presently cited, but you can find a reliable source that directly supports it, then that material is already verifiable, and we would like you to edit the article so that the material is both verifiable and also cited".

What do you think? WhatamIdoing (talk) 07:38, 20 June 2022 (UTC)

I wonder if you are just trying to handle a situation where a room full of pedants still frequently use lazy shorthand, expecting others in the room to understand their shorthand, rather than to get diverted by someone dismantling their failure to be utterly precise. In other words, is this an issue to bother most editors with? The lead does already state everything must be verifiable but only "material whose verifiability has been challenged or is likely to be challenged" needs inline citations. So the message is already present in the lead that there is that there is text that is verifiable already but lacks and doesn't need a citation. Your proposed guidance would perhaps suggest editors should spend their time citing that Paris is the capital of France if they can find a source for it.

Perhaps the problem is with BURDEN:

"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 the contribution."

is absolute. The two sentences combined make this bold text require one demonstrates verifiability with citations in all content, which is wrong. We could drop that first sentence that just repeats what the lead has already stated. Replace this short paragraph with

"If material has been challenged or is likely to be challenged, the burden to demonstrate verifiability lies with the editor who adds or restores that material. This is achieved by providing an inline citation to a reliable source that directly supports the contribution."

Remember this is the section on "Responsibility for providing citations" so a statement "All content must be verifiable." is not vital here. However, if editors wish to keep it, I think the caveat I've added would mean that first sentence could remain:

"All content must be verifiable. If material has been challenged or is likely to be challenged, the burden to demonstrate verifiability lies with the editor who adds or restores that material. This is achieved by providing an inline citation to a reliable source that directly supports the contribution."

-- Colin°^Talk 13:45, 20 June 2022 (UTC)

Or perhaps insert a sentence: All content must be verifiable. Content might be verifiable even while it is uncited, but sometimes it is necessary to demonstrate verifiability, by adding a citation. (No changes, just adding the second sentence.) WhatamIdoing (talk) 18:44, 20 June 2022 (UTC)

Way, way more than "sometimes". How much uncited text do you find in GAs and FAs? This isn't 2003 anymore where editors welcome people just plopping down whatever they believe to be true. The overwhelming majority of our text should be and is cited. Crossroads ^-talk- 01:24, 21 June 2022 (UTC)

Using GAs and FAs as a guide to what a binding minimum standard like WP:V requires doesn't seem well-founded. -- Visviva (talk) 02:04, 21 June 2022 (UTC)

There might even be a difference between when it is "necessary" to demonstrate verifiability and when it is "desirable" to do so. WhatamIdoing (talk) 04:19, 21 June 2022 (UTC)

An aspect to consider: say I add a paragraph of info to an article, but because I am baffled by WP's citation system, I include the URL to the source on the talk page or even as the edit summary. That is a completely valid verified piece of info with a source, just that the source isn't in an inline cite, and meets this guideline. Obviously, not including the citation will make it hard on other editors to track, so it would be expected experienced editors can help to include the cite, but just because the cite is absent doesn't make it unverified. --Masem (t) 19:34, 20 June 2022 (UTC)

Masem, I think the issue is about articles that end-up in the NPP queue or somehow make it into mainspace without any sources - totally unsourced articles - and the expectation is for NPP to find the sources that will (a) verify N, and (b) verify that it is not OR. NPP should not be burdened beyond identifying unsourced articles as being unsourced, and then send them to draft or PROD them for failing to meet GNG, and V. BURDEN states the onus is on the article creator to provide sources - NPP are not article creators and we have enough on our plates without having to finish the work that may be the result of a bot or UPE or an editor who just wants to take credit for article creation but doesn't want to do the work associated with V and GNG. NPP was not established to complete unfinished articles. Our job is to not allow them to be indexed, and to send them back to draft. We should not be encouraging the creation of unsourced articles by doing the work the creators failed to do, or were unable to do because sources don't exist, or the topic was not notable. Atsme 💬 📧 23:57, 20 June 2022 (UTC)

Sure, but that said, if I as a new editor created an article and still confused about citation style, dumped all my sources into the talk page, that's still "verified" for all purposes. Mind you, we absolutely accept bare URLs wrapped in ref tags as acceptable inline cites (they can be improved but they aren't failing to source their information) and a NPP patroller can inform the user how to do this quickly. Masem (t) 00:43, 21 June 2022 (UTC)

Even if you didn't dump any sources anywhere, the subject could be notable and the contents could be verifiable.

Have a look at Geriatric sexology, which I ran across yesterday. It's three sentences long. It's been unsourced since its creation about 17 years ago. But ask yourself: Do you reasonably expect someone to have written about this subject? Is there anything in those three generic sentences that sounds like the kind of thing that you'd never find in a reliable source? I don't see any problems here, and I suspect that the reason it remains unsourced is because nobody else thought it was all that important to have those three unsurprising, basic sentences followed by citations, either.

I'm not saying that the article benefits from being unsourced. I'm only saying that its completely uncited contents are also completely verifiable, even though it's obviously an article that made it into the mainspace without any sources. I'd expect a NPPer to glance at it long enough to see that it doesn't contain any inappropriate jokes, have enough general knowledge to realize that since old people have sex, someone's going to write about old people having sex, click the [Mark page as patrolled] button, and move on without giving it another thought. I would not expect NPP to spend time searching for sources (though if you want to, you can find several books on the subject as well as many journal articles), finishing the article (unless it just happened to be a subject area that interested the individual), or deciding whether it's worthy of being indexed by a search engine. WhatamIdoing (talk) 03:37, 21 June 2022 (UTC)

I would expect at minimum a competent NPP patroller to at least dump the term into Google and make sure its not something made up, because it is very easy to make a unsourced article with compelling enough language to seem legit. (I'm thinking of the hoaxes involving "dihydrogen oxide" being dangerous as a outside-of-Wikipedia example). The NPP doesn't have to read any more than 2-3 pages of results - a whole minute of work - to validate its a legit topic and then flag the article for "needs sources". But its like Schrodinger's cat - without sources present somewhere (talk page or bare url or the like), there's no way to know if the topic can be validated without at least identifying the potential for sourcing. Masem (t) 04:48, 21 June 2022 (UTC)

Masem, how about this idea - when NPP tags an unsourced article with a CSD or PROD, and an admin rejects the tag, they should go ahead and find the sources to satisfy Notability and Verifiability - it only takes a minute, right? And while they're at it, they can do some CE and make sure MOS is followed. I like that idea. Let the rejecting admins do the work and find the sources if they choose to not encourage article creators to make sure their articles are ready for mainspace, and are verifiable by adding RS since the ONUS is on them. Why won't that work? NPP has a 14k+/- backlog and that doesn't count AfC so it seems perfectly feasible for our admins to help out. After all, the article could be a hoax. j/s Atsme 💬 📧 12:37, 21 June 2022 (UTC)

This might be the only time you and I have agreed, Atsme. I don't know if you're being sarcastic or not but I do think that declining CSD/prods (by pretty much anyone) if it's largely unsourced/unreliably sourced, should absolutely put the onus on the decliner instead of allowing a vague notion toward "sources exist" which I know of at least 3 editors who mass de-prod, de-tag and de-csd articles but fail to even provide a single one of those "sources" that exist. PRAXIDICAE🌈 12:44, 21 June 2022 (UTC)

Not sarcasm, Prax – I've never faked a sarcasm in my life.^[FBDB] And we actually do agree on a lot more than you think, despite our occasional differences. We're both here to help build an encyclopedia. (PS: I love your new sig). Atsme 💬 📧 13:34, 21 June 2022 (UTC)

All I am saying is that asking for a NPP patrolled to do a 1-minute Google search for the topic where the new article lacks any type of source confirmation, just to make sure it doesn't appear to be a hoax, is a very reasonable check. They don't have to add sources, just be aware that potential sources exist. At worst, then.maint-tag the article for lack of sources before approving Masem (t) 13:21, 21 June 2022 (UTC)

It appears that the opposite is being proposed at Wikipedia talk:New pages patrol/Reviewers#Proposal: increase unreviewed new page search engine NOINDEX duration. We must prevent Google's users from seeing any content at all until we are satisfied that the subject is worthy of being shown to Google's users. WhatamIdoing (talk) 18:57, 29 June 2022 (UTC)

This sounds like a problem with NPP taking on a bit too much, frankly. The expectations at WP:NPPCHK are wildly excessive for the basic protections against spam, vandalism and other facially inappropriate material that we actually need patrollers to protect the project from. If reviewers are overloaded because they're taking on extra work checking cites, or because this voluntary project has been structured in a way that makes them think they have to do a bunch of extra work, that's unfortunate but not really something a core policy should be addressing. -- Visviva (talk) 02:04, 21 June 2022 (UTC)

I agree, @Visviva. NPP used to be our defense against CSD-worthy new pages. Now they're trying to stop anyone from getting an imperfect article into the mainspace. This is a hugely bigger scope than the original goal and a completely unfair expectation to put on them. WhatamIdoing (talk) 03:09, 21 June 2022 (UTC)

Your linked example of geriatric sexology is not, in fact, notable, and is a fitting example of why it's good that NPP enforces reasonable standards, and why we shouldn't be so forgiving of unsourced material. The phrase, in quotes, returns almost nothing on Google Scholar - there is no evidence that this is a distinct field of sexology. Rather, research on this topic belongs in the much-better article linked in the "see also" - Sexuality in older age. I intend to pursue deletion or redirection of this article once this discussion runs out. I would very much not want NPP to approve a lazy and misleading article like this. Crossroads ^-talk- 01:26, 22 June 2022 (UTC)

I'm a dedicated mergeist, so put me down as a support. Just please add a few sources while you're at it. ISBN 9789021905662 by Zwi Hoch and Harold Lief (quote: "The lifecycle subdivisions of sexology are: embryonal–fetal, infantile, child, pubertal, adolescent, adult, and geriatric sexology") is plausible, if old. You'll find a similar description in one of the books by John Money. But I point out that the question on this page isn't whether it should be handled as a Wikipedia:Separate, stand-alone article. The question on this page is whether someone is "able" to "verify" that this information comes from reliable sources. WhatamIdoing (talk) 03:51, 22 June 2022 (UTC)

I think it would be good if NPP could enforce those standards. The problem is that they can't. We don't have the volunteer capacity or the processes to quickly deal with unsourced material in the way that (some) NPP reviewers would like to, so they're getting frustrated and blaming the policies and processes that we do have. – Joe (talk) 09:13, 23 June 2022 (UTC)

The idea is that "other people" are entirely, hopelessly unable "to check that the information comes from a reliable source" unless a reliable source is presented to them on a silver platter. Doesn't matter. As soon as a good faith challenge occurs, it is the BURDEN of someone restoring a claim to demonstrate that the claim is verifiable. Why is the (frankly) bare minimum of "cite sources for your claims" so oddly controversial with some, getting derided as requiring sources "on a silver platter"? We should not water down BURDEN whatsoever with anything confusing about how uncited text is supposedly verifiable. Again, nowadays, we don't want people just lazily plopping down whatever they believe to be true. Crossroads ^-talk- 01:24, 21 June 2022 (UTC)

Whether an uncited fact is verifiable is really a separate question from whether an editor should do something.

Consider "The capital of France is Paris".

We Y fully agree that if someone makes a good-faith challenge for that sentence, then editors must cite a reliable source for that uncited sentence.

The question we're addressing here, though, isn't whether WP:CHALLENGE rules apply to CHALLENGEd material. We all already agree to that and support those rules.

The question at hand is whether that sentence is actually verifiable right now, despite not having a citation after it right now, or if – according to some editors – the sentence is unverifiable unless and until someone adds a citation (...which nobody is "able" to do, because it's "unverifiable", right?).

Do you have an opinion on whether editors are able to verify that uncited sentence? WhatamIdoing (talk) 03:17, 21 June 2022 (UTC)

Responded below. Crossroads ^-talk- 01:41, 22 June 2022 (UTC)

Perhaps it is the nature of the discussions I've been in, but I think I've had more difficulty with editors trying to satisfy verifiability with vague and unspecific citations, for example citing an entire 10,000+ page book, instead of citing a specific chapter or page within that book, than I have with editors trying to do what WAID has said when opening this section. Sideswipe9th (talk) 01:33, 21 June 2022 (UTC)

FWIW I have also observed that this confusion seems to be increasingly common. I think the distinction is very important, and should be addressed in the policy, since the lack of clarity on this point only serves to encourage ever more aggressive gatekeeping that harms the project. -- Visviva (talk) 02:04, 21 June 2022 (UTC)

• I mean, policy is already clear that citations are only required for things that are challenged or likely to be challenged. I'd be reluctant to tinker further in the way you suggest because I do think that when something is challenged, the citation needs to be clear - it is not sufficient, in the long term, to say "a citation exists on talk" or "there's a citation elsewhere in the article." I think that in situations where a citation can be easily produced WP:PRESERVE might lightly discourage massive wholesale removals, but ultimately we can't forbid them because the burden is on whoever wants to keep the text. So I don't think we want to encourage process-wonkery arguments about the difference between "verified" and "verifiable" - if you are at the point where you're actually arguing with someone, then the bottom line is that whoever wants to preserve the text is the one with the actual responsibility to affix an actual citation to the disputed text in question, end of story. Yes, on a proverbial silver platter, so to speak. If people are going around challenging massive swaths of text on the belief that absolutely all text requires citations even on very new articles, they're mistaken and policy is clear about that; but once that challenge is made I'd be very reluctant to do anything that could weaken WP:BURDEN. --Aquillion (talk) 03:32, 21 June 2022 (UTC)

  What I want is for highly experienced editors to not say that an uncited sentence, such as The capital of France is Paris, is unverifiable or a violation of this policy.

  It is very easy to follow this sloppy path, and I'm sure you've seen editors do it:

  • Everything uncited is unverifiable.
  • BURDEN says "All content must be verifiable".
  • That means all uncited content is a violation of the Verifiability policy!

  This conclusion is false, because it is simply not true that all uncited content is unverifiable. The capital of France is Paris is a verifiable statement even when it is uncited. WhatamIdoing (talk) 03:46, 21 June 2022 (UTC)

  If there's an editor who's routinely challenging such WP:SKYISBLUE-style statements, then you might be able to make an argument that it's a conduct matter rather than an issue with the underlying content being challenged. However, I'm not sure that this issue is endemic enough that a change to policy is warranted...though I might support further clarification on the differences between "verifiable" and "cited". In my personal experience, when I challenge "obviously true" statements, I tend to end up in arguments with editors claiming that the information doesn't need to be sourced because it's "obvious"...nevermind that the most productive option would be to simply produce a source rather than bickering over the need for one (maybe that should be in the policy), and then frequently the first source they (finally) provide is a bad source. DonIago (talk) 03:54, 21 June 2022 (UTC)

  I am looking only for a clarification on the differences between the words "verifiable" and "cited". I am not looking to change the policy on when or whether something should be cited. (Editors in those disputes should WP:Let the Wookiee win; it's faster to cite obvious information than to argue about it.) WhatamIdoing (talk) 04:25, 21 June 2022 (UTC)

  I agree with the logic behind the concern. Editors can easily misinterpret BURDEN to mean that all material must have an inline citation. Instead of tinkering with the first paragraph, perhaps change the third paragraph to say something like: "Any material lacking an inline citation to a reliable source that directly supports the material may still be verifiable. If challenged, the material may be removed and should not be restored without an inline citation..." (bold is only to emphasize the suggested change). Maybe that's not front and center enough to cater to the concern, but just a thought. --GoneIn60 (talk) 04:17, 21 June 2022 (UTC)

What would that change even mean? The removal itself is a challenge. The policy already says, in bold, that only things that are challenged or likely to be challenged require a source; anyone who ignores such a flat statement is not going to be convinced by turning the entire section into tortured language that reiterates that point again and again. Additionally, I would strenuously oppose anything that would add additional bars to removing unsourced material. We are not required to do so, but unsourced material 100% can be removed, at any time, by anyone, solely based on the fact that it is unsourced, and when that happens anyone who wants to restore it should find a source. (If it really is WP:BLUE, good for you; finding a source should be trivial.) I would oppose any change to that in strongest possible terms. Indeed, my position is that entirely unsourced articles should be removable at any time for any reason. It is absolutely essential that we encourage people to provide sources for the things they add, and the best way to produce that pressure is to make it clear that if you don't add a source then anything you add can be instantly removed with no further discussion or explanation. Per WP:PRESERVE it is not always required to remove it; often a CN tag is better. But I would never support anything that would prevent anyone from doing so or add any barriers to doing so. Removing unsourced text is always, without exception, valid, fullstop. --Aquillion (talk) 09:04, 21 June 2022 (UTC)

"The policy already says, in bold, that only things that are challenged or likely to be challenged..." The "only" in your statement is assumed; it is not explicitly stated in policy, hence the potential for confusion. You and I get it, but novice editors can easily interpret this to mean that no citation always means not verifiable, and therefore any statement lacking an inline citation violates WP:V and should be fixed, tagged, or removed. BURDEN shouldn't be seen as an instruction to do so, that some action must be performed, because it is possible for verifiability to exist without a citation. If there's a way to add that clarification without being overly repetitive, then I'm all for taking that into consideration. --GoneIn60 (talk) 10:11, 21 June 2022 (UTC)

I agree with Gi60 here. The other false implication is that anybody can challenge any material for any reason, and that is absolutely not true and totally ridiculous. You must provide a valid reason in the edit summary. The thing we are not taking into consideration is that perhaps a challenging editor has mistakenly removed uncited material without realizing the material actually has been sourced elsewhere in the article. The confusion arises when editors think "uncited" means "unsourced", and this is due to a lot of our tools and templates such as Twinkle, Redwarn, and the Real-Time Recent Changes which propagate these misunderstandings by using such language as "added unsourced content" or whatever. If editors see that enough times they will think they are the same. Most, if not all, arguments tend to assume that a challenge, once issued, is infallible with the ultimate authority to force an inline citation to be produced without question, and this is simply unquestionably absurd reasoning that is in direct conflict with the spirit of collaboration and consensus. I think I have seen almost everyone agree that once a challenge has been issued the citation must be added, but why? Who says the challenger must always be right, or that their challenge is "indisputable"? I find it to be extremely odd, and even a little bit inappropriate that we are putting such an inordinate amount of good faith in challengers to the point that Doniago has even suggested it would be a matter of conduct issue if we had one making the mistake of routinely challenging uncited material that is clearly sourced, but any newish editor could easily make that same mistake and even veterans too. Huggums537 (talk) 04:20, 5 September 2022 (UTC)

@Huggums537, can you explain the distinction you're drawing between "uncited" and "unsourced"? Those two words are usually used interchangeably.

As a side note – and perhaps this is a key example of what you're thinking of – there has never been any clear agreement about whether a fact repeated in several places in an article must have a citation in every one of those places (assume someone who didn't read the whole article is complaining about it). WhatamIdoing (talk) 14:14, 5 September 2022 (UTC)

I know you're not generally asking, but to me it only makes sense that a repeated fact be cited each time (though that does somewhat beg the question of why the repetition is occurring). It's not an unreasonable burden to place upon editors to ask that they reiterate a source (indeed, it's easily accomplished), but it does seem unreasonable to put readers and other editors in a position of needing to confirm that a fact has in fact been repeated, and if so, whether it's sourced at any of its iterations. This would especially apply to newly-added information that may or may not be a repeat of information elsewhere in the article and that may or may not have been sourced when previously added. As a gnome, I have little interest in perusing the entirety of an article to confirm that yes, the information is repeated, and yes, it's sourced at least once. DonIago (talk) 15:06, 5 September 2022 (UTC)

It should be noted that this would also complicate the removal of sourced text, in that editors would then need to review the rest of an article to ensure that they weren't unintentionally removing a source for information elsewhere in the article. DonIago (talk) 15:12, 5 September 2022 (UTC)

Well, it's complicated, right? On the one hand, people blank "unsourced" content all the time, and then are told that WP:LEADCITE doesn't require citations in the lead, and they'll find the citation they're looking for in the body of the article. Or they'll be told that it's cited in the text and duplicated to the infobox (or vice versa). So this happens, and it's usually resolve amicably. In other cases, editors duplicate the citation to the contested spot. Since people don't read articles from top to bottom, sometimes a contested claim means that people are skipping to a section and not seeing the previous material, and editors decide that it's helpful (or just easier) to meet the apparent demand by duplicating the citation.

On the other hand, there are citations that ...don't make sense to repeat. Elephant (an FA) makes about a dozen direct and indirect claims that elephants are mammals. Should every instance of this claim be followed by an inline citation? Class: Mammalia^{[elephants are mammals]}... the largest terrestrial mammal^{[elephants are mammals]}... Unlike most mammals,^{[elephants are mammals]}... in most other mammals^{[elephants are mammals]}... the largest known among mammals^{[elephants are mammals]}... an unusual trait among mammals^{[elephants are mammals]}... as in most mammals^{[still mammals]}... While most mammals have^{[Have you figured this out yet?]}

Experienced editors mostly seem to know how this should work, but I'm not sure how you would write a rule for this that wouldn't invite wikilawyering. WhatamIdoing (talk) 15:41, 5 September 2022 (UTC)

I'm not sure it's all that complicated on a practical basis, most of the time, but I accept that opinions might vary on that. In my above text I was mostly referring to repeating information in prose. I think most editors understand that the lead is a special case, and that in theory an infobox is only supposed to be summarizing information that's repeated in the body of the article as well, so there should be sourcing in the article for anything in the infobox. I don't really have an issue with editors repeating cites in the infobox, though if they do so in the lead it tends to draw more scrutiny from me...but then, sometimes the lead isn't repeating information in the article body, which is a separate concern.

I'm assuming your elephant example as-presented is just hypothetical, because if the information is being repeated within the same section, much less within the same paragraph, then it's hard for me to imagine that any reasonable person would insist upon repeat citations.

I don't know whether a good rule is possible either, but based on my above statement, it might be that, "there's no need to repeat a citation within the same section of an article for repeated information" would address the most common scenarios. If we even needed this kind of rule, and I'm not really clear that we do. DonIago (talk) 16:42, 5 September 2022 (UTC)

The quotations from Elephant are scattered around the article. Few come from the same section. Except for the first (in the infobox, unsourced), the main point of the sentence is unrelated to the fact that elephants are mammals. For example, consider this sentence:

"Unlike most mammals, which grow baby teeth and then replace them with a single permanent set of adult teeth, elephants are polyphyodonts that have cycles of tooth rotation throughout their lives."

You want the citation for this to focus on cycles of tooth rotation, not on elephants' classification as mammals. WhatamIdoing (talk) 17:01, 5 September 2022 (UTC)

Hmm. But in that example there is a distinction being drawn between elephants and other mammals, so I would guess the source itself notes that contrast, though perhaps it only supports the final clause? If it does only support the final clause though, then that's a little eyebrow-raising as that would mean the sentence is synthesizing.

Otherwise, it being an FA and all, and not an article I'm especially invested in, I'd be very reluctant to try to edit myself. If the sourced claim about them being mammals is fairly early on in the article though, then I don't really see issues with it not being sourced elsewhere. I'm not sure that's what I'd consider optimal, but there's a world of difference and a world of relativism between 'acceptable' and 'optimal' in any case.

Anyway, I feel we're going a bit far afield here. I don't really have any proposals to make, nor do I necessarily feel there's the need to change anything anyway; I just think that editors shouldn't be expected to read an entire article before redacting unsourced material because the same information might be sourced elsewhere in the article, especially with regards to newly-added information. Mistakes may occur, but those shouldn't be laid at the feet of said editors, assuming anyone is even inclined to raise the argument, "But it's sourced three sections further down!" Rather in such cases it should be a calm, "Hey, that's sourced further down, but I can see how you might have missed that. I've reinserted the info and copied the source." DonIago (talk) 00:41, 6 September 2022 (UTC)

The problem isn't putting extra work on challenging editors to read the whole article before removing, the problem is that the only thing required to remove is that the material doesn't have an inline source. It then becomes a catch 22 for the uncited material that might actually be sourced if the material should not have been challenged in the first place. You say, "I don't want to do the due diligence", and I say, "I don't want my sourced material forced into the same category as quotations that require inline citations just because you refused to look at the fact it was possibly sourced elsewhere." Huggums537 (talk) 02:32, 6 September 2022 (UTC)

Depending on the size of an article, it can be a significant ask to require editors to read an entire article before they can remove any material that's apparently unsourced, especially if it's material that's been newly-added to the article.

Can you please establish how it's placing a significant burden on editors who wish to retain material lacking an inline citation to simply restore it with an appropriate citation if and when the material is challenged? Or, ideally, including inline citations at the time they're adding the material to begin with? DonIago (talk) 03:07, 6 September 2022 (UTC)

Well, that depends on what the editor values doesn't it? If the editor values their time, then they will simply add the citation and be done with it so your point would a good one that it would be asking more to check the article (you don't have to read it all) than to add the citation. However, if the editor values their principles enough to spend the extra time hashing it out on a talk page with several editors over a period of days or weeks about why the citation was not needed in the first place, then I would say that could turn into a much more significant task than simply checking an article for existing sourcing. Huggums537 (talk) 03:28, 6 September 2022 (UTC)

It just seems to me like we are expected to read the article and do due diligence to add material before writing, so why would it be any different for removing? We aren't supposed to just add willy nilly, so why remove that way? The problem is that most people don't think removing material is willy nilly if it doesn't have an inline citation, but they think it is if it isn't added with an inline citation, and that is a huge problem because no matter if it is easier or harder to do, it simply is not true. If I had any idea we were going to base rules on the easiest way to do things no mater what the truth is, then maybe I might feel differently. Huggums537 (talk) 03:45, 6 September 2022 (UTC)

You claim that we are expected to read the article and do due diligence to add material before writing. I don't think this is true. I think that the community's actual as-seen-in-practice policy is not to read the whole article unless you're planning to take the article to GA or FA.

For example, I edited Kings Dominion the other day. It's about 4200 words long. I checked (mostly with ⌘F) to see whether the content I wanted to add was already there, and I skimmed a couple of sections to figure out where it belonged. That's normal, expected, and acceptable.

I don't think that's a case of basing the rules on the easiest way to do things. I think that's a case of doing what's relevant to the task at hand. WhatamIdoing (talk) 04:02, 6 September 2022 (UTC)

I would like to resound those statements. Wikipedia has generally been very open and collaborative, and I support that approach to the maximum extent possible. Altanner1991 (talk) 04:37, 6 September 2022 (UTC)

Right, well I should have said "check" the article to keep in line with my previous statement, but your point is taken and I hope you see that the normal, expected, and acceptable amount of proofreading for adding material is not any more than I would be asking someone to do before they automatically yank some content just because it doesn't have an inline citation. Also, yanking content solely for that reason alone without even doing so much as a check to see if it could be sourced does nothing to support collaboration if the result is that it forces the content into a category of content that is now required to have inline citations such as quotes. If it only forced a mere discussion on the matter, but didn't force it into this category, then it would be collaborative. As it stands, I can essentially require any content (not just uncited) to have an inline citation no matter if it is already sourced or not even if I know it is sourced. This means that the choice of referencing or sourcing style is effectively taken away by rule. All one has to do to force this style is challenge for any reason, and it does not even have to be a valid challenge. In fact, it will be in favor of the challenger if they don't make a perfectly valid claim because once the challenge is made, others will say you can't put it back until you have argued both that the challenge was not valid, and for your reason to keep or that you simply give in to their demand to conform to their referencing style. Huggums537 (talk) 07:21, 6 September 2022 (UTC)

Another point I would like to make is that I would be perfectly fine with doing things the "easy way", if the easy way wasn't "inline citations only". IF users were offered more options for sourcing the uncited material other than strictly "inline citation only", then sure, I think it would be much easier and fair for everyone involved than asking someone to go through the trouble of going through the whole article to find sourcing, but we aren't asking editors to simply "verify the source" when we remove, or choose a way to reference the material. We are saying, "inline source or nothing!" It would be just as easy to ask to verify with a hidden comment as it would any other form of reference. I'm not saying hidden comments should take the place as references by any means, but it is a valid example for the sake of argument. Huggums537 (talk) 12:02, 6 September 2022 (UTC)

I guess the difference I am trying to draw between "uncited" and "unsourced" would be very similar to the same distinction you are drawing between Verifiable ≠ cited. It could more easily be explained by saying that by "uncited" I meant "without an inline citation", and by "unsourced" I meant that most editors editors erroneously think that if the inline citation is not present then the material is not otherwise sourced elsewhere in the article or anywhere in the world. So, they call it "unsourced" just because it lacks the inline citation, but that might not even be true since it could be sourced elsewhere in the article or somewhere in the world if that makes any sense. Huggums537 (talk) 03:05, 6 September 2022 (UTC)

I have also sought to distinguish "sourced" and "cited" in the policy page, and yes I would add "verifiable" to that list of often confused wording.

Unsourced should indeed be clarified: does it mean unverifiable or uncited? Altanner1991 (talk) 04:29, 6 September 2022 (UTC)

@Huggums537, is your list looking something like this?

verifiable

Someone could find a reliable source that says this, if they tried to.

cited

A very nice editor has made it really easy for others to find a reliable source that says this, by adding an inline citation.

sourced

Someone could find a reliable source that says this, and instead of starting with their favorite web search engine), they only need to look in the list of already-cited sources in this article to find at least one that Wikipedia:Directly supports this claim. (No very nice editor copied that citation over to this particular spot yet, though.)

The "un-" versions are the opposite: nobody could find it, no very nice editors have helped out, and the other sources already cited in the article for other material don't also cover this point.

If that sounds like what you're thinking of, then I think you need a different name for it. WhatamIdoing (talk) 05:06, 6 September 2022 (UTC)

Yes, that sounds right. Huggums537 (talk) 06:42, 6 September 2022 (UTC)

Editors generally don't make a distinction between the first and last state (you can find a source by asking Mr Google vs you can find a source by looking in the ==References== section), and they do use the words cited and sourced interchangeably.

Back in the day, we might have described this situation as using the ==References== section as (implicit) general references for the material that didn't have inline citations. WhatamIdoing (talk) 02:20, 7 September 2022 (UTC)

On rechecking I see that the policies were actually fixed sufficiently—no more need. Altanner1991 (talk) 07:47, 6 September 2022 (UTC)

As I've said before, the extreme overwhelming majority of our content does not consist of widely-known things like "Paris is the capital of France", so writing policy with that in mind is misguided. Rather, our content consists of things that most people don't know off the top of their heads. My concern is the 99.9999...% of content that tells most readers things they don't know, and that material should ideally be cited (and we definitely should not move toward saying or implying it's okay not to). And in that process, someone can quick cite the CIA World Factbook or something for where the capital of France is, rather than these lengthy debates about enabling failure to cite sources. Crossroads ^-talk- 01:41, 22 June 2022 (UTC)

I don't think that makes any difference. Consider:

• Here is information you know, and you can find a reliable source if you try. You are "able" to "verify" it, so it's verifiable.
• Here is information you don't know, and you can find a reliable source if you try. You are "able" to "verify" it, so it's verifiable.

Same end result, right?

I could agree that it's worse to have that second type of information unsourced, but I can't agree that it's unverifiable. The question here isn't whether uncited information is bad. The question here is whether uncited information is unverifiable. WhatamIdoing (talk) 04:19, 22 June 2022 (UTC)

As I've said at length at WT:NOR, I don't think that whether sources "exist" in some vague sense, but are never cited, really matters in practice. Trying to make the policy go on about that more is just confusing. Crossroads ^-talk- 23:44, 23 June 2022 (UTC)

• To be honest, other than the two sentences at the top of the BURDEN section (which I think should be fixed because their juxtaposition suggests verifiable = cited), I keep seeing WP:V reminding me that "Verifiable ≠ cited". Read the sentences beginning "When tagging or removing..." Those would make no sense if uncited meant unverifiable. I've read a lot of completely made up and untrue things about polices in the last few months but I don't think adding more text to those policies always a solution. If NPP needs guidance is how best editors should handle utterly uncited articles, say, then perhaps there should be some guidance for NPP, rather than a policy text change. -- Colin°^Talk 09:01, 21 June 2022 (UTC)

  To be frank, ^{because I'm always being honest,} the problem is not that NPP doesn't know what to do or cannot discern the difference between cited and verifiable. Verifiable means that the article, in its current state, should include RS that establish notability, verifiability and NOR. If it doesn't, then it fails and should not be allowed in main space until those issues are addressed. The policy needs to more clearly state that RS are required to establish both V, N and OR before an article or stub will be published in main space and indexed. Otherwise, V will conflict with all our other PAGs. What some may not be taking into consideration is whether or not the article includes any sources at all, it's not simply about inline citations. Stubs/articles that are sourced and have no inline citations are simply tagged by NPP with either ^{[citation needed]} or a big tag at the top of the page. NPP reviewers know the drill. My concerns are not about material that is added to an article that already contains RS. For all we know, some of those thousands of articles sitting in main space that are not sourced at all, and have slipped under the radar, could have been created by a UPE who has written a really nice promotional article that reads like an advertisement for a BLP or company they represent. How do those articles make it into main space? Could have been theft of a redirect, or an autopatrolled editor helped make it happen – I don't know, but we eventually find out if it's an admin or NPP reviewer with autopatrolled rights. There's a reason the decision was made to stop automatically giving all admins autopatrolled rights, and for having new reviewers either attend NPPSCHOOL or qualify via experience before they are given NPP user rights. Unsourced articles are much more serious than what some of the arguments have presented. It is not in the best interests of WP for any of us to turn a blind eye and simple accept unsourced articles under an assumption that they qualify. It's time to stop (a) putting bandaids on gaping wounds, and (b) encouraging editors to create articles in main space without RS or having first established V, OR and/or N. Perhaps some of our problems stem from a lack of qualified AfC reviewers, I don't know. I've read some of the comments by DGG who is inundated with new articles at AfC and carries more than his weight. He's beginning to burn out, and the same happens at NPP. Atsme 💬 📧 14:32, 21 June 2022 (UTC)

  Atsme is absolutely right. An article without any RS has no right to exist here. Independent RS notability has not been established. There is no evidence it's not a hoax or OR. It doesn't have to be fully referenced, but there must be enough RS to first establish V, OR and/or N. Period. Without RS we should treat it as the opinion of the author, and we don't allow such content in any article here. Send it back to private userspace or draft space. The initial burden of establishing V, OR and/or N is on the article's creator. -- Valjean (talk) (PING me) 21:14, 21 June 2022 (UTC)

  @Valjean, I disagree. @Atsme says:

  • "Verifiable means that the article, in its current state, should include RS that establish notability, verifiability and NOR."

  The actual policy says:

  • "verifiability means other people using the encyclopedia can check that the information comes from a reliable source."

  These definitions are not the same. Worse, one of them is wrong. If you read the next bit, you see her say 'The policy needs to more clearly state that "RS are required"...'. This indicates to me that Atsme's first sentence is a statement of what she wishes the definition of verifiability were, rather than what the definition actually is.

  It is not necessary to say that uncited information is technically unverifiable or a violation of this particular policy to say that unsourced articles are a blight upon humanity. If you want to establish rules that say unsourced articles are bad or banned (i.e., even when the contents are technically verifiable because someone could add citations), then I suggest making a proposal at WP:NOT. Please insist that at least one of those sources be WP:INDY. WhatamIdoing (talk) 03:32, 22 June 2022 (UTC)

  WhatamIdoing stated: "verifiability means other people using the encyclopedia can check that the information comes from a reliable source." That is exactly what it says, but it doesn't mean what you think it means. See Use-mention distinction which may be at issue here, so forgive me for using bold text and for highlighting the important aspects of the policy that are either being overlooked or misinterpreted. Based on what I've gleaned, the position in some of the arguments is that we should keep unsourced articles in main space and not worry that the article isn't sourced or is missing inline citations because it can be verified. The "how" of verification is what appears to be missing or misunderstood, as does the use of "material" which may include the entire stub/article. The following very important aspect of V supports my position: Any material that needs a source but does not have one may be removed. It goes on to say: Please immediately remove contentious material about living people that is unsourced or poorly sourced. Granted, not every sentence in an article requires a citation provided the article is properly sourced...only challenged material or material that is likely to be challenged are affected. If those sources are not provided the material can be removed and that comes straight from our core content policy as does the following: {{tq|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[2] the contribution.[3] NOTE 2 is also extremely important: 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. What part of the former and latter are causing editors to believe that unsourced articles are ok, and should not be removed from main space? But wait...there's more. Unsourced articles are automatically challenged as OR - and any experienced editor should already know that, especially if they've edited in a controversial topic area or BLPs. If a source is not cited, and/or there are no sources listed in the article at all, is it expected that our readers should leave the article to find those RS in order to verify the information in the article? If that's the case, why read WP? Here's a somewhat comparable comparison: a grad student turns in a master thesis without references or citing any sources, and then simply tells the prof not to worry about it – it's verifiable. What do you think happens? When an article is unsourced, it's more than just a matter of simply finding a source to verify that the topic is a valid one, and not a hoax. It involves finding multiple RS to satisfy GNG, V, and OR, as I've stated repeatedly, and am being criticized over. Far too many arguments are failing to acknowledge that the sources used must support what the author wrote, so if the author failed to provide the sources they used, and failed to provide inline citations for the material they added that is likely to be challenged, how are we supposed to find those same sources that were used to add that material? Cited sources and citations are how we separate the wheat from the chaffe. Bottomline: if the article creator failed to provide the sources they used and/or failed to include citations for material that is likely to be challenged so that our readers could easily verify a statement, opinion, or fact in that article it should not be allowed in main space. Atsme 💬 📧 12:39, 22 June 2022 (UTC)

  I think you're missing the point. It's not about whether articles should be sourced. It's about whether you say that uncited material is impossible to verify. This is a use-the-correct-jargon problem. Whenever these steps can happen:

  • Alice adds uncited information
  • Bob slaps a fact tag on it
  • Alice adds an inline citation to a reliable source that Wikipedia:Directly supports the information.

  then that information was never "unverifiable". That information was "verifiable, but uncited". In the third step, Alice now demonstrates verifiability, but the information already was verifiable from the very beginning. The policy says that all information has to "be" verifiable; it does not say that all information has to "be demonstrated to be verifiability".

  The "how" of verification has never been defined or restricted. The policy does explicitly state that information does not need to be "easily" verifiable by anyone.

  This is long, so perhaps numbers will help in case someone wants to pick out a bit for further exploration.

  1. "The position in some of the arguments is that we should keep unsourced articles in main space and not worry that the article isn't sourced"
     • ...is irrelevant. This discussion is about whether you declare something to be "uncited" vs "impossible to verify", not about whether it's a good idea to have completely unsourced pages in the mainspace.
  2. We all agree that "Any material that needs a source but does not have one may be removed."
     • But there is no sentence in this whole policy that says all "uncited" information is "impossible to verify".
     • There is also nothing in here that says completely unsourced pages are banned, and you've produced no evidence that all unsourced pages always contain material that needs a source. Have a look at User:WhatamIdoing/Christmas candy. Do you see anything in there that qualifies as contentious matter about BLPs? And direct quotations? Any material that has been challenged? How about any material that you think is Wikipedia:Likely to be challenged (i.e., by anyone who isn't trying to get blocked for bad behavior)? My answers are nope, nope, nope, and nope, and I bet yours are, too. If you can't find some material in that page that falls into at least one of those four WP:MINREF categories, then the whole page does not contain "Any material that needs a source", and therefore the whole article does not (technically) need any sources, because the whole article already complies with WP:V even without them. BURDEN doesn't apply.
  3. Granted, not every sentence in an article requires a citation provided the article is properly sourced...
     • The last phrase is wrong. Not every sentence in an article requires a citation even if the rest of the page is garbage. Whether this material is possible or impossible to verify does not depend on whether the rest of the article is properly sourced.
  4. NOTE 2 is also extremely important: 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. What part of the former and latter are causing editors to believe that unsourced articles are ok, and should not be removed from main space?
     • I think you might be reading more into Wikipedia:Directly supports and NOR than is really there.
     • I don't see anyone claiming that unsourced articles are "ok". I see them claiming that unsourced material is not always impossible to verify. We might be able to get an agreement that unsourced pages are Very Extremely Extra Bad – just not "unverifiable". Most of our unsourced pages contain verifiable information.
  5. "Unsourced articles are automatically challenged as OR"
     • Um, nonsense? NOR is the policy that actually says Articles that currently name zero references of any type may be fully compliant with this policy. Anybody who is automatically challenging unsourced articles as a NOR violation needs to go read the policy. If you run across an unsourced article, please don't put Template:Original research on it. Please use Template:Unreferenced instead. You might find this table of problems and their matching templates handy.
     • (If you wanted the equivalent statement for this page, I think it would be "Articles that currently name zero references of any type may be fully compliant with this policy, assuming that they contain almost no information at all.")
  6. "how are we supposed to find those same sources that were used to add that material?"
     • WP:V doesn't require that "those same sources" be cited, assuming anything has to be cited at all. If (and only if) we need to demonstrate verifiability, then we need to cite any reliable source that directly supports the material. It does not have to be the original source. In fact, there doesn't even have to be an original reliable source. The "original source" might be an editor's own memory, or something on Twitter. You should still cite a reliable source whose contents match the claims you're putting into the Wikipedia article. We do not insist on a strict interpretation of WP:SAYWHEREYOUGOTIT (so long as you actually did read the source that you put in the article).

  The TLDR:

  • The absence of an inline citation doesn't actually mean that people aren't "able" to "verify" the material.
  • This policy permits (some) uncited material.
  • This policy permits (a few) completely unsourced pages.
  • Editors should call uncited material "uncited" or "unsourced" or "bad", but not "unverifiable".
  • Editors should use the word unverifiable to describe material that is impossible for anyone, no matter how determined, to find reliable sources that could support it.

  WhatamIdoing (talk) 21:47, 22 June 2022 (UTC)

  Why does this hair-splitting of terminology matter? When content is CHALLENGED and removed it makes no material difference whether, unbeknownst to everybody, a reliable source exists out there somewhere ("uncited") or not ("unverifiable"). Either way, the unsourced content must not be reinstated, and the editor(s) supporting it must cite RS that directly support the claim. Really, what we should do is update WP:V's clauses from like 20 years ago where Wikipedia was an obscure site that lacked content and welcomed people adding whatever they thought was true, and bring it in line with current practice by formally deprecating adding material without citation. Atsme's comparison with the student turning in a thesis with no citations is the perfect reductio ad absurdum of this outdated idea of 'verifiability is just when sources exist somewhere out there'. Crossroads ^-talk- 00:06, 24 June 2022 (UTC)

  What you call "hair-splitting of terminology" is what other people call "clear communication". WP:Policy writing is hard, and it doesn't get any easier when people use the same words to mean significantly different things. WhatamIdoing (talk) 01:49, 24 June 2022 (UTC)

  @Atsme, with regards to your highlighted points; please note that they are only referencing material that needs a citation or material that has been challenged. It does not apply to all unsourced material. This is per later in paragraph where WP:BURDEN tells us what needs to be sourced: Attribute all of the following types of material to reliable, published sources using inline citations: all quotations, all material whose verifiability has been challenged, all material that is likely to be challenged, and all contentious matter about living and recently deceased persons. Huggums537 (talk) 00:52, 1 September 2022 (UTC)

• Citations do not verify; they merely facilitate. The process of verification is inherently difficult because, even when you have some evidence, analysis and understanding is required to confirm that the evidence corresponds with the assertion. The process of analysis itself requires verification and so you then get an infinite regress per What the Tortoise Said to Achilles. At some point, the reader just has to accept that something is true and that requires an act of faith. Often this faith is lacking and this generates much activity at WP:RSN.

As an example, consider the main page of Wikipedia and notice that none of the facts there are cited. But we mostly accept them on trust that someone has done the relevant checking to ensure that they all stand up. But this is a fallible process and so there is regular activity at WP:ERRORS.

And, even if everything has been done right, there will still be a disclaimer footnote which says emphatically that "WIKIPEDIA MAKES NO GUARANTEE OF VALIDITY".

So, as we are not claiming that any of our pages are exactly correct, we should just relax and save our energies for the most controversial and complex cases. It's like the dictum of Frederick the Great, "He who defends everything, defends nothing."

Andrew🐉(talk) 13:37, 21 June 2022 (UTC)

Andrew, I'm just going to jump in here and leave a comment. Tell me (ping) if I'm close to the point. You write: "Citations do not verify; they merely facilitate." I agree. We need to differentiate between "verifiability" and "verified". The initial content should be accompanied with a verifiable source. Hopefully, it also verifies the content. If it doesn't, then we have a case of "failed verification" and the "source" should be removed. The next question is what to do with the now unsourced content. If the content isn't seriously controversial or a possible BLP violation, it can be tagged with a "cn" tag. If controversial, then the content can be deleted. This is a matter of editorial discretion.

My point is that maybe we need to say something about the difference between "verifiability" and "verified". The existence of a citation/ref is not the end of the process, because some editors are careless or even sneaky vandals. Other editors should seek to "verify" the "verifiable" source. -- Valjean (talk) (PING me) 02:01, 10 July 2022 (UTC)

I can think of an example where it would reasonably be the end of the process, and that would be in the case where it is assumed that the subject of the article is the primary source like a book, and there is a plot section where it is obvious that the book (or the subject of the article) is the source, so a reference would not be needed for the plot section since anyone could verify from the book. See: Wikipedia:When to cite#When a source or citation may not be needed Huggums537 (talk) 23:46, 31 August 2022 (UTC)

• Question - Something that needs clarification for this discussion to continue productively: are we discussing the situation where an article has NO citations (at all)… or are we discussing the situation where an individual fact (within a sentence or short paragraph) does not have a citation? I think these are two different issues. Blueboar (talk) 14:55, 21 June 2022 (UTC)

  My understanding is that it's the latter; sentences or short paragraphs that do not have an inline citation. I agree that these are two totally separate issues, and judging from some of the responses so far, it looks like there may be some confusion. --GoneIn60 (talk) 21:26, 21 June 2022 (UTC)

  Unless it's a 2 or 3 sentence stub like this one: Jagatipala. Or you have to deal with articles like 3rd Alley, originally deleted back in 2008. Right now there are 182 pages in the NPP queue with no citations/sources. From what I've gathered, some editors are of the mind that as long as an article is verifiable it meets V, but there's no guarantee that V aligns with N, or OR. And doesn't using V in that context also suggest that an article's notability is also verifiable if sources are automatically verifiable? Why even bother to add sources if it's ok to make these assumptions? We could just let WP run as an aggregator, I guess.^[stretch] + ¯\_(ツ)_/¯ Atsme 💬 📧 22:10, 21 June 2022 (UTC)

  Information that (truly) meets V will always meet NOR. But there are no guarantees whatsoever that something complying with WP:V is a notable subject. If the only sources ever published about a business are the business's own self-published website, you could write a fully verifiable and even fully cited article, but it would not be a notable subject. WhatamIdoing (talk) 03:40, 22 June 2022 (UTC)

  It doesn't really matter. Uncited information that could be cited to RS is verifiable. We are "able" to "verify" it; it is "verifiable". This is true whether we are talking about the only three sentences on the page, or the only uncited three sentences in an otherwise well-sourced article.

  There might be some differences (e.g., whether it is a minor problem or a serious violation of all that is holy), but the presence or absence of other citations elsewhere on the page makes no difference in whether someone is actually able to verify that particular information. WhatamIdoing (talk) 03:13, 22 June 2022 (UTC)

  Uncited information that could be cited to RS is verifiable. We are "able" to "verify" it; it is "verifiable". I like this. Effectively explains in a nutshell why verifiable ≠ cited. Sometimes it's the simple analogies that really hit home. Perhaps some variation of this is the answer. --GoneIn60 (talk) 05:49, 22 June 2022 (UTC)

  Nope, it's misinformation - read the highlighted text above. Atsme 💬 📧 12:44, 22 June 2022 (UTC)

  Atsme, do you really think that the statement: “Trenton is the capitol city of New Jersey” becomes misinformation if we say it without citing a source? There are thousands of sources that support this fact (atlases, almanacs, history books, tourist guides, news papers… etc, etc). It is an extremely verifiable statement, even if we don’t actually cite any of them. Blueboar (talk) 13:10, 22 June 2022 (UTC)

  Blueboar, I've never even hinted at such nonsense – it defies common sense – but since you and I are usually on the same page in such matters, if you've seen something I've said that motivated your question, please point it out to me so I can fix it. My position throughout this discussion remains the same because I have not been convinced that it needs to change. I have made valid statements that are unambiguously supported by our core content policies, or at least I thought they were unambiguous. My highlighted text above remains as the crux of my position. Atsme 💬 📧 13:32, 22 June 2022 (UTC)

I think that there are two different important things being discussed here. The second raised one is the lobsided "prove/establish a negative" workload related to look for wp:GNG sources which is causing the in-progress collapse of NPP. It's not simply a google search. Even the easier ones (where a search of English language / Arabic character set sources is enough,) that still involves evaluating every hit for GNG sourcing criteria. And quadruple the person-hours required when the "source area" where we need to "prove a negative" has a non-english language with a different character set.

On the first topic raised, when there is a sincere concern about the verifiability/veracity of the material, I think that we need to strongly keep the burden for sourcing on the person seeking to include the material. But if you look at the problem cases, I think that a common theme is that there is no actual concern about the verifiability / veracity of the material, and none expressed, often resulting in deletion of uncontested sky-is-blue material. Probably the three most common cases of this are:

• POV warriors wanting to knock out material as a way of pursuing that end. And using this rule in synergy with "RS" classifications.
• Somebody that is having a pissing war with the editor involved
• Somebody being obsessive, thinking that an overkill approach is "just enforcing the rules"

I've long advocated for a measure that I think would dramatically reduce these problems without upsetting the apple cart. And that is to require any removal on such a basis include a statement of concern about the verifiability/veracity of the material. That alone would constitute fulfillment of the challenge criteria. Sincerely North8000 (talk) 13:12, 22 June 2022 (UTC)

You're talking about a completely different problem. Compare these two:

• Alice posts “Trenton is the capitol city of New Jersey”. Bob says "You put unverifiable information in Wikipedia! That's bad!"
• Alice posts “Trenton is the capitol city of New Jersey”. Bob says "You put uncited information in Wikipedia! That's bad!"

Bob's first statement is wrong. Bob's second statement is correct. Are you able to understand why the first is wrong but the second is correct?

NB: The difference is not in whether it's bad. Also, it has nothing to do with NPP or notability. It is only about whether Bob is wrongly declaring that it's actually impossible for someone to find that information in a reliable source, or rightly declaring that this statement was not followed by a little blue clicky number. WhatamIdoing (talk) 20:19, 22 June 2022 (UTC)

I think that something is confused here. First, I noted the second topic infused into the discussion (NPP and verifiablility) mostly to identify it as separate but I did not post on that, and then I posted on the the main topic which is basic wp:ver. And, of course I understand that distinction, I understood it in my first 5 minutes of reading wp:ver over 12 years ago. Now, on applying my point to your examples....the case that I noted is neither of those. It is where Bob is a POV warrior that wants to minimize coverage of Trenton. So they just delete the material, saying "unsourced" without making either of those two statements that you described. Or if it is sourced but to an attack able source, they just delete it saying "not RS sourced" And what I advocate is requiring Bob to make one of those two statements that you describe (either the right one or the wrong one) which would cut down on what could be called bad faith deletions or even a weaker one like "I have a concern that the statement might not be verifiable" North8000 (talk) 13:39, 23 June 2022 (UTC)

Sure, and we've agreed on this for years. But that's outside the scope of this conversation. I really just want people to understand that a particular bit of content can be technically, according to this policy, actually totally verifiable even if there are no sources named on the page. That's it. I want your POV warrior to say "unsourced" instead of "unverifiable". Getting to the point that we CHALLENGE only material that we have an actual reason to doubt is for the advanced class.

Perhaps I should put this another way: I do not want any editor who reads this policy from top to bottom to be able to say that this policy says that everything is unverifiable unless it is cited. Right now, we have a few good(!!!) editors who are saying that. They genuinely believe, or at least fervently hope, that this policy bans unsourced articles (in particular) and that uncited material is a violation of this policy. I want good people to read this policy and discover that "All material must be verifiable" does not mean "All material must be cited", or even that "WP:V says that every article must name at least one source". WhatamIdoing (talk) 23:02, 23 June 2022 (UTC)

• Thing is… even if Bob is acting in bad faith, the quickest and easiest way for Alice to tell Bob to shut the fuck up is to return the Trenton statement with a citation. Don’t worry about why Bob wants a citation. Give him one and move on. Blueboar (talk) 14:43, 23 June 2022 (UTC)

That happens after the bad-faith deletion and so is not related to my idea which is a simple soft requirement that would reduce bad faith deletions. BTW the most common case isn't for a totally unsourced item, it's when the POV wikilawyer uses this provision on an already-sourced item in tandem with a wikilawyer attack on the source or use of the source. North8000 (talk) 14:56, 23 June 2022 (UTC)

Exactly. That prevents this Bob this time, but does nothing proactive to prevent the next string of Bobs down the line... Huggums537 (talk) 06:21, 29 July 2022 (UTC)

• There was a lack of sophistication in the initial guidelines, where it was assumed that any source on the internet would be satisfactory verification; considering that good sources accessible on the internet were many fewer than now, it produced generally poor quality verification by current standards. Most articles from that period that have not been extensively revised have inadequate sourcing by current standards. W :What is the operative definition of verifiable? I think the usable meaning is that we have many analogous articles, whicch we have sourced properly, and know that there will be similar sources available. But it can also mean an article translated from another WP, without copying their references (a great temptation, because the formatting usually needs to be reworked completely, and because some excellent WPs, such as deWP, accept extremely general references, such as "for references, see the entry in the national bibliography"). I think the safest course is to insist on at least one reference that verifies eexistence and the general notability -- which is the current standard at BLPPROD. DGG ( talk ) 18:27, 23 June 2022 (UTC)

  @DGG, I think the operating definition of verifiability is "If I go to the trouble and expense of truly searching, I am able to find a reliable source for this specific claim".

  This really isn't about notability standards. I'd love to see a rule that says all newly created articles must contain at least one source that is (a) reliable for at least one statement in the article and (b) financially independent of the subject matter. We don't have that rule; I suggest proposing it at WP:NOT. WP:NOT#UNSOURCED would be a good shortcut. In five or ten years, we can start apply it retroactively.

  But really: All I want to achieve today is that people will stop saying "Aaaaarrgghhh, it's another stupid unverifiable sentence" and start saying "Aaaaarrgghhh, it's another stupid uncited sentence" whenever they have a reasonable expectation that someone could add a reliable source at the end of that sentence. That's it. Change one word. I want people to know that if someone is "able" to "verify" it, then it is "verifi-able". This really should not be a difficult concept. Maybe the simplicity of it is the actual problem? It's so patently obvious, so we need to drag something controversial or difficult into it, like NPP's challenges or how to establish notability or what to do about rules-lawyering POV pushers? But that's not what I want to accomplish. I want to tell people that "All material must be verifiable" means that it someone must be able to find a source, and not that someone must have already done so. WhatamIdoing (talk) 23:13, 23 June 2022 (UTC)

  I have to quibble… I believe that it is the responsibility of those who wish to add material to actually have found a reliable source for it… before they add it. However, they don’t always need to cite that source (the “Paris is the capitol of France” exemption). Blueboar (talk) 11:54, 24 June 2022 (UTC)

  As do I, and quibble I will because it appears to me that there is an attempt to lower the bar for verifiability in that material included in an article doesn't have to be corroborated if one RS has been cited. Granted, the primary purpose of V is to demonstrate that the material is not OR, and that content is determined by previously published information rather than the beliefs or experiences of editors. However, none of what has been proposed has taken into consideration the fact that there actually may be opposing views, and that we must maintain a NPOV. Furthermore, any effort attempt to dilute BURDEN requires community-wide consensus, especially the part that states Any material lacking an inline citation to a reliable source that directly supports[2] the material may be removed and should not be restored without an inline citation to a reliable source. That statement should not be altered, but I'm finding that this "no sources are ok" position to be rather disconcerting because it contradicts long standing policy and opens the door to contentious material and relentless debate. Atsme 💬 📧 01:40, 25 June 2022 (UTC)

  @Atsme, you said "none of what has been proposed has taken into consideration"...

  What, exactly, has been proposed? WhatamIdoing (talk) 18:06, 25 June 2022 (UTC)

  WhatamIdoing, if you need "exactly", then may I suggest a keyword search using "propose", or some form of it? This discussion is now tl;dr for me, and I'm not inclined to keep repeating what I've already explained. I think we're pretty much on the same page on some issues with the exception that quite a few of us don't think the policy needs to be changed. The opening sentence is quite clear, which you've also quoted but I've added some bold underline: ...verifiability means other people using the encyclopedia can check that the information comes from a reliable source. It doesn't say search online for a RS if you want to verify it. The RS that the material came from has to be with the article, hopefully in the reflist. It's the only way we can check where the information came from, otherwise we must assume OR, or that it's possibly rumor, a hoax or taken out of context. If we have time, we can search for a RS to verify that the material is supported, but then it may require a rewrite as it may not exactly support what's stated in the article, not to mention that it could be taken out of context. The options we have include tagging it ^{[citation needed]}, or removal. If there are no sources with the article at all, we have options for deletion. It's pretty simple. This diff supports my position, and I also agree with what Aquillion stated in this diff. There's nothing more I can add. Atsme 💬 📧 01:39, 26 June 2022 (UTC)

  Why do you think this policy requires editors to cite "the RS that the material came from", vs, "that the information comes from a reliable source" – "a" reliable source, as in "any reliable source in the whole world, even if it's not the specific RS that the material came from"? WhatamIdoing (talk) 04:50, 26 June 2022 (UTC)

  P.S. "This diff supports my position" isn't a diff. It also doesn't say that we need to cite the exact RS that the material came from. WhatamIdoing (talk) 04:51, 26 June 2022 (UTC)

  @Blueboar and @Atsme, I really do think that your "quibbles-n-bits" about a volunteer always being obligated to find a source before they add it is just a little bit of "Purina" bull stuffing because of what I said just a little earlier about how plot sections can easily be verified with the subject of the article (primary source) without a citation. The statement made by @Aquillion does not explain the fact that people like to argue that if something really is blue, then finding a source should be trivial, but in this case the source is already assumed, and easily verifiable so there isn't any need to cite it unless someone just wants to challenge it. Huggums537 (talk) 00:29, 1 September 2022 (UTC)

I don't feel we should use the current contested state of how we cite plot summaries as an example of anything. There have been a vocal number of people arguing that we don't need citations for plot summaries (or that "implicit" primary citations are sufficient); to my knowledge, discussions on that have never reached a clear consensus. For the record I strenuously disagree that plot summaries are in any way exempt from the clear requirements of WP:V or WP:PRIMARY - nor do I believe it would be possible for a consensus to override the most basic requirements of WP:V, since it is not subject to consensus; a challenged plot summary always requires a citation, fullstop. Primary sources can sometimes be used to answer that challenge, but only if the plot summary is not a "large passage" (or worse yet, the entire article), since that would violate the restrictions WP:PRIMARY places on the use of primary sources. But even the people who argue that plot summaries are somehow exempt from WP:V's requirement for citations when challenged or the WP:PRIMARY limitations on the use of primary sources have always held that they are a special case - I think that that's already a wrongheaded and deeply-mistaken argument, but I would strenuously oppose using that longstanding dispute as an argument to further erode our sourcing requirements elsewhere. If people are going to make that mistake, then I think we need to start being more aggressive about challenging uncited plot summaries that may not perfectly reflect the book, and to trim or reduce ones that have reached the point of violating the WP:PRIMARY restriction on using primary sources for large passages. --Aquillion (talk) 03:54, 1 September 2022 (UTC)

But, the whole point is not that we don't need the citations or that plots should be exempt. If some jerk wants to challenge a perfectly verifiable plot, then fine let the rules apply to the plot, but the point is that a simple straightforward plot is obviously verifiable through the primary source without even adding the citation. One could argue that a plot is already essentially sourced and cited by a general reference that isn't officially listed at the bottom in a reference section, but it does exist in the form of attribution to the title, author and so forth in the body of the article. By this logic, sources are not required if the appropriate sourcing is already provided elsewhere in the body of the article, and let's face it, if there is a full article on the topic, then both the primary and secondary sources likely already exist in the article for the plot, and whoever challenges, or argues it should be challenged is probably just being stubborn about their philosopyhy that every little "thee" or "thou" needs a cite. Huggums537 (talk) 20:58, 1 September 2022 (UTC)

• @WhatamIdoing: OK, I'm starting to see your main point in this thread. (Although if I read your last post precisely, it's to get people to change a wording in a sentence that I don't see people say.) But I'm starting to see it as raising awareness of the fact that the general standard is verifiable, not verified. And I think a non-controversial small step in that direction would be adding a sentence to WP:ver that adds emphasis to that point. But going beyond that small step starts to get complicated. The most common removal verbiage isn't what you described, it's simply like "removed unsourced material" and the second most common is something like "not a WP:RS" with the WP:ver based removal material. And usually the person wants to remove the material, and merely used wp:ver as a justification. What would you propose regarding that? My idea is to require challenges to include an expression of concern such as the sentence in your post. North8000 (talk) 12:54, 24 June 2022 (UTC)

Re WhatamIdoing's question on is the uncited sentence Paris is the capital of France verifiable, yes obviously. But if anybody challenges it then it still needs a citation. Even if you, or I, think that challenge is silly. Or even if made by a Nice supremacist who is POV pushing. Like Blueboar said above more poetically, the quickest way to dispatch that request is getting the requested citation, not argue over if it is technically verifiable or what the other editor's motives are. nableezy - 13:38, 24 June 2022 (UTC)

If there's a pattern of behavior in good faith editing that's tied to BURDEN, as observed by the OP, then it might be worth pursuing some additional clarity. However, it's a good point that the fix is often quicker than deciphering the motive. Maybe we need a clearer picture of the scale of the problem to help us weigh the benefit of action over inaction. --GoneIn60 (talk) 23:09, 27 June 2022 (UTC)

Also… as I have mentioned elsewhere in these threads… Wookies are MUCH more receptive to explanations about “why a source isn’t actually needed” AFTER they have “won” the challenge and a source has been provided. The conversation shifts tone from being confrontational “drama” to being friendly “education”. Blueboar (talk) 23:38, 27 June 2022 (UTC)

WE're talking bout the more reasonable common use of this which is on an unsourced item. But keep in mind that the most common disputed use by wikilawyer POV warriors is on a sourced item in tandem with nitpicking the source. North8000 (talk) 19:30, 29 June 2022 (UTC)

If a source does exist (or did exist and the editor is aware of it) but an editor is disputing it, then that isn't really the same thing as what we're talking about here - at that point you can point to and defend the source on talk, or take it to WP:RSN, or start an RFC over the specific and clear question of whether the source is sufficient for that specific statement. None of those things make sense if someone has removed a completely unsourced statement (ie. one that nobody has attempted to provide any source for at all at any point), because at that point there's nothing to constructively discuss - if you refuse to provide a source, all you can do is argue that the statement falls under WP:BLUE, which is overtly accusing the other editor of bad faith or incompetence (since BLUE is supposed to be immediately obvious.) That's not helpful, and if you do think one of those things are the problem you should take it to ANI rather than article talk. That said, even when there's already a source that someone found objectionable, unless the complaint seems glaringly absurd, I will usually first make an attempt to find an additional source - if I can find one, then doing that is faster and easier, and the article is ultimately better for having additional sourcing on a statement someone found objectionable, which will also avoid further rehashes of the discussion from others who might have the same objection. If you can't then sometimes you have to wade in and argue over the one source you have, but I don't think there's a harm in encouraging people to at least make an attempt to find a second source for contested statements first. --Aquillion (talk) 04:08, 1 September 2022 (UTC)

I'm a big Star Wars fan, and I think Let the Wookie Win is cute as hell, but I think letting someone know about a source you found for the challenged material or why it is verifiable on the talk page is just as fast, and probably ends up being just as productive as providing the citation. Just as N8000 suggested, you might be dealing with a POV warrior who could just as easily find some other excuse for removal after you add the citation, and you would just as likely find yourself ending up on the talk page anyway, so you might even save yourself some time by skipping adding the citation... Huggums537 (talk) 00:43, 1 September 2022 (UTC)

... specifically to Wikipedia:Verifiability#Self-published or questionable sources as sources on themselves, which mentions this as a key element of that branch of the policy. Seems like it would be navigationally useful. I already requested this at WP:AfC, but it occurs to me that as it'd presumably involve a corresponding edit to this page, editors here might have a view either way. 109.255.211.6 (talk) 10:21, 3 October 2022 (UTC)

Oops, apologies! Wrong section. Fixed. 109.255.211.6 (talk) 13:54, 5 October 2022 (UTC)

I Will Create An Article With This Name SpyridisioAnnis (talk) 10:51, 6 October 2022 (UTC)

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

Amir k wurma (talk) 08:32, 8 October 2022 (UTC)

The best way Amir k wurma (talk) 08:33, 8 October 2022 (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 and provide a reliable source if appropriate. MadGuy7023 (talk) 08:37, 8 October 2022 (UTC)

does wikipedia have a mechanism by which a user can declare that they have verified content against provided references?

In vanilla wikipedia there is no way to know if information is accurate without consulting references. This can be a time consuming process, for example if many sources have to be cross-referenced.

Like a crowdsourced publishing service.

As it stands, there is no reliable way to tell if a particular piece of information was entered by a dilettante or by a highly meticulous person. Or to even tell if there had been a sneaky act of vandalism some time in the last year.

If users vouch for accuracy, track records are generated that can be used to estimate the chance that a certain claim of accuracy is true.

if verified content on wikipedia is linked to by other wikipedia articles, the discovery of an error can be escalated downstream.

If such a mechanism was in place, users are probably going to be more motivated to ensure accuracy.

Effort spent verifying information can be used like a currency. Users could verify content and in return be awarded with an effort of verification of content of their choosing.

If a record of this activity existed, these efforts would not go unnoticed. The quality of content would increase. Nowakki (talk) 14:12, 17 October 2022 (UTC)

Presumably, the editor that adds a citation is vouching for that citation. Of course, some citations that currently exist were added more than 20 years ago, and the editor that added one may have stopped editing almost that long ago. Citations to web sites generally have an access date, and editors who have clicked through to the web site and found it alive and relevant are encouraged to update the access date. Editors who find problems with citations can try to fix the problem. They also have a number of templates they can add to the article to indicate a problem they have not fixed. In the end, however, Wikipedia is edited by volunteers who are, more or less, anonymous. The closest thing to a reward system on Wikipedia is the reputation that editors acquire through their contributions to the project. In general, the community has rejected proposals that complicate editing, or smack of bureaucracy. Donald Albury 15:59, 17 October 2022 (UTC)

> Presumably, the editor that adds a citation is vouching for that citation.

There is no way of telling if an editor is reliable. There is no way of knowing which iteration of an article an author deems reliable. wikipedia editors do not apply the same level of meticulousness to all their edits.

> The closest thing to a reward system on Wikipedia is the reputation that editors acquire through their contributions to the project.

nobody is stopping an editor to go into a quid-pro-quo with other editors.

> In general, the community has rejected proposals that complicate editing

vouching for a piece of content is optional and anyone can add content just as before.

The community has not rejected the general proposal to make a reliable encyclopedia. And i think the community would be willing to pay a price to achieve it.

In my opinion, a reliable wikipedia is possible, and as it stands now wikipedia by itself cannot provide any reliability, unless external sources are verified every time or you already know that something you read is true. Nowakki (talk) 17:26, 17 October 2022 (UTC)

There is equally no way of telling if the user who wants to declare that they have verified content against provided references is reliable. We assume that Wikipedia:You are not a reliable source – neither the person who adds the content originally, nor anyone who claims to have verified it.

Additionally, we don't necessarily want people to verify the content "against provided references"; it's often more important to verify them against sources that aren't cited in the article. What if someone cherry-picked the one source that said the Moon is made of green cheese? Or the one source with an unfortunate typo in a critical statement? Inline citations are convenient, but if you want to know whether the article is good, you often need to look beyond them. WhatamIdoing (talk) 05:09, 19 October 2022 (UTC)

> There is equally no way of telling if the user who wants to declare that they have verified content against provided references is reliable.

yes there are ways.

• you may know the verifier for some time, because you are working on the same project and quacks like a good faith helper bee
• the track record of a verifier can be used otherwise, a percentage of verifications that have until the present day not been falsified

there is no guarantee, but at this point you have to take into account that the reference is used to write another wiki article, and that new article will point to the original verified claim and the new article can be flagged as needing maintenance if the verification is falsified.

> it's often more important to verify them against sources that aren't cited in the article.

you can and should still do that. verification of proper citation to make wikipedia more dependable is not required to solve the problem of accuracy in every conceivable way. it is just another tool to be used to achieve the goal. Nowakki (talk) 10:19, 19 October 2022 (UTC)

sorry forgot one thing: the original editor can introduce an insignificant error deliberately. a verifier can never know whether they are being audited and would have to be always meticulous. under such conditions if a verifier has a track record of 98% after 200 verifications, that would be good enough to rely upon in matter less severe than life and death. Nowakki (talk) 10:31, 19 October 2022 (UTC)

Please don't use Wikipedia for matters of life and death. While Wikipedia is relatively reliable, and in my opinion just as if not significantly more reliable than other encyclopedias that aren't Wikis (per the famous Nature study, more reliable than Brittanica), we don't allow citations to Wikipedia because it's not itself reliable enough as a source. So always follow the citations and pay attention to the edit history and content tags if you care about veracity to a high standard. Andre🚐 21:23, 19 October 2022 (UTC)

i am making a proposal for a change to wikipedia.

why do you explain the current state of wikipedia as if i am living under a rock?

if a table made from 10 crappy looking pdfs can be used to sum one column of it, without first having to go through 10 crappy pdfs, that would be an improvement.

wikipedia 1.0 will look different from wikipedia 0.4 we have now. Nowakki (talk) 21:55, 19 October 2022 (UTC)

i made a mockup

User:Nowakki/sandbox/Reliability

what do you think? if you were writing an article about the shipbuilding history of california in the 20th century on wikipedia, would you dare to link to this section with a <ref>?

your article will get notified when it changes.

if you instead copy over 5 references to your article, if one of the references is later proven to have been errorneous, you can only hope that the person who discovers that did so while browsing your article. Nowakki (talk) 11:31, 21 October 2022 (UTC)

Not sure if this is the right place to discuss this or not… but I have noticed that recently we are getting more discussions at RSN about the reliability of document hosting websites (ie websites that compile and republish documents). It is obvious that many newer editors do not understand HOW to cite the documents hosted on such websites (ie cite the original document, but also include a note that the website is a “courtesy link”). Do we need to clarify this at WP:V or WP:RS? (Or someplace else?) Blueboar (talk) 16:28, 22 October 2022 (UTC)

It should be in WP:CITE even if it is elsewhere too. Zero^talk 22:09, 22 October 2022 (UTC)

I don't think that it's normal to explicitly note that a URL is a Wikipedia:Convenience link. I think we've always assumed that editors were smart enough to figure that out. WhatamIdoing (talk) 16:46, 27 October 2022 (UTC)

Per the directive made here I am requesting comments for the proposal being made at: Wikipedia talk:Inline citation#Unverified statements about valid challenges Huggums537 (talk) 21:23, 29 October 2022 (UTC)

I have commented there… but I think one flaw in your RFC is that you start off by defining certain forms of challenges as “valid” (and thus, by implication, other forms of challenge as “invalid”). First, I doubt there is consensus on what constitutes a “valid” challenge (and, thus, by implication what would constitute an “invalid” challenge), and second… I am not sure that making such a distinction is appropriate to begin with. As I see it, there are multiple ways of challenging - all of them are “valid”. I suppose you could say that some are weaker/stronger than others (a “cn” tag would be a weaker challenge than blunt removal)… some are more/less helpful than others (a removal accompanied by a full explanation on the talk page is certainly more helpful than an unexplained removal)… but all are “valid”. Blueboar (talk) 15:15, 4 November 2022 (UTC)

(Also, it's not an RFC.) WhatamIdoing (talk) 21:24, 4 November 2022 (UTC)

Dang, I just realized that. What a goof! I will replace the "RfC" in the title with "Proposal" instead... Huggums537 (talk) 00:38, 5 November 2022 (UTC)

Thank you for commenting, @Blueboar. Why don't you enlighten us with one of your history lessons about where this idea about "valid challenges" originated from? 😁 Huggums537 (talk) 01:13, 5 November 2022 (UTC)

Also, I responded to your thoughts about defining "valid" and making such distinctions at the proposal... Huggums537 (talk) 01:49, 7 November 2022 (UTC)

There is a request for comments at Wikipedia_talk:Inline_citation#Alternate_proposal_for_valid_challenges on a proposal to clarify the differences between content removal related to missing citations, and the ordinary removal of content not related to missing citations. Your voting decision is appreciated. Huggums537 (talk) 20:44, 7 November 2022 (UTC)

Given all of the concerns that have been recently raised related to Twitter's new way of operating, I'm wondering how this would affect things such as WP:TWITTER and WP:BLPSELFPUB. It's seems that up to now, verifiable tweets from account with a blue check mark were assumed to come from the subject of the account. However, there've been quite a number of reports in the news how this can no longer be necessarily assumed since a fee was put into place to get accounts verified. There also appears to be quite a bit of upheaval going on at Twitter since new ownership took over which means whatever problems it had before per WP:RSPTWITTER might've increased, particularly with respect to citing verified accounts or subject-matter experts. -- Marchjuly (talk) 21:41, 13 November 2022 (UTC)

I think we can probably wait a couple of weeks before updating policies. Editors are generally pretty smart about this kind of thing. WhatamIdoing (talk) 22:10, 13 November 2022 (UTC)

• Agree that this is concerning enough that we should pay attention to what is happening, but also agree that it is too soon to change the policy. “Wait and see” seems like the best approach. Blueboar (talk) 22:20, 13 November 2022 (UTC)

At the pump: Wikipedia:Village pump (policy)#RFC: change "verifiable" to "verified" Levivich (talk) 16:53, 19 October 2022 (UTC)

@Levivich, to follow up on that short-lived discussion, would you like to talk about a mini Wikipedia:Glossary for this policy? I think we could start with something like this:

verifiable

Someone (other than the person who originally added the material) is able to find this claim in at least one published, reliable source. There are no limits on the amount of expense, effort, or prior knowledge required to verify content. A claim can be verifiable even if it is not cited.

cited

There is an Wikipedia:Inline citation to a source associated with this claim. The cited source may or may not be reliable for this claim. A claim can be unverifiable even if it is cited (e.g., due to errors introduced by subsequent editors); similarly, a claim may be verifiable even if the cited source does not verify it (e.g., due to accidentally citing the wrong source).

verified

A Wikipedia editor (other than the person who originally added the material) compared the contents of the cited source(s) against the contents of the Wikipedia article and determined that they have the same meaning. A claim that is not verifiable in the cited source(s) may or may not be verifiable in other, uncited sources.

(Alternatively, we could just copy from Wikipedia:Glossary#Citation, Wikipedia:Glossary#Source, Wikipedia:Glossary#verifiable, and Wikipedia:Glossary#uncited.)

I think that one major source of opposition to your proposal is that people didn't like the word you chose to describe it. Standardizing the jargon could help if you want to revive that idea in the future. WhatamIdoing (talk) 20:54, 14 November 2022 (UTC)

Thanks for the invite but I'm gonna pass. Levivich (talk) 21:31, 14 November 2022 (UTC)