Wikipedia talk:Verifiability/Archive 77

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

Archive 70

←

Archive 75

BURDEN and the removal of tagged content

Latest comment: 1 year ago452 comments24 people in discussion

We have a well-intentioned, experienced, and useful editor, Fourthords, who has received multiple talk page messages complaining about their removal of non-dubious content that has been tagged as citation needed for what they perceive as too long. They're not trying to check for a source first and are using BURDEN as their reason. I get that by a strict interpretation this is within policy, but in my opinion it's not helpful. I havent been able to convince them, nor have the multiple other editors who have complained. They just keep repeating BURDEN until the other person goes away.

I think this is a real problem, and it's been going on for years. There are NINE separate conversations on their user talk alone right now, and that only takes us back to February 2021. The last time I talked to them about it (two months ago), I was able to find a source for the content they'd removed in under two minutes.

My concern is that this kneejerk removal of nondubious, noncontentious, nondisputed content means now no one can ever actually FIX the problem because the unsourced content is now simply gone. No one can ever come along and think, "Oh, I know where to find that." I think we need to revise that policy to make it clear that content tagged as citation needed is not the same as content being in contention, which is what BURDEN is actually about. Valereee (talk) 11:47, 17 October 2022 (UTC)

Remember that our articles are never “finished”… material can be freely added - and removed. It happens all the time.

If something important has been removed from an article, someone will eventually come along and add it back in again (hopefully with a source this time). If it isn’t important, then its removal did not hurt the article. Blueboar (talk) 12:28, 17 October 2022 (UTC)

If the removal is done in a thoughtful manner, maybe, but when a paragraph or more is cut from an article without realizing that that information was referenced earlier in the article or when what is left is fragmentary and disjointed, then both makes the article less helpful to the reader and harder for an editor to fix. {{CN}} tags exist for a reason; they cue the reader to information that may not be accurate or verified and they cue editors to a problem that needs addressing. —Carter (Tcr25) (talk) 12:45, 17 October 2022 (UTC)

Well... not in theory. In theory, maintenance tags exist to prompt people to fix them. They aren't supposed to "warn the readers" (or even "cue" them), because we have Wikipedia:No disclaimers in articles. WhatamIdoing (talk) 04:53, 19 October 2022 (UTC)

But some of that information may not be easily available. Many older articles were likely written by people sitting in libraries while accessing sources only available there. I have a whole list of books to consult next time I get to my main branch because they're not available for checkout.

If material is tagged, fixing it takes:

Someone comes along
Sees the tag
Thinks, "I wonder if I can fix that?"
Goes and finds the source
Adds the citation

If material is simply removed, it takes instead:

Someone comes along
...who has subject matter expertise enough to notice something is missing
Reads the article carefully enough to notice something important is missing
Thinks, "I can fix that!"
Goes and finds the missing information
Adds the content back
With a source, so it doesn't get immediately removed or eventually tagged again

Valereee (talk) 13:12, 17 October 2022 (UTC)

I share Valereee's concern. WP:BURDEN does make clear the need for added content to be cited, but it also places an obligation on editors removing uncited content to attempt to fix the problem or to flag it. Removal is allowable if the editor truly believes the information may not be verifiable, but they should explain that. Justifying the removal based on BURDEN's obligation for the adding editor overlooks that the standards for citations have changed over time. The places where I've seen this problem most acutely are articles that were created a decade or more ago when sourcing practices and requirements were looser. Should those articles be improved and better sourced? Of course, but blowing them up by simply deleting chunks of content doesn't achieve that goal. In many cases, a general reference may already be on the page that covers the material in question. In other cases, edits over the years may have moved the proper citation to a different point in the paragraph. Looking at the article, considering it carefully, helps discover those sorts of issues and make for an easy fix. Cutting material because there isn't a footnote at the end of the sentence or paragraph doesn't. Similarly, just because a paragraph is footnoted doesn't mean the source is correct or appropriate, so a singular focus on the lack of a footnote (instead of actually reading and editing the content) is going to leave incorrect or dubious content in place. TL;DR: BURDEN requires thought and consideration, both on the part of someone adding content and on the part of someone removing content. If you aren't willing to look for sources and to make edits within the context of the whole article, then add a {{CN}} tag and jump to another article that's of greater interest to you. Cutting content that isn't a BLP violation or obvious bunk for the sole reason that it currently lacks a source in close proximity to the statement does more damage than good. —Carter (Tcr25) (talk) 12:59, 17 October 2022 (UTC)
Yes, and honestly I've had a general policy to add citation needed tags pretty generously if an article isn't in my general area of interest. If adding a tag means someone is more likely to remove the content than fix it, I'm rethinking that. Valereee (talk) 13:18, 17 October 2022 (UTC)
edits over the years may have moved the proper citation to a different point in the paragraph. I see this all the time. Even just deciding a paragraph needs to be split in two can mean an entire paragraph looks like it's uncited. Then someone drops a tag on it, and eventually someone removes it for being unsourced. Now we need our subject expert to wander through again before the content gets re-added. It's not good. Valereee (talk) 13:22, 17 October 2022 (UTC)

As the person who suggested to Valereeee to make a thread, I share her concerns. I believe that Fourthords' behavior is problematic, but I think the arguably black-and-white wording of the verifiability policy also has something to do with it. Going by Fourthords' strict interpretation, it boils down to "When editing, you and only you must add sources to back up your statements. If there's no tiny number at the end of a sentence, it must be removed", which is antithetical to the spirit of Wikipedia. I propose a rewording of the verifiability policy to make it clearer that things should be verifiable, not verified, and to be something along the lines of "When editing Wikipedia, you are encouraged to add citations to reliable sources, but you are not required to do so unless you are editing biographies of living persons or material that you find contentious. Even if you fail to add citations, your content must be able to be verified by other editors, although be aware that uncited material can potentially be removed. All statements are to be taken as potentially true; when you see uncited material, you are encouraged to check for sources; if you fail to do so, leave a citation needed tag; only delete them if you strongly feel a source does not exist, it is original research, or a hoax. Note that several sentences may be cited only once, or they may use a format different from the traditional superscript number." 2001:4453:532:1900:1C9F:D7D0:47CB:E0E8 (talk) 14:57, 17 October 2022 (UTC)

As proposer, I support these changes, but I am open to slight rewordings, as long as the gist of it still remains. 2001:4453:532:1900:1C9F:D7D0:47CB:E0E8 (talk) 14:57, 17 October 2022 (UTC)

If you could show us what you mean by something like this:

Change: lorem ipsum

To: lorem dolor

It would help people understand what you're proposing. From a brief look, though, 'only delete them if you strongly feel' is not likely to fly. Valereee (talk) 16:52, 17 October 2022 (UTC)

I think BURDEN as it is currently worded is fine. When I try to add citations to unsourced material, I often cannot find any reliable source for the material, or find that I have to extensively revise the material to conform to what can be supported by the reliable sources I do find. Even where citations exist, they are often to sources that are not reliable, and/or do not support the material they are cited for. The point of BURDEN is, if your do not want certain material removed from an article, find reliable sources to support it. And, or course, if someone removes material from an article, it is not lost forever, it is still in the page history. - Donald Albury 16:12, 17 October 2022 (UTC)

Per Donald Albury, WP:BURDEN is fine as is. If you don't want someone to remove text from an article you like, it's your responsibility to find sources for it. No one is under compulsion to do your work for you; if you want to write something in an article, you have to find the sources. You can't write something without a source, and then demand others find sources for you. It is a courtesy to allow a cn tag; but no one is under any obligation to add sources. Wikipedia does have a policy that sources should be cited, however, and "until the heat death of the universe" is not a reasonable amount of time to leave an unresolved cn tag. --Jayron 32 16:20, 17 October 2022 (UTC)

@Jayron32, this isn't about an article I like. If I like an article and it has cn tags, I find the citations. When I write an article, I source it. This isn't about me. This isn't about the heat death of the universe; for one thing it's impossible for any of these tags to be more than about 15 years old, I assume? I'd like to see them fixed in a reasonable time, too, but I'd rather we not delete likely-verifiable content simply because it's been tagged too long.

This is about whether we should be interpreting this policy as intending to refer to noncontentious, nondubious, nondisputed, nontrivial content that we think is probably correct. Valereee (talk) 16:47, 17 October 2022 (UTC)

You're allowed to find references and place them. That is entirely you're right. Another user is allowed to remove the information if they prefer to do that. Note, that WP:BURDEN already recommends that "noncontentious, nondubious, nondisputed, nontrivial content" should be tagged and given a non-trivial amount of time for someone to find references. However, "enough time to find references" is not "forever". Those are not synonyms. If something is tagged for a few hour or a few days, then that's probably inadequate, and I would say that is not enough time for people to find references, if they had wanted to. However, if something has been sitting around, unreferenced, for years and years, at some point, it's shit or get off the pot time. If someone wants information to remain in an article, they are the only person required to find a reference. If someone wants some bit of text removed, they are not required, ever, to do any extra work. The work is only done by the person who wants to add the text to the article. --Jayron 32 18:10, 17 October 2022 (UTC)

And who is the judge of whether some content is "noncontentious, nondubious, nondisputed, nontrivial". Opinions will vary. I do often revert newly added, unsourced material that I consider at all dubious or inappropriate in context. I rarely remove existing material unless I think it is very obviously a problem (or, I have searched for and have not found any reliable source that supports it), but I recognize that other editors may disagree either way with my assessment. Disagreements about whether content should be kept should be settled by whether or not reliable sources can be found to support the content.

Donald Albury 17:09, 17 October 2022 (UTC)

Well, that's the point, I guess. If an editor isn't removing the content because they think it's dubious or trivial or because they object to it, then it's not contentious content, surely? If they're removing it solely because it's been tagged for what they believe is too long a time, that doesn't make the content itself contentious. If they've tried to verify and can't, that's one thing. But removing it soleley because it's been tagged for too long surely is different from that. Valereee (talk) 17:20, 17 October 2022 (UTC)

Sigh… We have had this conversation dozens of times. So I will simply say what I always do: Yes, it is annoying when someone stands upon a strict interpretation of WP:BURDEN and removes material that could be fairly easily cited with some effort. But that annoyance is better than leaving potentially inaccurate information in an article.

And if someone removes easily cited material, just return it yourself with a citation. Arguing about whether it should (or should not) have been removed in the first place takes MUCH more effort, time and angst than simply shrugging and returning the material with a source. When someone stands upon a strict interpretation of WP:BURDEN… just “Let the Wookie Win”. Blueboar (talk) 17:06, 17 October 2022 (UTC)

Please don't sigh, I'm here in good faith. My concern is that if someone is systematically removing anything that has been tagged for "too long", then maybe we need to discuss whether that's what this policy actually means. Valereee (talk) 17:21, 17 October 2022 (UTC)

Let me simplify it for you. Yes that is what the policy actually means. While there's lots of explanatory text, the policy is a single sentence which is fairly straightforward. It is so central it is put in bold on the page: "The burden to demonstrate verifiability lies with the editor who adds or restores material" Notice, the person who removes the material is under no obligation to even do a simple Google search. They don't need to do anything. That is, in fact, the policy. Best practice is to only add material alongside a citation. If a person can't be bothered to do that, they can expect their added text to be removed at some point. --Jayron 32 18:13, 17 October 2022 (UTC)

And is the lack of a footnote on a sentence the sole criteria that should be used then in deciding that something is unverifiable? Is there any obligation to look at the surrounding text and sources to see if the information may only appear to be unsourced? An article that has been around since the early 2000s likely has portions that don't meet today's standards for sourcing. Is the best approach one that seeks to cut material without examination? I'd contend that all editing requires some thought and work, whether you're adding or removing content. Perhaps the bolding of that sentence in BURDEN is part of the problem; people stop reading and don't consider what other actions might be taken to address the apparent deficiency. This isn't a problem of unsourced material being added today; if something is added and reverted in 2022 because it's unsourced, that seems fine. The issue is with long-standing content that may have been sourced to an older acceptable standard or where subsequent edits may have moved a source away from a statement. BURDEN shouldn't be a defense for indiscriminate removal of content simply because someone doesn't see a footnote where they expect one, unless they do some work to examine the content. And if they can't be bothered to make that effort, then add a CN tag and move on. Carter (Tcr25) (talk) 18:53, 17 October 2022 (UTC)

This isn't about an editor adding or restoring material. This is material that has been there for years. This is about removing material that has been unchallenged for years. It's been tagged for a citation. Valereee (talk) 19:35, 17 October 2022 (UTC)

That something has been wrong for years doesn't make it any more right. --Jayron 32 12:27, 18 October 2022 (UTC)

However, when it was added may have been before inline sources were the standard. Or intervening edits may have moved the source from the now apparently unsourced content. A number of the comments here seem to view this through the lens of "lazy" editors adding content today, but that fails to recognize that the expectations of editors and the project have changed over time. That's one reason I find unexamined text removals solely because the source isn't evident troubling. WP:WIP, so updating content to meet today's standards should be expected, and I think that's better done by flagging content with {{CN}} tags (even if they linger) than by cutting material that isn't clearly bunk or a BLP violation. I think it is reasonable to expect someone questioning a statement to make the effort to check their intuition. If an editor removes something, particularly long-standing content that likely was added under different standards, after making an effort to determine verifiability fair enough, but the lack of an inline citation next to a statement shouldn't alone equal a presumption of OR, misinformation or laziness that must be excised. — Carter (Tcr25) (talk) 13:48, 18 October 2022 (UTC)

I agree, and the automatic assumption that something is wrong just because it's tagged as CN without even checking the material or sources is the biggest part of the problem others are describing. I think WP:USI is useful and WP:CRV should be upgraded to a guideline or whatever. Huggums537 (talk) 14:35, 18 October 2022 (UTC)

No one is saying it's right. Let's get off that toggle switch: it's not ideal to include unsourced content. But if the content is not dubious/not contentious, why are we putting a deadline on getting the citation in there? Valereee (talk) 21:00, 19 October 2022 (UTC)

Wikipedia:List of hoaxes on Wikipedia. Without citations to verify, how do we know anything is true? Chris Troutman (talk) 21:02, 19 October 2022 (UTC)

It is pretty funny. :) Yeah, I know. There are vandals. But seriously, do you think the upset editor at Rowing is a vandal, inserting hoax info? I don't. He might be inserting info because, gosh darnit, the little guys deserve to be listed here too! But I don't think he's a hoaxer. Valereee (talk) 21:10, 19 October 2022 (UTC)

No one is saying it's right. Let's get off that toggle switch: it's not ideal to include unsourced content. That is correct, and this is true of myself as well, but someone could say it is right with a BLUESKY argument if they wanted to. However that was not the purpose of my comment. The purpose of my comment was not to say that it's right, but to demonstrate that editors have this faulty inbuilt thinking process that says uncited content is inherently wrong just because it doesn't have an inline citation or just because it has a CN tag and keeping that view in their mind before they have even reviewed the material or sources.

@Chris troutman to answer your question: Without citations to verify, how do we know anything is true? You stop counting completely on Wikipedia citations for editorial research, and check for sources (outside Wikipedia). Huggums537 (talk) 19:02, 20 October 2022 (UTC)

Also, I fail to understand this strict obsessive adherence solely to the flawed and bolded part of WP:BURDEN when it is perfectly clear from context of reading the whole thing that it isn't unsourced or uncited material that is supposed to be removed, but unverified material as defined in bullet point # 2. It seems most editors arguing for the other side seem to wrongly interpret bullet point #2 as meaning "anything that gets challenged for any reason whatsoever". Huggums537 (talk) 00:07, 19 October 2022 (UTC)

@Valereee: Re: "if someone is systematically removing anything that has been tagged for "too long", then maybe we need to discuss whether that's what this policy actually means".

I think, yes, that is exactly what the policy means. When someone puts "cn", the main goal is to draw attention to the problem. This problem may be resolved, for example, (i) by removal of the "cn" tag if a consensus will be achieved that the tagged statement is obvious, and, therefore, it doesn't need a citation. Another way is (ii) to replace the tag with a reference to some source. However, what if nobody takes any action? In that case, any user may remove the tagged statement. Does this user have to take any attempt to fix the problem via "i" or "ii" methods? No. Remember, Wikipedia is being written by volunteers, so if someone sees a problem (a "cn" tag), but they have no time to dig into sources, they may choose the simplest way to fix the problem: just remove the problematic text.

However, that is not the end of the story. Removal of the text is another way to draw attention of other users. If the removed context is valuable, then anyone has a right to restore it, provided that the "cn" tag will be replaced with some real source. Otherwise, if nobody cares about the removed text, then, probably it was not a too valuable piece of information, so its removal causes no harm.

It summary, routine removal of the content tagged with the "cn" tag is just a normal part of the procedure, which works as follows:

1. If you see some text that needs to be sources, put the "cn" tag.

2. If other people believe the tag was not justified, and they achieved consensus about that, the tag will be removed.

3. If tag is not removed, someone will replace it with a reference to some RS.

4. If no reference was placed in a reasonable time, and nobody tried to discuss removal of the tag, that means nobody cares about that content, so everybody is welcome to remove it.

5. If removal is reverted, and the reference has been added instead of the "cn" tag, then the removal served its goal: attention of other users has been drawn to the problem, and the problem has been fixed.

6. If removal of the tagged content has not been noticed, then the tagged sentence was either unimportant, or (and) other users have been unable to find any reliable source. In that case, again, removal has played its role, and Wikipedia has been purged of a questionable content. Paul Siebert (talk) 03:39, 20 October 2022 (UTC)

Jayron32 hits this exactly on the head and I'm disappointed there is a divergence of views about this. I have realized that many editors only come to Wikipedia to write what they want read and gain self-indulgent validation from doing so, sources be damned. This will never be a real encyclopedia so long as we tolerate deviance from our professional ethic. Chris Troutman (talk) 18:44, 17 October 2022 (UTC)

I don't see this as deviance from our professional ethic. I see it as how do we best improve content that doesn't meet today's standards. Carter (Tcr25) (talk) 18:55, 17 October 2022 (UTC)

Editors today are adding content without adding reliable sources; this isn't just about articles created 15 years ago. If we don't insist on sources for content (old or new), then Wikipedia is just a webhost. For me, I would rather have no article at all than have a poorly-sourced one. Chris Troutman (talk) 19:03, 17 October 2022 (UTC)

This isn't about stuff being added today. It's about stuff that has been tagged for some time. Stuff that gets added without a source today gets immediately removed, as it should. Valereee (talk) 19:31, 17 October 2022 (UTC)

But we have one standard. We don't grandfather in old content and say it is exempt from policy because it is longstanding. Chris Troutman (talk) 19:34, 17 October 2022 (UTC)

Of course not, and the content has been tagged. The question is whether the simple fact it's been tagged is reason enough to remove. Valereee (talk) 19:35, 17 October 2022 (UTC)

BURDEN says it is. Why would we want to retain un-sourced content? Chris Troutman (talk) 19:40, 17 October 2022 (UTC)

See Rowing on the River Thames. Zero inline citations. One very poor source listed. Been that way for fifteen years. I think it's something that needs inline citations, and I marked it that way this morning when I discovered it. However, it doesn't seem to be dubious, the content is undisputed. I think we should retain it, myself. Valereee (talk) 19:45, 17 October 2022 (UTC)

So you posit we should retain the content because it is probably true and could be verified if we looked hard enough? Why not just delete the article? Is that a real loss? Sports articles like these proliferate across Wikipedia because you have fans who are only motivated to see tables of info but have zero interest in being responsible encyclopedists. Shall I send it to AfD? Chris Troutman (talk) 19:58, 17 October 2022 (UTC)

Chris, if you think it's not notable, AfD it. I suspect it's notable but doesn't include inline sources. Valereee (talk) 21:21, 17 October 2022 (UTC)

It's about process. Should content that appears unsourced be removed on sight? Or should an editor at a minimum assess the content? For example, this edit at A-ca-oo-mah-ca-ye cut half of a paragraph because there wasn't a source on the second half. A quick check of the source on the first half showed that the information was contained there. I don't remember why I placed the ref at that point in the graf four years ago (probably because it originally was written in bullet points with a single, general reference supporting things); it clearly should have gone at the end. But just cutting the text because it appeared unsourced without evaluating the material didn't improve things. Checking the source and moving the ref or adding a CN tag would have seemed the more sensible approach to me, and the one that BURDEN suggests over simply cutting. Carter (Tcr25) (talk) 19:59, 17 October 2022 (UTC)

Past discussions of this exact point include [1], [2] [3], [4], [5], [6] (familiar names there, hi!). I think it's fair to summarize the consensus as "if contested, source it or lose it". However, WP:PRESERVE (policy) makes it quite clear that the process is: fix problems if you can, tag or excise them if you can't. That directly contradicts Jayron32's claim that the remover is under no obligation to do anything: if you can fix it, you should. I empathize with Valereee's argument; good content being lost is a shame, but I feel PRESERVE already discourages that. Note that this also applies to tagging, so Valereee's comments that this may disincentivize cn-tagging due to fears of deletion are IMO also already addressed. Going through the uncited-backlog is an essential maintenance task, and if you make hundreds of such edits, you're bound to ruffle a few other editors; I don't support any change to policy that could increase the likelihood of such maintainers being accused of disruption, or ending up at WP:ANI for sorely-needed maintenance. Giving explicit permission to such editors is necessary. The best thing for editors like User:Fourthords to avoid any accusations of excessive deletion is to make it clear in the edit summary that you can't find support for the source; or to move such claims to the talk page if they're plausibly correct (or too technical for you to judge) and don't involve BLPs; it's far more likely someone will eventually come across it in the talk page archives, and hopefully source it and re-add, than if it was lost to the revision history. DFlhb (talk) 20:43, 17 October 2022 (UTC)

I note that no one ever quotes the second sub-section of WP:PRESERVE, which states: Several of our core policies discuss situations when it might be more appropriate to remove information from an article rather than preserve it. Wikipedia:Verifiability discusses handling unsourced and contentious material…. WP:PRESERVE clearly defers to WP:V (and thus BURDEN) in this situation. Blueboar (talk) 21:22, 17 October 2022 (UTC)

WP:PRESERVE seems to be supporting my objection here. It says

Instead of removing content from an article, consider:

Rephrasing or copy-editing to improve grammar or more accurately represent the sources
Formatting or sourcing on the spot
Tagging it as necessary
Correcting inaccuracies, while keeping the rest of the content intact
Merging or moving the content to a more relevant existing article, or splitting the content to an entirely new article
Adding another point of view to the existing points of view to make the article more balanced
Requesting a citation by adding the ^{[citation needed]} tag, or adding any other appropriate cleanup tags to content you cannot fix yourself
Doing a quick search for sources and adding a citation yourself
Repairing a dead link if a new URL for the page or an archive of the old one can be located
Merging the entire article into another article with the original article turned into a redirect as described at performing a merge
Fixing errors in wikitext or formatting

It seems to offer a dozen alternatives to removing content that is noncontentious, nondubious, but doesn't have an inline citation. Valereee (talk) 21:27, 17 October 2022 (UTC)

It's kind of like how the When tagging or removing material for lacking an inline citation, please state your concern that it may not be possible to find a published reliable source, and the material therefore may not be verifiable. If you think the material is verifiable, you are encouraged to provide an inline citation yourself before considering whether to remove or tag it. and the footnoted direction Also, check to see whether the material is sourced to a citation elsewhere on the page. For all these reasons, it is advisable to clearly communicate that you have a considered reason to believe the material in question cannot be verified. portions of BURDEN get lost. Carter (Tcr25) (talk) 21:29, 17 October 2022 (UTC)

I find this inclusionism inappropriate. Yes, we could search for sources and improve articles, but V is a policy and BURDEN clearly does not impose upon us to improve articles. Your argument to retain content ignores the harm done by keeping un-sourced content. Chris Troutman (talk) 21:38, 17 October 2022 (UTC)

This isn't inclusionism. I am not an inclusionist by any means. You can check my AfD statistics; I'm not a major contributor there (fewer than 300 AfD !votes) but I've made ~60 noms. When I see an article I think isn't notable, I nom. I've stripped large sections from multiple articles -- usually 'In Popular Culture' stuff or other trivia or promo or whatever. I get complaints regularly for removing stuff that is trivia. I am absolutely not an inclusionist. My interest here is in preventing us from losing useful content simply because it's been tagged "too long". Valereee (talk) 21:44, 17 October 2022 (UTC)

Here's one example of my non-inclusionism. That was a couple days ago. I do that regularly. Valereee (talk) 21:52, 17 October 2022 (UTC)

Agreed. I might have taken Jayron32 too literally now that I think about it; WP:PRESERVE by my reading doesn't create a requirement to do anything before deletion (which would be impossible for other editors to verify adherence to); I see it as a set of suggestions for serial-deleters to avoid the risk of being seen as disruptive. DFlhb (talk) 00:12, 18 October 2022 (UTC)

You think people peruse the history looking for stuff that was deleted and could be re-added if a source is found? Valereee (talk) 21:28, 17 October 2022 (UTC)

I oppose any watering down of BURDEN and agree with Blueboar, Jayron32, and Chris Troutman. If someone wants text to stay, then they can add a citation, and the person who added it originally should have done so instead of just typing in stuff without a source. If something important was removed, then someone else will come and re-add it in due time with sources if it is truly important. In the meantime, it's not the end of the world if it isn't in Wikipedia, specifically. On the other hand, containing false or misleading information is harmful, and such misinformation can easily propagate, even via WP:Citogenesis, and any modification of BURDEN would make it more difficult for editors to remove it, or discourage them from doing so when they should. Crossroads ^-talk- 23:58, 17 October 2022 (UTC)
Is it watering down BURDEN to expect editors to evaluate the information before removing it? This isn't about a lazy new addition (at least the points where I've encountered the issue), it's with long-standing content that may have been added when the sourcing standards were different or where accumulated edits have broken the connection between a statement and a source. If an editor is actually evaluating the content before removal and making sure the article makes sense after the removal, then I don't have a problem with it. But if they are just looking for a ^[#] at the end of a paragraph or sentence and cutting when they don't find one, then that doesn't actually help suss out misinformation; it just leaves choppy and disjointed articles. A {{CN}} tag does a better job of cluing readers to potentially questionable content and cluing editors to something that should be fixed. Carter (Tcr25) (talk) 00:13, 18 October 2022 (UTC)
This section was started about someone removing material that had a Citation Needed tag already. Some material is unfixable because it is erroneous, or was some rando's opinion from years ago that no one noticed them adding, and the only way to handle such material is removal. Just because content is long-standing doesn't mean it is any good - old hoaxes and other junk is discovered all the time. We have no grandfather clauses. Ideally someone does 'evaluate' before removing, but there is no way to verify that someone did so, and attempting to do so would discourage removal of junk. Such evaluation can be as simple as 'I have familiarity with this topic and this strikes me as implausible'. They have every right to do so. The person(s) who think that text should be included should have cited a source to ensure its sustainability - "anyone can edit", not "anyone can write"; it cuts both ways. Crossroads ^-talk- 00:20, 18 October 2022 (UTC)
Agreed, there are times that material should be removed, but the cases that I'm thinking of that I've seen heavily defended by BURDEN have mostly been cases where the source was there, just not at the end of the paragraph or it was there in a general reference but not inline (as was common practice in the past). That no work was done to assess it was clear because a quick check verified the material ... in at least one case by the citation on the previous sentence. There's a right to remove content, but what I'm looking to see better stated is the responsibility to do so in a considered manner that improves the project. Carter (Tcr25) (talk) 00:53, 18 October 2022 (UTC)
It's bad-faith to assume that editors are not doing any work to assess whether material can be verified before removing it, but I'll assume you meant to say that you have concerns that that may be occurring. If you're aware that a statement could be verified by a citation used for the previous sentence (or elsewhere in the article for that matter), then it's trivial work to reinstate the sentence prior to the citation that verifies it or duplicate the citation. However, it seems unrealistic to me to expect editors cleaning up articles to review every citation already included within an article to ensure that none of them verify information currently lacking citations, especially if that information has been tagged for needing a citation long-term. DonIago (talk) 01:36, 18 October 2022 (UTC)
Apologies if that came across as bad faith. It could also be seen as bad faith to assume that a seemingly unsourced comment is due to a lazy editor (and not evolving standards or a poorly made intermediate edit). Yes, in some instances it's a trivial task to restore and place a citation when it's caught at the time of removal. When they're done on a large scale or a low-traffic article, that may not be the case. While I would love it if someone did actually look at deletions (and additions) wholistically, that may be unrealistic. I don't think, however, it's unrealistic to expect at least an attempt to check verifiability before deletion of material that isn't a BLP violation or obvious bunk. Carter (Tcr25) (talk) 01:48, 18 October 2022 (UTC)
It absolutely is not bad faith to assume that editors are not doing any work to assess whether material can be verified before removing it, because they don't even have to in many cases. When I first got unblocked, I used to do a lot of vandalism work to "prove myself", and I can assure you that most unsourced content that is added is flagged as potential vandalism, and so the whole anti-vandalism system including the warning templates and tools such as twinkle and redwarn actually encourage this lazy behaviour. Only editors highly conscious of what they are doing will realize this, and do any work to assess to material. Otherwise, they will simply follow what they see is recommended in the tools and templates to other editors as I made the mistake of doing my own self before I woke up and started thinking for myself. I think @Doniago is the one who owes @Tcr25 the apology for trying to make them feel guilty about something they have every right to think and feel. Huggums537 (talk) 15:14, 18 October 2022 (UTC)
Are you saying you don't believe that assuming editors are doing no work to assess whether material can be verified before removing it is a bad-faith assumption?

We can also talk about your own bad-faith assumption that I was trying to make Tcr feel guilty, if you'd like, especially given that I immediately followed up by saying that I was assuming Tcr had simply misspoken. DonIago (talk) 15:29, 18 October 2022 (UTC)
Sure, we can talk about the fact that I didn't know what your exact intentions were, and maybe it wasn't to make Tcr feel guilty, but it still would not change the fact that whatever those intentions were, they were wrong no matter if was making Tcr feel guilty or some other well meaning intention that paves the road to hell. Tcr certainly didn't seem to take it you meant they had simply misspoken by virtue of the fact they apologized for the bad faith part rather than the misspeaking part. ~~and if you weren't trying to make Tcr feel guilty, then you most certainly tried to do same to me because I really see no other good intentions at all behind such a comment.~~ Huggums537 (talk) 15:55, 18 October 2022 (UTC)

Also, to answer your first question more succinctly, I believe what my previous comment already states is that I know (not believe) that assuming editors are doing no work to assess whether material can be verified before removing it is not a bad-faith assumption because I have been guilty of doing it myself in the past by blindly following tools, and what I see templates encourage other editors to do. If it can happen to an editor of average intelligence as myself, it can happen to anyone. Huggums537 (talk) 16:13, 18 October 2022 (UTC)

@Doniago I do acknowledge your call out of my bad faith assumption in spite of my comment above. It is hypocritical of me to require an apology from you and then refuse one when called out on my own stuff. I was wrong to assume your intentions, and should have just stuck to the fact that I thought your view was not right. I will also be stricking the last part of my comment. Huggums537 (talk) 16:30, 18 October 2022 (UTC)

I'd settle for at minimum requiring an edit summary that explains that the reason they're removing it is because it's been tagged cn for X years, or whether they tried to verify and failed. That would at least give other editors a clue as to whether that work has already been done. Valereee (talk) 15:53, 18 October 2022 (UTC)

I don't think a clarification is equivalent to watering down. The goal here should be improving the encyclopedia, which ideally means removing anything dubious or unverifiable and adding an inline citation for anything that doesn't have one. Assuming something is unsourced because it doesn't have an inline citation (especially for anything created before we started requiring inline citations for all new content) doesn't account for the fact that it was common to just list all references. Citation needed doesn't necessarily mean unsourced. It just means we need to add an inline citation. Valereee (talk) 14:52, 18 October 2022 (UTC)
I think there's two things to consider here. First, policy is clear that all material that is challenged or is likely to be challenged needs a citation. Second, this wording makes it clear that we are not supposed to universally remove all uncited content on sight. It's perfectly fine to challenge a lot of content; and I don't think we should add specific restrictions about what a "valid" challenge is. It's not required that the person removing it do a source-search first, or even that they believe no source exists; and setting those requirements would be a mistake, since it's important to ultimately maintain pressure towards WP:V for anything likely to be challenged. But I think the policy implies that you do need some reason to challenge it beyond "there's no source" (otherwise it would say all material requires a source) - I don't think it is normally necessary to articulate it; given the weakness of an unsourced statement, even a gut feeling is enough per Hitchens's razor, so I think it would be a mistake to add barriers to challenging a source. But if an editor is plainly taking the position that absolutely all material without citation must be removed on sight and is challenging and then removing literally every unsourced statement they see based solely on the fact that it is uncited, to the point where it's clear they have no other specific objections or doubts at all, then that plainly goes against policy. --Aquillion (talk) 15:41, 18 October 2022 (UTC)
When information has been tagged as needing a citation for a reasonable amount of time, I think it is logical to assume that a citation can not be provided… and to remove the material.

Now, the remover may be wrong in that assumption… but, if so, the burden to demonstrate that the assumption was wrong falls on those who want the material to remain in the article, not the remover. Blueboar (talk) 16:12, 18 October 2022 (UTC)

I don't actually think we can assume that, no. There are millions of articles created before we required inline citations. Some of them are barely watched. Some of them may not have been edited since the cn tag was placed. We can't assume anyone with both an interest in that article and the knowledge of how to fix a cn tag has even seen it. Valereee (talk) 17:28, 18 October 2022 (UTC)

Someone had to be watching the article, otherwise no one would be complaining that the tagged information had been removed. Blueboar (talk) 17:53, 18 October 2022 (UTC)

Fair point, although the flip side is that any articles that aren't being watched wouldn't get the complaints. :D Valereee (talk) 18:31, 18 October 2022 (UTC)

To put this another way… always ask: “how can I improve this article?”… rather than asking “how can I make the other guy improve this article?”. I can control what I do, I can not control what the other guy does. If the other guy removes uncited info that I think belongs in the article, it is my job to improve the article by returning it with a citation… it is not my job to chastise the other guy for removing it. Blueboar (talk) 16:50, 18 October 2022 (UTC)

As someone who works on scientific and technical topics, I find discussions like this positively frightening. A large part of Wikipedia's content in these areas was added at a time when references were either not thought necessary or not necessary for material covered in textbooks. (That did turn out to be a problem, particularly in physics articles with amateurs adding their own theories.) When references began to be required the feeling was that existing unreferenced material should only be removed if the removing editor thought it was not accurate. Now the feeling seems to be shifting to removing everything that is not well referenced or has had a citation needed tag for a while. This would be devistating for the part of Wikipedia that serves as a technical and scientific encyclopedia. We do not have enough editors with the expertise or time to watch and defend all these articles from the removal of accurate material that is just unreferenced. StarryGrandma (talk) 20:22, 18 October 2022 (UTC)

@StarryGrandma: Talk about frightening! How can one claim to have a scientific background but then say that we can just trust statements are correct even if they're not referenced? When you say you worry about having to "defend all these articles from the removal of accurate material that is just unreferenced" gives ever impression of ladder-kicking behavior: old timers wrote what they liked and they still intend to get away with it decades later, even while newer editors are prohibited from doing same. Maybe this is the editor retention issue? Chris Troutman (talk) 20:28, 18 October 2022 (UTC)

It sounds like you do not trust our existing scientific articles - many of which were written at a time when graduate students were more common on the internet than the general public and knew what they were doing and kept an eye on each other. Why do you think so badly of the early editors? There were plenty of discussions. Once the general public became involved physics articles ran into personal theories of relativity, etc. and from then on sources, and reliable sources, came in with new material. And material that was "likely to be challenged" got removed over the years. Sources have been a requirement for at least a decade. Now we have discussion about removing any material without an inline source, no matter the history of the article. StarryGrandma (talk) 21:02, 18 October 2022 (UTC)

How are editors who may not have the same level of technical or scientific familiarity as the editors who wrote an article supposed to know that material is accurate if it's unreferenced? If anything, I think it could be argued that the more uncommon knowledge is likely to be, the more urgent it is that references be provided. DonIago (talk) 20:29, 18 October 2022 (UTC)

It was problems with scientific articles that pushed Wikipedia to be much more careful about requiring sources for material. But someone who is unfamiliar with a topic should not be removing longstanding material from an article just because it was written before references were required. Only remove existing material if you know it is not accurate. People do watch RecentChanges to see if incoming material is sourced these days. StarryGrandma (talk) 20:47, 18 October 2022 (UTC)

You certainly have the right to feel that editors shouldn't remove material unless they know it is inaccurate, but current policy does not restrict the removal of unsourced material in this manner. DonIago (talk) 20:52, 18 October 2022 (UTC)

Not just current policy, but policy for at least a decade. But no one has promoted removing all material without an inline source until now. And the current policy exempts material that is not likely to be challenged from its list of what must have an inline source. StarryGrandma (talk) 21:11, 18 October 2022 (UTC)

I was around in those days… Part of the reason why we agreed to have an “unwritten exemption” for older articles was to give editors TIME to fix them (by adding the citations we required for newer articles)… that was 10 years ago. We have had more than enough time to locate and add citations. It is now time to lift that “unwritten exemption”, and to be a bit harsher when it comes to older articles that need proper sourcing. Blueboar (talk) 21:32, 18 October 2022 (UTC)

So what's the best way to achieve that goal of proper sourcing? Wholesale cutting of material in hopes that it prompts someone to restore it with citations? Tagging content that lacks sources? Engaging with WikiProjects that cover the subject matter? Use the talk page to ping editors who've contributed to the article in the past? You made the statement earlier always ask: “how can I improve this article?”… rather than asking “how can I make the other guy improve this article?”. Is cutting material that you don't question the veracity of an improvement? Again, I don't have problems with people cutting material they honestly believe cannot be verified, so long as that's made clear in the edit summary and they do take the time to assess the content. That some people see the policy as Notice, the person who removes the material is under no obligation to even do a simple Google search. They don't need to do anything. throws BURDEN out of balance. I think part of assuming good faith should imply doing that simple Google search even for recent additions, but I think it should be clearly required for removal of long-standing content.

The presence of a ^[#] isn't a guarantee of accuracy; nor is its omission a certainty of misinformation. If editors adopt a policy of cutting seemingly unsourced content because there isn't a proximate footnote, it will remove informative content that should be retained under PRESERVE. That doesn't make Wikipedia better or more reliable. If it's not done with care, it can and does leave disjointed text, broken statements, and poor quality articles. I'm not asking the other guy to improve the article; I'm asking the person looking to cut content to ask themself if making that cut actually does improve the article or if there is a better way to handle it. Carter (Tcr25) (talk) 13:27, 19 October 2022 (UTC)

@Valereee, I think these conversations usually end up in one of two places:

An individual is being WP:POINTY, which is a behavioral problem, which needs to be handled elsewhere.
A proposal that BURDEN specifically require people blanking unsourced content to say that they have a reasonable belief that it would be impossible for anyone to find a reliable source for the material they're blanking. (The enforcement mechanism here is "Then you need to self-revert, because Wikipedia isn't a game of 'Simon says add the sources right now or I get to wreck the article'", rather than blocking.)

The policy has said for years that "In some cases, editors may object if you remove material without giving them time to provide references", but this doesn't seem to be actionable, because the handful of editors who do this routinely shrug and say "How about that: Editors can object all they want, but I still have an absolute, unfettered right to blank anything not provided with a citation as fast as I think it should have been, and I frankly don't care who 'may object'." WhatamIdoing (talk) 05:04, 19 October 2022 (UTC)

That is your interpretation. It is true that "I still have an absolute, unfettered right to blank anything not provided with a citation as fast as I think it should have been, and I frankly don't care who 'may object'". If you are serious about writing an encyclopedia, then citing sources should be no problem Fact is, perhaps half of us aren't serious about it. They like using Wikipedia to say what they like, sources be damned. Is that you, WhatamIdoing? Chris Troutman (talk) 20:38, 19 October 2022 (UTC)

This is an extremely uncharitable straw man statement. Mass blanking is disruptive if there is no substantive rationale provided. There are many justifications and qualifications required. Wikipedia is a collaborative project. If you blank a section that is tagged with a cite needed tag and I revert saying I will get the book from the library tomorrow, are you saying that you will then revert back? Because that feels uncollaborative. Andre🚐 20:50, 19 October 2022 (UTC)

If you'll get the book tomorrow, why can't you wait to revert until tomorrow? DonIago (talk) 21:04, 19 October 2022 (UTC)

That depends, but I've seen situations where people come through and start plowing through removing material which then drastically changes the article, and then the article could go through additional changes since then. Since we allow material that is "verifiable" but not necessarily "verified," there might be a situation where a reasonable delay is reasonable, and I believe the policy might even explicitly allow for that somewhere. Andre🚐 21:09, 19 October 2022 (UTC)

In all my time here I don't believe I've ever encountered a scenario where I removed material only for someone to revert it and say, "I'll add a source tomorrow" rather than just coming back and reinserting the material with a source (though not-infrequently an unreliable one, at least the first time around). It's certainly possible that I've tended to focus on lower-traffic articles, but I also don't really feel I've seen many situations where an article was so active that a reinsert the next day would be unwieldy or especially inconvenient, given that article history is readily available. If anything, I'd be more concerned that an article would be active enough that after the uncited information was re-added subsequent edits would bury the addition. I don't think it should be incumbent on editors concerned about unsourced material to have to set a reminder for themselves to ensure that a citation was ultimately added to material they'd already previously challenged. All in all, this seems like a scenario where a note at the article's Talk page might be the most practical approach. DonIago (talk) 02:58, 20 October 2022 (UTC)

The scenario is hypothetical, but I disagree with the idea that these situations don't come up. There have been several in this discussion and I've encountered them over the years. The point is that it's a give and a take. Folks familiar with the topic and willing to do a little legwork should work on articles that need cites. We shouldn't be mass removing cn tags that are "too old." Andre🚐 14:29, 20 October 2022 (UTC)

@WhatamIdoing, in the case of the removals by the editor who got me here, not I don't think it's actually pointy. Or at least not intentionally so. I think this editor sincerely believes that 'X years is long enough' and when they see something that has been tagged for however long, they remove it. My concern is more that if the content isn't objectionable in any other way, what's our hurry? If anyone reading it thinks, 'sounds about right', why remove it simply for having been tagged "too long". Yes, I do think people should be providing some reason for removal of noncontentious content other than "too slow." Valereee (talk) 20:55, 19 October 2022 (UTC)

I agree, WP:NODEADLINE, WP:NOTPAPER, WP:AGF. Andre🚐 20:56, 19 October 2022 (UTC)

@Andrevan: I'm fine waiting for someone to get a cite although they could re-add the content when they finally get the cite. The problem I find in articles like Rowing on the River Thames is an assertion that cites aren't needed. Maybe this mindset tracks to an earlier era on wiki but it is wrong now. I've written 30-odd articles and I never had a problem sticking to sources. Chris Troutman (talk) 20:58, 19 October 2022 (UTC)

I think that's an unusual editor, one who edits quite frequently but maybe hasn't been much involved in or aware of policy changes. See their editing history. They seem to be unusually upset over being asked for citations. Valereee (talk) 21:03, 19 October 2022 (UTC)

Yes, that seems to be an entirely different situation. It's true that in 2003 or whatever year, there was no hard sourcing requirement, but those days are obviously long gone. However, there are definitely still some articles from the old days that lack inline sources. For example I would write articles 10 years ago with a book or two, on my lap and then just write them as citations in the references section. Andre🚐 21:15, 19 October 2022 (UTC)

Ok, so we had different expectations 10 or 15 years ago… What have you been doing since then? You have had more than enough time to convert your references into in-line citations. Blueboar (talk) 22:01, 19 October 2022 (UTC)

I have gone through my old articles and updated them but there's no hard deadline AFAIK. Andre🚐 22:06, 19 October 2022 (UTC)

Ok, but my point is that surely 5 to 10 years is a reasonable soft deadline for such a conversion… or for fixing a cn tag. There does come a point when we can say… “right, enough time has passed… I’m forcing the issue”. Blueboar (talk) 23:52, 19 October 2022 (UTC)

I think the point is you need a valid reason. The point of the referencing requirement is that statements "likely to be challenged," at least I believe that's what the language used to say, should be removed if no reference is provided. If you're just challenging statements because they've been tagged for a long time, that doesn't really provide a valid rationale for removing the statement, if there's reasonable reason to believe a citation will be or can be provided for the statement and the statement itself is not in doubt. If the statement is controversial, that's not what I'm referring to here. Andre🚐 00:23, 20 October 2022 (UTC)

If a statement was tagged, then clearly at some point someone had concerns about it, and tagging is in and of itself a challenge as per WP:MINREF. It could be argued that an editor removing tagged statements is merely performing a maintenance task on a statement that was challenged and subsequently never cited (though circumstances may vary). I don't think it's reasonable to believe that a statement tagged for years should be left in an article because, "maybe someone will provide a citation tomorrow..." when it's already been in the article for hundreds of tomorrows.

TL;DR it rather sounds as though your beef is with the tagging editor rather than the removing editor. DonIago (talk) 03:07, 20 October 2022 (UTC)

No, I don't agree, because my view is basically eventualism. If we think a statement is unverifiable or it seems dubious and it's not sourced, it should be removed. If it seems possibly like a verifiable statement and there's nothing about it that seems problematic or controversial, it can be tagged with a cn for an indeterminate period of time. I don't have a beef per se, but if I had one, it would be with someone deleting statements which seem verifiable without trying to verify them or research them. Andre🚐 03:13, 20 October 2022 (UTC)

I don't think many editors are going to agree with your apparent viewpoint that statements that have been tagged for needing citations should be allowed to remain in articles in that state because they are possibly verifiable. One could argue that any statement is possibly verifiable, and the very fact that someone tagged the statement in the first place, unless you want to assume bad-faith, indicates that they felt the statement, without citation, was problematic. DonIago (talk) 05:35, 20 October 2022 (UTC)

There was a recent situation where someone was mass blanking technical articles about radio antennas. There were statements that were basically SKYISBLUE to a knowledgeable person but were tagged needing references. References were easily obtained but that individual didn't even try a BEFORE. It's not bad faith but it was in my opinion rash and wrong. Andre🚐 14:28, 20 October 2022 (UTC)

@Doniago, I think that a few editors are tagging and blanking statements not because they think the individual statements have any problems at all, but because they believe it is always and inherently problematic for Wikipedia articles to contain any paragraphs (outside the lead) that do not contain an inline citation. WhatamIdoing (talk) 16:43, 20 October 2022 (UTC)

I agree. I tag stuff all the time that I have zero concerns about. If I had an actual concerns -- thought the statement was dubious for instance -- I'd at minimum tag it as being dubious, but if it was dubious enough I'd remove it with an explanation that I thought it was dubious. To be clear, I have zero objection to anyone tagging or removing something they find dubious. Valereee (talk) 16:52, 20 October 2022 (UTC)

In the radio case Andre mentioned above, the editor was placing a significant number of CN tags across the article, waiting 30 days, then blanking anything that hadn't had a source added because enough time/warning had been given. (The removed material was placed in a collapsed section of the talk page.) Things got further complicated and POINTY when editors reverted the blanking to begin adding sources but BURDEN objections were made that restoration could only be made bit by bit as sources were added. (It's similar to the restore with a I will get the book from the library tomorrow comment.) There was some SKYISBLUE pushback involved too, but based on BURDEN the demand was for everything to be sourced in full. In the end, all the requested (and more) sources were added, but only after the blanking was reverted. Carter (Tcr25) (talk) 17:22, 20 October 2022 (UTC)

Another fact that does need to be taken into account besides just years is number of views and edits during that time. A cn tag placed 10 years ago, but where the article has only seen < 1000 views in that time and practical no edits is not the type of tag to remove. A 10-yr old cn tag on an article that gets >100,000 views a year with thousands of edits each year, on the other hand, is prime for removal. Masem (t) 00:23, 20 October 2022 (UTC)

Is it? Maybe a hundred editors saw it and thought "Oh, my, someone must be a real idiot to think that actually needs an inline citation. Reminds me of the time that IP wanted a source for the number of digits on the human hand. Probably just a kid who didn't know what a digit is in anatomy. Still, if someone feels like adding a citation there, I won't stop them..." WhatamIdoing (talk) 16:49, 20 October 2022 (UTC)

But if a hundred editors saw it and thought that, wouldn't at least one of them just remove the tag? Valereee (talk) 17:03, 20 October 2022 (UTC)

If you take the tag off, then Wikipedia:Citation Hunt can't see it. It falls off lists that editors use to find statements where citations were requested (example). You'd only want to remove such a tag if you think that there really shouldn't be a citation there. WhatamIdoing (talk) 17:29, 21 October 2022 (UTC)

Exactly, and a cn for 'the human hand has five digits' should be removed, surely? Valereee (talk) 17:42, 21 October 2022 (UTC)

One would hope so, but I've been told SKYISBLUE is an essay and WP:V as policy outranks an essay. Because WP:V says any material whose verifiability has been challenged or is likely to be challenged, must include an inline citation to a reliable source that directly supports the material, so if someone requests a citation for "the human hand has five digits" then that's considered a challenge and it must be cited. It seems ridiculous, but under the letter of WP:V that's what's demanded no-matter the spirit of the policy. Carter (Tcr25) (talk) 18:23, 21 October 2022 (UTC)

A challenge itself has to be reasonable, rational, and based in policy. You can't challenge something because you don't like it. The rules are principles so you should not wikilawyer the letter of a policy, because WP:5P ask you to use good judgment and WP:CIR as well. Andre🚐 18:48, 21 October 2022 (UTC)

"A challenge itself has to be reasonable, rational, and based in policy" is not in this policy. We tend to forget that IAR is also a co-equal policy. Also, we'd want editors to make challenges based in common sense, not just policy. WhatamIdoing (talk) 19:46, 21 October 2022 (UTC)

Exactly. What WP:V says is Any material lacking an inline citation to a reliable source that directly supports the material may be removed and should not be restored without an inline citation to a reliable source. If people were applying reason and rationality to some of these actions, this wouldn't be a problem, but that doesn't appear to be happening. Carter (Tcr25) (talk) 20:04, 21 October 2022 (UTC)

Well, I think it's implied that every policy should be interpreted in good faith with logic and common sense. But maybe we should modify this policy to say that the "challenge" must have a reasonable basis. Andre🚐 20:18, 21 October 2022 (UTC)

I agree, page views and number of edits made since the tag was added is important. If an editor is doing rote maintenance, they may be visiting articles that get 3 pageviews a day and 3 edits a year. So an article which was created using sources but without inline citations might get a maintenance visit from an editor who while doing other maintenance removes anything with a cn tag.

I'm not objecting to that maintenance editor telling anyone who complains, "then fix it per BURDEN". What I'm concerned about is that while that editor might get a dozen complains and generate 100 fixes, there might be 1000 articles where no one even notices because the article is basically unwatched. The creator may have stopped editing before the cn tag was even added. The complaints are a symptom of a problem, not a problem in and of themselves. Valereee (talk) 17:02, 20 October 2022 (UTC)

'X years is long enough', and we have at least one editor who says '30 days is long enough'. I've seen that editor remove uncited material such as "For simplicity, this section describes a basic type..." – something that can't be cited, unless you think you can find sources that tell you what you're going to include in a section of a specific Wikipedia article. I think that edit, which blanked more than half the article, also removed about 20 images, which are not normally required to have citations anyway.

But basically we end up in the same place: either we have an unlimited right to remove anything and everything that is unsourced, even when we personally know the content to be accurate and verifiable, or we don't. So far, BURDEN says we have the right to blank unsourced content. As you can see, there are costs to this policy, but this is the policy. WhatamIdoing (talk) 00:40, 20 October 2022 (UTC)

I don't see it as an absolute right: I think it's contingent on having a good reason. Andre🚐 00:50, 20 October 2022 (UTC)

So far, we have not been able to get the necessary level of agreement to say that unsourced content can be tagged but not blanked if you are unwilling to say that you genuinely believe no reliable source contains this content. Until that happens, you are allowed to blank statements like "Water is wet at room temperature" and "Human hands normally have five digits" unless they are cited. WhatamIdoing (talk) 16:19, 20 October 2022 (UTC)

A lot of people here are making a faulty argument "If you want your edit to stay in wikipedia YOU should provide the best sources". This is in direct contradition with WP:NPOV which state that you should provide the best coverage of ALL the relevant points of view and general facts. What if I do not want ? WP:NPOV Do not say : "You are allowed to avoid talking about anything you don't like so long as no one else bring a source". Removing something in an article for which an NPOV-compliant source is accessible should be considered vandalism. That implies that the remover of a passage should be knowledgable enough on the subjet to recognize an WP:OR or know to whom to attribute the WP:fringe. Everyone do mistakes but, there should be concequences to editors who regularly remove verifiable material. The same way a pusher should expect to eventually get in trouble and for the same reasons. Iluvalar (talk) 09:13, 20 October 2022 (UTC)
That's the point that @Andrevan was making earlier. Imagine that you discover a disastrous article about some pseudoscientific product. Maybe it claims that their particular brand of juice cures cancer – a truly WP:EXTRAORDINARY claim.

What can you do quickly? You can slap some maintenance tags on it. You can beg for help from WP:FTN or WP:MED. You could also add a sentence like "This product has not been scientifically proven to have any medical benefits for cancer patients", because you know that no juice product has ever cured any kind of cancer, even if you don't happen to have a source at hand about this.

There isn't an editor of this page who would tell you that you are doing something serious wrong by adding that. Editors who focus on NPOV issues would cheer you on, because your uncited addition moves the article closer to a neutral point of view.

But CHALLENGE says that this addition can be removed by the touts, and that you can't stick it back in until you add a source. Removing it is permissible under CHALLENGE but would violate NPOV.

Of course, it cuts both ways: If the tout starts spamming uncited sentences into other articles to say things like "This kind of cancer can be cured by Our Brand", then you can remove those per CHALLENGE, and they can't stick them back in without a source. But removing their content wouldn't violate NPOV, whereas removing yours would. WhatamIdoing (talk) 16:31, 20 October 2022 (UTC)

As a general rule, WP:DUE weight is relative, and this is especially true for fringe topics per WP:PARITY. So if someone removes a qualifying statement you added, and you feel the qualified statement is not NPOV without it, one solution is to remove the whole thing. Obviously if the statement you were qualifying was itself unsourced then this is common sense. When you were qualifying a sourced statement using an unsourced one, though, it gets a lot trickier, especially if there's at least some degree of consensus behind the statement you want to qualify - you have to examine the source and argue that there's something wrong or WP:FRINGEy about it, or provide something rebutting it. Fortunately boards like WP:FRINGEN or WP:NPOVN exist if you think that there's a glaring problem that needs immediate resolution and don't have the time to do it yourself. But Cromwell's rule also applies to an extent; some things do genuinely require time and effort to resolve and we can only shortcut this so far, no matter how clear-cut it might seem to one of the parties involved. --Aquillion (talk) 19:52, 20 October 2022 (UTC)

comment: I think one of the biggest problems with this policy is that it appears to apply to uncited content more than anything else to a grossly disproportionate degree, and so you've ended up just repeating the same discussions over again. When you go through and think about the responsibility that is required to remove [unsupported] cited content, and compare that to BURDEN, then you find that BURDEN takes on a whole new meaning that is rather lacking in many respects as you consider how it would apply to [unsupported] cited content when you read it. I think there is much responsibility placed on editors to cite added material, and also to do the proper work to be responsible in removing material that is already cited, but [unsupported]. So, to me if editing is this much responsibility for both adding and removing, I see no valid argument for why that should be any different just because the source isn't in one particular spot located at the end of a sentence called an inline citation. I've heard the argument that adding it at that spot is "easier" than asking others to take the responsibility to do the work that would normally be required of us anyway, suggesting it isn't fair, and somehow must be "harder", but the job is the exact same no matter who does it if we are talking about articles that old because whoever added the content is probably long gone. More importantly, they forget that one person can delete several times the amount of uncited content in the same time it takes for another editor to properly cite and restore just one item they've deleted. So, who is really asking who to do more work, and who has the right to cry "no fair"? I think the biggest improvements to BURDEN would be to fix the areas where it seems to only talk about uncited content, and update all of them in such a way where the whole section is able to talk about and cover any challenged material no matter if it's cited or not. This might get a lot out on the table so we can finally put it all to bed. Huggums537 (talk) 21:01, 20 October 2022 (UTC)
Note: I could not really find any policy/guidance anywhere on removing cited content that is unverifiable/inaccurate. I really do think this a huge amount of the problem. If we could clarify the removal of all content such as it seems to do in a general way at essay: WP:REMOVAL rather than an obsessed focus on uncited content at BURDEN it would solve a lot. Does anyone else see this as problematic? REMOVAL has only a small section dedicated to unsourced content while our freakin' core policy main focus is almost exclusively uncited content to the near exclusion of everything else. Wtf? Huggums537 (talk) 21:26, 20 October 2022 (UTC)

If something is cited, if the citation supports the content, then the content is by definition verifiable. Since this is a page about verifiability, of course it focuses on whether content is verifiable or not. There are other policies which deal with others sorts of problems. If you'd like to propose a new single policy that draws together all of the potential issues that might warrant removal of content, go ahead - but this isn't it. Nikkimaria (talk) 03:16, 21 October 2022 (UTC)
Either you have misunderstood, or I have not explained clearly. Read my note above more carefully. I'm not talking about cited material that is supported by references, I'm talking about unverifiable/inaccurate/unsupported material that has been cited, but not only that, I'm also talking about the unverified material (mentioned in bullet point #2). This policy totally overlooks defining that "unverified material" could just as easily mean material that isn't supported by the reference as it does material completely lacking one. However, that is just one glaring example. If you read my post again with this in mind, you might begin to see that is only the tip of the iceberg. Huggums537 (talk) 15:09, 21 October 2022 (UTC)

To illustrate further, one could go through BURDEN and replace all of the examples of "lacking an inline citation", "not having an inline citation", or "unsourced" with other language such as "lacking a supporting inline citation", "not having a supporting inline citation", or even follow in the footsteps of bullet point #4 about the removal of contentious material, and suggest "unsourced" or poorly/improperly sourced. This would be a simple proposal that would be an easy fix so this is it right here. Huggums537 (talk) 16:07, 21 October 2022 (UTC)
In other words, I completely agree with your assessment that this is a page about verifiability, so of course it should focus on whether content is verifiable or not, but in the current form, BURDEN is only focusing on whether something is cited or not. That is the big difference that @WhatamIdoing tried to bring up in this discussion. If the section is glossing over the fact that "unverified material" could just as easily mean material that isn't supported by the reference as it does material completely lacking one, then it isn't focusing on verified/unverified material, but just focusing on the fact it is uncited, and that is not what the page is supposed to be about according to what you yourself a much more experienced editor have told me. Huggums537 (talk) 16:44, 21 October 2022 (UTC)
BURDEN is limited to uncited material to keep it from becoming too strong. For example, there's an editor on one of the COVID pages who has been arguing recently that a source that actually says "We believe it is safe" (this is a direct quotation from the source) should be cited for a sentence the editor wants to add claiming that it's not safe.

BURDEN requires one (1) source that you (=the adding editor) genuinely believe to be reliable and you genuinely believe to support the claim. That's it. Once that hurdle has been cleared, we move on to real editing, which includes full consideration of all aspects of verifiability, neutrality, good writing, pertinence to the subject, etc.
So think about that scenario:
Alice adds "It's safe" to the article with no source.

Bob removes it per BURDEN.

Alice restores it with a source that says "It is safe".
Bob might be able to fuss about whether it "is" safe or if it's more accurate to say that Prof. I.M. Portant "believes it is safe", but overall that's okay, right? Now:
Bob adds "It's not safe" to the article with no source.

Alice removes it per BURDEN.

Bob restores it with a source that says "It is safe".

Alice removes it per WP:V
If BURDEN's strict mechanical rules were applicable to cited material, we'd see this:
Alice adds "It's safe" to the article with a source that says "It is safe".

Bob removes it per BURDEN, claiming that when the source says "It is safe" they actually meant "It's not really safe, but we'll lose our jobs if we tell the truth".

Alice doesn't know what to do, because now BURDEN apparently requires her to find a source that satisfies Bob, which can't be done.
Limiting BURDEN to uncited material avoids a lot of needless drama and limits POV pushers. WhatamIdoing (talk) 20:07, 21 October 2022 (UTC)
Okay, I see your meaning, but then BURDEN seems to be more confused than I am at this point since we do have the bullet point #2 which does suggest it already applies to cited material whose verifiability has been challenged since it does not specify if it talking about either cited or uncited material. Then, we also have the contentious blp part that clearly says even poorly cited material may be removed so it seems to me there really isn't any limit except for the sake of talking about if a citation is present on some material or not. Huggums537 (talk) 20:41, 21 October 2022 (UTC)
Are you thinking of the beginning, which says this?

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.

It ...kind of is correct. It might be clearer if we wrote something like this, though:

All content must be verifiable. Additionally, certain material is required to be supported by an inline citation. The burden to demonstrate verifiability lies with the editor who adds or restores material, and it this burden is satisfied by providing an inline citation to a reliable source that directly supports the contribution even if this source is not ultimately accepted as fully demonstrating verifiability. After that first source has been provided, it is every editor's equal duty to determine whether the material is both verifiable and also otherwise suitable for inclusion in that article, and to remove it if it is not.

It's kind of wordy, though. WhatamIdoing (talk) 00:16, 22 October 2022 (UTC)
"Certain material", meaning 99.99% of our content or so. Seems like misplaced emphasis. Crossroads ^-talk- 00:20, 22 October 2022 (UTC)
Exactly how much content needs an inline citation depends on the subject matter. I would expect fewer inline citations in an article about an album, and more in an article about COVID. WhatamIdoing (talk) 14:50, 22 October 2022 (UTC)

I was actually meaning the more simple bullet point #2, but now that you mention it, I think what I was talking about could apply to what you are saying though. Also, I don't think "Certain material" is misplaced emphasis or that it means 99.99% of our content "or so". I think it refers to the four bullet points which is the certain material required to have an inline citation, and I think it is very helpful to clarify this. Huggums537 (talk) 03:21, 22 October 2022 (UTC)

There's a simple solution to this which would also solve many other problems. As background, the policy basis for such a removal is that the removal is implicitly a challenge. We simply have to say that such a challenge must include a statement of concern that the material may not be verifiable. Few people are going to want to look stupid by saying that they have a concern that a sky is blue type statement may not be verifiable. If there is interest in exploring this, pleas ping me and I'll make a specific proposal. Sincerely, North8000 (talk) 18:13, 21 October 2022 (UTC)

Even simpler solution: treat the removal itself as a valid challenge, and act accordingly. Blueboar (talk) 18:32, 21 October 2022 (UTC)

@Blueboar, that doesn't really sound like a solution, but just more like a suck it up and accept the flawed version the way it is by finding ways to rationalize it. I mean, I guess that technically is a solution, but is it a good one? Huggums537 (talk) 18:49, 21 October 2022 (UTC)

It's not a flawed version. Removing uncited content is working exactly as envisioned. It's a feature, not a bug, of this policy. Look, try to imagine that the removed text was never in the article to begin with. Like, let's just put aside that it was removed, and lets say what would happen if the bit of text in question never actually existed. What would you do if you were properly to add the information to an article where the information was missing? You'd add it, and properly cite your source. Removing uncited text presents no undue burden on anyone, because if the text had never existed it would have required the same amount of work to properly add new information. --Jayron 32 18:54, 21 October 2022 (UTC)

I think part of the problem is that we have dumbed down the policy in many ways in an attempt to make it foolproof for any editor to participate, but this has backfired on us in a major way because many editors come to rely on it too much, and don't do enough thinking for themselves anymore. I think we should start to trust editors more. Most of them are not as dumb as some of us think they are. Your comment focuses on ~~uncited~~ content [being cited or not]. That is my main gripe about the major flaw of the policy in my comment and "note" above. Huggums537 (talk) 19:39, 21 October 2022 (UTC)

I think this is a good idea in addition to the one I mentioned above too, and I believe @Valereee mentioned something about it before. However, I think my change is far easier to implement, but maybe not as effective in dealing with the problem as a well as this idea does. Huggums537 (talk) 18:39, 21 October 2022 (UTC)

@North8000 what kind of proposal did you have in mind? Huggums537 (talk) 19:53, 21 October 2022 (UTC)

North has frequently suggested adding a sentence to this policy along the lines of "a challenge must include a statement of concern that the material may not be verifiable". In other words, blanking is a CHALLENGE and BURDEN applies if you say "I don't think any sources say this anywhere". Blanking would not be a CHALLENGE and BURDEN would not apply if you just blanked the material (e.g., with no explanation).

Another approach would be to follow the Wikipedia:Deletion is not cleanup model, and say that it's not okay to systematically tag and blank material for the purpose of rushing other editors into providing citations, on threat of TBANs. WhatamIdoing (talk) 20:10, 21 October 2022 (UTC)

And the third approach is to not waste time wikilawyering about whether a removal constitutes a “valid” challenge or not… when it is FAR quicker and MUCH less stressful to quietly fix the article by adding a source. Don’t focus on what the other guy did wrong… focus on what YOU can do right, to make the article better. Blueboar (talk) 20:45, 21 October 2022 (UTC)

The onus is on whomever added un-cited content. Chris Troutman (talk) 20:48, 21 October 2022 (UTC)

But would either of you ever argue the same thing about cited content? Bullet point #2 says all material whose verifiability has been challenged so if someone added cited content and you challenged the verifiability of it, would you then refuse to give any statement of concern about the verifiability, and just simply say "onus is on you" deal with it, or "stop wikilawyering and focusing on the fact I never gave a valid reason, and just provide a good source since it is faster"? Because, if you would then you are just being butt-hoes in my humble opinion. (Not meant to be a real opinion of anyone. I've interacted with Blueboar a lot. I hope that is taken in good spirit). If I'm being a butt-hoe, let me know. 😀 Huggums537 (talk) 21:05, 21 October 2022 (UTC)

I guess the point I'm trying to make is that if we would normally give a valid reason for our concern for the removal of cited content when we challenge the verifiability of it, then what would be the difference for uncited content? The argument that it is faster to add the source doesn't make any sense to me when we are really already required to do the right thing and give valid reasons in edit summaries anyways so it would be faster to just do that and avoid the wikilawyering to begin with. Also, I just realized WP:ONUS isn't really so much about uncited content as it is verifiable content that doesn't have consensus to belong, which could be any disputed content including either cited or uncited content, but the point is that you would also normally almost always include an edit summary to dispute or remove such content so any such proposal from N8000 would be right in line with current practice about the removal of content. Huggums537 (talk) 21:38, 21 October 2022 (UTC)

The difference is that cited content is exactly that - now you're in a discussion about sources, reliability, due weight, etc. With unsourced content, it's just hanging there. A bureaucratic "I hereby declare I do not believe any source anywhere on the planet makes this claim" or whatever is not needed. Crossroads ^-talk- 00:15, 22 October 2022 (UTC)

I'm not sure I understand your logic. Are you saying that uncited content isn't also potentially "a discussion about sources, reliability, due weight, etc."? It seems to me you are propagating the idea that uncited content is inherently wrong just because it doesn't have the citation at the end as discussed earlier. Huggums537 (talk) 03:01, 22 October 2022 (UTC)

ONUS isn't about anything because its about everything. - Butwhatdoiknow (talk) 06:10, 23 October 2022 (UTC)

That's a novel interpretation that seems fair enough on the face of it, but a quick glance to do an overview of that discussion leads me to conclude very quickly and easily that the whole thing was flawed from the beginning by virtue of the fact that it takes one single sentence out of context in an attempt to define the whole meaning of ONUS. Huggums537 (talk) 09:20, 23 October 2022 (UTC)

Flawed or not, taking the sentence out of context is what most editors seem to do. And the community resists any effort to modify that sentence to clearly tie the sentence to the prior text. - Butwhatdoiknow (talk) 14:37, 23 October 2022 (UTC)

Sad, but true. Huggums537 (talk) 15:11, 23 October 2022 (UTC)

I guess I support the North idea. Andre🚐 21:01, 21 October 2022 (UTC)

I don’t support requiring edit summaries. While I do think they are helpful (and usually try to leave one), sometimes I forget and submit my edits (both to add and remove) without doing so. So should my unsummarized edits be considered somehow “invalid”? Should they be automatically reverted?

More to the point: Should we waste everyone’s time arguing about whether my edit is somehow “invalid” … simply because I didn’t follow a minor bureaucratic procedure? My answer is not just no, but HELL NO.

I consider discussion/arguments about the fact that someone didn’t follow procedures to be some of the most disruptive discussions we have here on WP. In the case of removals for “no citation”, the arguments are far more disruptive than the initial removal was… and definitely more disruptive if someone can quickly and quietly fix it by adding a source. Blueboar (talk) 22:28, 21 October 2022 (UTC)

I am very much in agreement with the general idea of limiting bureaucratic and procedural arguments, as well as not generally requiring edit summaries, but I also don't think that people should just be removing others' contributions if they don't otherwise think there is a problem that they are addressing. Andre🚐 22:36, 21 October 2022 (UTC)

Just assume they are trying to address a problem. Assume they don’t believe the material is verifiable. Assume they are an idiot if you want to… in all cases, the simplest, least disruptive fix is to quietly return the material with a citation. Blueboar (talk) 23:06, 21 October 2022 (UTC)

Agreed. Why else would they remove it if they didn't think it was wrong? Maybe it was part of a series of edits and they didn't want to repeat the edit summary over and over. Crossroads ^-talk- 00:13, 22 October 2022 (UTC)

Fine, but, what if they remove something, then are reverted and the only reason given is WP:BURDEN? Andre🚐 00:14, 22 October 2022 (UTC)

I'm not sure I understand. Someone should not be restoring unsourced content. They should cite a source for the claims they are thereby endorsing in restoring the content. Or else leave it be and trust that if it's actually true and important someone else will add it properly in the future. Crossroads ^-talk- 00:17, 22 October 2022 (UTC)

Why do people remove content that they don't think is wrong, or even that they know to be correct? Because some editors believe that it is inherently and always wrong to have any paragraph in an article without an inline source. Their goal isn't accurate content; their goal is to make other editors to provide sources that they can't be bothered to find themselves. BURDEN allows them to do this.

Also, because some editors find BURDEN very convenient for POV pushing. Imagine someone searching for a particular politician and removing every uncited sentence that has the "wrong" POV. BURDEN allows them to do this. WhatamIdoing (talk) 00:23, 22 October 2022 (UTC)

Sure, assholes exist. However, we are supposed to assume good faith, so we assume that they had a legitimate motivation to remove the content. And even if we strongly suspect bad faith, our fighting about it is what they want… so don’t give them the satisfaction of goading you into disruptive arguments. Just quietly return the material with a source. Win-win… the article is that much better, and they go away unsatisfied. Blueboar (talk) 00:36, 22 October 2022 (UTC)

@Blueboar I have to strongly disagree because I think it is worth fighting for even if it might not be easier. When I was doing vandalism and checking recent changes I used to come across unsourced content that was flagged, and so I would just remove it without even thinking about it because I somehow learned it was the "Wikipedia way", but then one day just by chance the uncited content piqued my interest so I actually read it, and perused the article a bit, and found to my surprise that the content wasn't all that bad, it just needed citations from a newbie who might not have known better. That's when I realized I was becoming too Wikipedified, and I needed to start thinking for myself by making my own judgement calls instead of being lazy and just letting the system do the thinking for me. Sadly, editors who promote the idea of just assuming good faith would never question my intentions and motives, and they might be right in doing so, but they would also only enable me to continue in my lazy Wikipedified ways if they never bother to fight me or call me out on it. I was only lucky to gain the insight on my own. So, it isn't always about assholes because I wasn't trying to be one - I honestly thought I was doing the right thing until I thought differently. Huggums537 (talk) 03:40, 22 October 2022 (UTC)

@Blueboar, the editor that brought me here has NINE complaints on their user talk. Don't get me wrong, this is an experienced, well-intentioned, useful editor. But they believe removing content simply because it's been tagged is valid. The fact there are nine complaints likely means something along the lines of "and a hundred articles got improved by their tagging and maybe on a thousand, no one noticed because the original creator no longer edits and the article is not well-watched." Valereee (talk) 17:49, 22 October 2022 (UTC)

Any BLP subject to strong POVs, like politicians, should be purged of unsourced material. Doesn't matter if the person doing so does it because they love the politician or whatever. Crossroads ^-talk- 01:04, 22 October 2022 (UTC)

No one is arguing that. Everything on a BLP should be sourced, and unless it is completely innocuous (the date they graduated from college or whatever) removed on sight. Valereee (talk) 17:50, 22 October 2022 (UTC)

Suggestion: We could close the door on bad actors with the edit summary idea proposed by N8000 and Valereee, but it seems some editors are not willing to pay the cost of inconvenience, and limited freedom to do this. However, we could still shut the door on bad actors with minor inconvenience, and limits on freedom by simply requiring that a removing editor explain their actions adequately either in an edit summary or on a talk page, etc. That way, the price to pay for shutting the door on bad actors becomes reduced, and much easier to pay because the inconvenience , and limit to freedom would then be minimal since we already have the freedom to explain ourselves where we choose anyway. Huggums537 (talk) 09:43, 24 October 2022 (UTC)
Is "I removed this because it was fact-tagged more than a month ago, and nobody added citations in the meantime" an "adequate" explanation, under this scenario? It is, AFAICT, a reasonably complete and honest description of the actual motivations and actions by a (very) few experienced editors. WhatamIdoing (talk) 03:19, 27 October 2022 (UTC)
That's a pretty good question. "I removed this just because I hate content that doesn't have inline sources" is also a reasonably complete and honest description of the actual motivations and actions by a (very) few experienced editors, and it might qualify as adequate. So, your question raises the idea about how we are defining "adequate". Should it mean "valid" or "honest"? I believe we have been thinking in terms of "valid", but I think the important thing is that an edit summary or explanation is there for others to judge if it is valid or not in the first place. I also think there are enough people complaining about the summary you quoted above to say that maybe it has been "honest", but not "valid". The fact they have said something is a great thing, but I think editors are seeking even more clarification, and it is only one step. Huggums537 (talk) 03:41, 27 October 2022 (UTC)
Actually, I think I explained that all wrong. A better example would have been to say that, "I'm just blanking this" is also a reasonably complete and honest description of the actual motivations and actions by a (very) few experienced editors, but it would be more than fair to say that it isn't adequate by any means since it doesn't explain anything. In fact, it isn't any better than than simply blanking without a summary at all because just telling us what you are doing is not any less obvious than just doing it without saying anything. I think the problem other editors are having with statements like the one you quoted above are essentially the same - they don't really explain anything, they just simply say what is being done, i.e. "I removed this because it was fact-tagged more than a month ago, and nobody added citations in the meantime" roughly translates to: This was tagged. Nobody added a citation. I removed this. Yes, these were all things that were done, but that doesn't really explain anything at all about anything in the situation. Why was it tagged? Why was a citation not added? Why did you remove it? It's essentially a longer version of "I just blanked this". It isn't adequate. Huggums537 (talk) 11:54, 27 October 2022 (UTC)
It sounds like you want an explanation that is related to the substance/content, rather than the form/bureaucracy. If we imagine an uncited sentence like "Joe Film is a famous actor^[fact]", we could imagine multiple reasons for an editor to remove it:
I'm not sure he's famous, and someone else tagged it, so they're probably uncertain about that, too.

I'm not sure he's a really an actor (e.g., maybe he's primarily a film director).

This information doesn't belong in the article (which is about something else, or perhaps it's trivia).

It's bad writing (e.g., puffery or redundancy).

It's uncited, and I assume (without checking?) that if nobody added a citation, it's because it can't be done.

I blank content to manipulate other editors into increasing the number of citations in an article.

It sounds like you want to hear more of the first reasons, and maybe ban the later ones. WhatamIdoing (talk) 16:08, 27 October 2022 (UTC)
Sure, I want more of the first, and less of the later ones, but a ban is not really doable. I think just any suggestions that focus on the cn tags might be best then. Huggums537 (talk) 20:54, 27 October 2022 (UTC)

BURDEN - arbitrary break

I have realized where some of my frustration with this (and similar) discussion(s) stems from… I think we are conflating two separate questions. The first question is:

What should I do when I come across unsourced information in an article?

The second question is:

How should I react when someone else has come across unsourced information in an article, and decided to remove it?

My answer to the first question is that I should follow best practice by hesitating before removing the content and attempt to improve the article by adding sources. However, best practice is a higher standard than allowed practice, so it is forgivable if (sometimes) I simply cut the content.

My answer to the second question is more complex… I can not force someone else to follow best practice. Trying to do so is a waste of time and energy. Besides, in the situation we are discussing, it is already too late for that… the other guy has already followed allowed practice and cut the content.

I suppose I could use the situation as a “teachable moment”, and explain best practice to the other guy… but they won’t be receptive to that if I am being antagonistic and trying to force them to edit to my high standards.

However, best practice for me remains the same: try to improve the article by adding sources.

Then, if I want to use my edit as a “teachable moment” I can reach out to the other guy and explain best practice in a non-confrontational way.

This is why I keep coming back to: “don’t wikilawyer… just fix”. Blueboar (talk) 13:25, 22 October 2022 (UTC)

@Blueboar, honestly I'd be happy with adding a statement of best practices. As I've pointed out before, the particular situation I find myself in is with a well-intentioned, experienced editor who I'm sure would want to follow best practices and likely thinks they already are because we haven't made it clear. Valereee (talk) 17:54, 22 October 2022 (UTC)

We already have a statement of best practices at WP:PRESERVE. Blueboar (talk) 21:15, 22 October 2022 (UTC)

Unfortunately, BURDEN doesn't make reference of PRESERVE and the language at PRESERVE appears to defer to BURDEN, so it becomes moot. Carter (Tcr25) (talk) 05:29, 23 October 2022 (UTC)

My gripe is that PRESERVE seems to encompass both cited and uncited materials, but BURDEN, and statements made like the one Blueboar made above that only asks the two questions about unsourced material seem to completely ignore what "best practices" are for anything else except the unsourced stuff exclusively. This seems to clearly show that editors are confused about thinking BURDEN is solely about unsourced material even when they say we have best practices to guide us about content at PRESERVE. I think BURDEN would be very well served making a reference to PRESERVE as a reminder that unsourced material isn't all there is to think about because apparently even experienced editors are not totally aware without it. Huggums537 (talk) 09:44, 23 October 2022 (UTC)
In other words, I think Blueboar should reformulate the points he is trying to make with what the actual best practices at PRESERVE are, and not just strictly how it applies to only two questions about unsourced content because until then they are incomplete ideas since there are parts in BURDEN that suggest it also applies to the removal of cited material whose verifiability has been challenged just as suggested in PRESERVE, and that might be part of what is so frustrating. Huggums537 (talk) 09:59, 23 October 2022 (UTC)
Blueboar, please forgive me for making an example out of your post, but it made a prime target for the type of ideas I am trying to promote. I mean you no harm by it. Huggums537 (talk) 11:43, 23 October 2022 (UTC)
I guess it is important for me to make it clear why I have been making such a big stink about cited material that isn't supported by the references in this thread. It is because I think it might be just as problematic (even if not as prevalent) that there is a possibility someone can tag a statement as dubious if they think the reference doesn't support it, and then someone else can just come along and remove it based on the tag without even trying to find a source to support the statement so it would be a very similar issue to the CN tag problem, but I never see anyone asking any questions or even thinking about cited content at all when it comes to BURDEN. It is very troubling and disturbing to me that the focus sooo much on uncited content only. Huggums537 (talk) 11:01, 23 October 2022 (UTC)

Simplified flow chart for removing bad content from English Wikipedia

Huggums, I've read all of this, and I'm really not understanding why it's important for badly cited material to be "removed per BURDEN". So please have a look at this little flow chart. Imagine that your goal is to remove some bit of content. This chart gives you three different explanations for removing the content, depending on the circumstances.

There is the circumstance in which it's uncited, and you claim BURDEN while removing it.

There is the circumstance in which it's cited badly (includes all problems related to sourcing: inadequate, misrepresented, unreliable, inappropriate, etc.), and you claim WP:V while removing it.

There is the circumstance in which it's cited properly, and you claim NPOV (or really, any other policy or guideline except WP:V) while removing it.

Every single time you get to remove it. Why are you insisting that you must be able to claim BURDEN specifically – that no other policy can possibly be adequate – when you still get to remove the content? WhatamIdoing (talk) 06:24, 27 October 2022 (UTC)
I haven't insisted any such things. I'm puzzled and perplexed as to where you're getting such ideas from. If you think you are getting them from me, then either I have failed miserably at trying to get my ideas across, or there is some other idea at play that I have not yet grasped myself. It isn't so much important to me that any material be removed by BURDEN so much as it is that BURDEN give consideration to cited material when it is explaining itself. In the current form, BURDEN explains itself mostly from the point of view of strictly uncited content. I think that this highly selective form of explanation by neglecting the point of view of cited content skews the very meaning of BURDEN in an extremely biased way. I really don't know how to explain this any better. Whenever I can nail down exactly what those biases are, and precisely how it skews the meaning, then maybe I'll be better equipped to explain it, but for now, all you get is: talking strictly about unsourced content is bad, but that doesn't in any way mean I favor any kind of strict claims of BURDEN for any kind of content. Huggums537 (talk) 12:26, 27 October 2022 (UTC)
Why does BURDEN (specifically BURDEN) need to give any consideration at all to cited material? Almost the entire rest of the policy is about badly cited material. Does badly cited material need to be addressed in every single section? Why can't we have one section in this policy that is exclusively about uncited material?

I think I'll understand it better if you can fill in the blank at the end of this statement: "Right now, BURDEN applies to uncited material only. This means that if your material is uncited, I can blank it, and you can't put it back unless you add one source that you believe(!) is decent. I want to make BURDEN apply to already cited material. If we do that, then if your material is already cited, I can blank it, and you can't put it back unless ___________".

Right now, it seems to be that the thing that would belong in that blank is "revert me with exactly the same source that's already in the article, because you apparently believe that source is decent, or you wouldn't have added it in the first place." WhatamIdoing (talk) 16:17, 27 October 2022 (UTC)
I mean it's kind of impossible for the section to be exclusively focused on unsourced material isn't it? My view is that it already does talk about cited content, but just not nearly enough to the extent that it suppresses those ideas in the unbalanced favor of ideas geared toward the removal of unsourced content. From your perspective, you fear that talking about cited content, and bringing those ideas more to the surface will make removal of cited content just as problematic as the removal of unsourced content is, but from my perspective, I see the ideas about cited content as already being firmly in place, but just not expressed as freely as the ideas about unsourced content so I have no fear of how it affects the removal of cited content, but I believe it would change the way editors view uncited content if they are forced to see those same views compared side by side with the ideas about cited content. They are currently unable to do this because of the suppression of the one, and the over-emphasis of the other. For example, consider all of the differences between how you perceive your own statement, and how I view it:

You: Right now, BURDEN applies to uncited material only.

Me: Editors apply BURDEN mostly to uncited material, but clearly it also applies to cited material such as contentious BLP violations that are poorly sourced, or any cited content whose verifiability has been challenged.

You: This means that if your material is uncited, I can blank it, and you can't put it back unless you add one source that you believe(!) is decent.

Me: This means that it doesn't matter if your material is cited or not, I can blank it (based on a challenge of verifiability), and you can't put it back unless you add one source that you believe(!) is decent.

You: I want to make BURDEN apply to already cited material.

Me: I believe BURDEN already applies to cited material (whose verifiability has been challenged), but it has been suppressed while other points of view have been over-emphasized.

You: If we do that, then if your material is already cited, I can blank it, and you can't put it back unless ___________

Me: If your material is cited or not cited, and I challenge the verifiability of it by blanking it with my explanation, you can't put it back unless you've verified it.

You: Right now, it seems to be that the thing that would belong in that blank is "revert me with exactly the same source that's already in the article, because you apparently believe that source is decent, or you wouldn't have added it in the first place

Me: I would absolutely agree with that, but I would also add that you could revert with the same source that was used to blank no matter what rationale the blanking editor gives for blanking.

Please note that in none of the "Me" responses do I ever once indicate that the proper rationale used for blanking is BURDEN. You could still easily blank based on V or NPOV, and that would not change whether BURDEN applies or not. Nobody says you have to use BURDEN or even that you should. Huggums537 (talk) 21:47, 27 October 2022 (UTC)

I guess my previous comment before this one is a good example of how I've explained myself stupid in this thread to the point of being reduced to caveman responses - BURDEN bad. Me fix BURDEN. Lol. Anyway, maybe you haven't been understanding my views because I've been making them too complicated. In a nutshell, I think BURDEN talks way too much about whether content has the inline citation at the end of the sentence to the expense of the more important issue of verifiability, which could be referenced by other means. That's a very simplified version that leaves out other side issues. I thought you of all people would understand, and you have given me a pretty gnarly bone to chew on by wrongly thinking I had other goals in mind, but your concerns about it have got me to thinking that maybe some of my ideas might have a double edged blade to them I had not considered so I don't know if I should gnaw on that gnarly bone for a while or just go bury it somewhere and dig it up later... Huggums537 (talk) 08:20, 29 October 2022 (UTC)

@WhatamIdoing, I know this is simplified, but what's the flow chart when it was cited, but to general references? I'd like to see inline citations added, but it's for me not so urgent that we should remove likely-good content. Where does that fit? Valereee (talk) 23:00, 4 November 2022 (UTC)
I don't think we've ever suggested a pathway for that. It would probably require a step in there that says "somehow, already know that one of the sources already named on the page contains the uncited information". WhatamIdoing (talk) 22:04, 6 November 2022 (UTC)
In other words, I think the questions about how I act, and how I should react to others becomes more complex when we have to talk about cited content as well as uncited content, and I think that when you add that to the mix the dynamic of how you act and react takes a new direction as well. We should be discussing how to act and react to the verifiability of all content in general on this forum, and not just uncited content. When you look at things through this new lens, it then becomes nearly impossible to make such bold statements such as I can not force someone else to follow best practice. and Besides, in the situation we are discussing, it is already too late for that… the other guy has already followed allowed practice and cut the content. For example, unexplained content removal is most often seen as a form of vandalism when the content is cited, and I can force the removing editor to comply with the best practice of leaving a valid edit summary. Huggums537 (talk) 13:06, 23 October 2022 (UTC)

Except that we can’t force someone to leave an edit summary… all we can do is react to the fact that they didn’t. How you react is up to you. You can react in a confrontational way, or in a non-confrontational way. I think best practice is to be non-confrontational. I understand that being non-confrontational is not always easy (and if the other person is looking for a fight, may not even possible)… but it is still best practice. Blueboar (talk) 14:59, 23 October 2022 (UTC)

Well sure, I stand corrected on that point. I can't truly force the offending content remover to leave an edit summary if they choose to explain themselves some other way like the talk page or something, but I guess I could have explained it better by saying that unexplained content removal is usually automatically considered invalid as a form of vandalism, and is almost always automatically rejected in the form of a reversion until the error is corrected in some way and this action essentially forces the best practice of making the removing editor explain the removal in some way or keeping the content. Huggums537 (talk) 15:38, 23 October 2022 (UTC)

In other words, there aren't very many options for an editor who leaves an inadequate edit summary when removing content, and "it had a cn/dubious tag" is not an adequate edit summary. It's a piss poor edit summary that really amounts to unexplained content removal since, "it had a CN/dubious tag" doesn't "explain" anything other than the fact that someone else placed a tag on it for some reason that has not yet been explained. If I saw an edit summary like this I would revert it as unexplained content removal or inadequate edit summary. Huggums537 (talk) 16:23, 23 October 2022 (UTC)

That is certainly an understandable (and common) reaction… but it isn’t best practice. Best practice would be to be less confrontational, and to simply return the material with the requested citation. Blueboar (talk) 16:53, 23 October 2022 (UTC)

Per WP:MINREF, if someone tagged material in the past, that was a legitimate challenge to the material, edit summary or no. If you feel otherwise, it seems to me that the appropriate thing to do, short of simply providing a source to moot any argument, would be to initiate a Talk page discussion regarding whether a source is really needed. Simply reverting the removal of tagged material, especially if it was tagged long-term, seems to me to be far from the optimal response. Unless there's reason to believe otherwise, the good-faith assumption would seem to be that editors removing tagged material are doing so because it was tagged in the past and nothing had consequently been done to address the concern. DonIago (talk) 17:38, 23 October 2022 (UTC)

Wow. I do have a reason to believe otherwise already. I would say the less than optimal thing is MINREF. That thing is so far wrong it is scary. How can any statement be regarded as a "valid" challenge simply because it has been removed, when most material that has "simply been removed" without explanation is considered an act of vandalism? So, now we are regarding acts of vandalism, and inadequate edit summaries as "valid challenges" to content? Because, that line of reasoning essentially says removal for any reason is perfectly valid, but it really isn't acceptable, allowable, or best practice at all. It is allowable only to the extent that anyone even without an account can get on here and start blanking shit, but that would not be sustainable since they would be indef blocked for vandalism in short order. I would say MINREF needs to be modified to say something more along the lines of: Any statement that has been challenged (e.g., by being questioned on the talk page, in an adequate edit summary, or other reasonable form of notice.) I will concede that the question about tagged material might be a debatable one, but the question about material that has just simply been removed absolutely is not in my mind. Simply removing material without explanation should always be considered an act of vandalism until it has been proven otherwise. If the other editor made a mistake, they will let you know. Nobody says you have to be confrontational with anyone just because you think blanking content without reason is vandalism until proven otherwise. That isn't bad faith, that is good patrolling. Right? Hahaha. Huggums537 (talk) 21:25, 23 October 2022 (UTC)

Hold on… Removing material for no reason is not the same as removing material without stating a reason. Talk about assuming bad faith! Blueboar (talk) 21:38, 23 October 2022 (UTC)

If it is your belief that not making the distinction between removing material for no reason, and removing material without stating a reason could be an act of bad faith, then MINREF is in even worse condition than I feared since it also commits the "bad faith" act of failing to make any such discernable differences about the removal of content as being "a challenge". Huggums537 (talk) 00:09, 24 October 2022 (UTC)

I think it is also important to point out that MINREF seems to at least partially share my views in that respect, as well as by making this statement: For example, section blanking may be considered vandalism, rather than a demand for inline citations.

It is vitally important that editors understand the implications of saying that the simple act of removal amounts to a valid challenge. This gives trolls and vandals the ability to blank content with impunity without any fear of consequence since all they have to do is say it was an accident they didn't leave a summary, then come up with any silly reason whatsoever to say it was actually a challenge, and it won't matter if the reason is valid or not because it then turns the blanking action from a behavioral issue into a discussion about the challenge, where the worst thing that could possibly happen to them for their actions is that they turn out to be incorrect in the discussion. It is a loophole for trolls, vandals, or really any bad actor to avoid consequences for blanking content in bad faith. Huggums537 (talk) 08:45, 24 October 2022 (UTC)

Editors are allowed to use their best judgement and some common sense, e.g., to decide whether an unexplained section blanking is or isn't an intentional CHALLENGE. If someone blanks a section that happens to be uncited, and you decide it's vandalism, they're free to come tell you that you're wrong and insist that it was a CHALLENGE. Until then, though, you're free to use your brains.

We don't seem to have a significant problem with this. (Also, MINREF [assuming it's still up to date] merely reflects and explains the rules that were written in WP:V and WP:BLP. It does not make any rules itself. That's why MINREF begins with the words "Wikipedia's content policies require..." instead of "This informational page recommends...") WhatamIdoing (talk) 06:37, 27 October 2022 (UTC)

@WhatamIdoing, I don't have a problem with that either because it means they had to come and tell me it was a challenge. In other words, they had to explain themselves before the blanking became a challenge, otherwise it still remains vandalism, and I believe that is absolutely the way it should be. WP:BLP only applies to living people, so where else other than MINREF does it say that the mere act of removal is a challenge? I've seen a number of editors say that this is the policy in several discussions, but I have failed to find it or see anyone actually quote it until Doniago linked us to MINREF. I mean I hate to buck several well established editors, but it seems they are just assuming that is policy rather than going by anything actually in it. Maybe I have been missing something? Huggums537 (talk) 13:06, 27 October 2022 (UTC)

There are many reasons to remove content from articles that don't constitute vandalism (e.g., splitting an article, copyediting, NPOV/DUE emphasis).

Where in MINREF does it say that the mere act of removal is a challenge?

WP:CHALLENGE says "Any material lacking an inline citation to a reliable source that directly supports the material may be removed and should not be restored without an inline citation to a reliable source", but I see nothing in WP:MINREF that says this. WhatamIdoing (talk) 16:21, 27 October 2022 (UTC)

There are many reasons to remove content from articles that don't constitute vandalism (e.g., splitting an article, copyediting, NPOV/DUE emphasis). All of those reasons for removing require an explanation or else they can be considered vandalism without it. It isn't until the explanation has been given that the idea of it being vandalism goes away. Splitting an article, or copyediting isn't always obvious, but sometimes maybe copyediting might be.

Where in MINREF does it say that the mere act of removal is a challenge?

In the explanation box thingy it says: Any statement that has been challenged (e.g., by being removed, questioned on the talk page, or tagged with ^{[citation needed]}, or any similar tag) As I've explained elsewhere, this is so vague it literally allows the edit of a blanking vandal to be considered a "valid challenge". Bad, very bad.

WP:CHALLENGE says "Any material lacking an inline citation to a reliable source that directly supports the material may be removed and should not be restored without an inline citation to a reliable source" Ok, here I see it saying that material can be removed, but I don't see it explicitly saying that this removal counts as being the challenge itself?!? Are we just supposed to assume that this is the implied meaning, and why should we especially given the fact that making a mere removal a valid challenge essentially greenlights vandal blanking as I described above? Huggums537 (talk) 22:47, 27 October 2022 (UTC)

Something I want to add to this comment is that not only do I not see it explicitly stated that a removal itself counts as a challenge, but I also don't see an edit summary, a tag, or even someone having the mere thought that this material lacks a citation so it should be removed counts as a challenge. In other words, I just see it simply saying it is ok to remove, and not ok to put back until the blue clicky number is there, but no reference whatsoever indicating any kind range for what defines the challenge itself. If we are assuming it is the removal, then that certainly invalidates the theory that a tag is a challenge, but if other editors are claiming a tag is a valid challenge, then that invalidates the theory that the passage above means "the removal". I personally feel like both are invalid since neither are explicitly stated in policy anywhere. Edit summaries are automatically valid challenges because they are indisputable proof a challenge has been issued no matter if it is enshrined in policy or not. Removals, and tags in and of themselves are very circumstantial, and disputable unless accompanied by an adequate edit summary. Huggums537 (talk) 08:59, 29 October 2022 (UTC)

Tagging for citation needed isn't necessarily a statement that the material is being challenged. Usually it's simply a statement that the material requires a reliable source. Valereee (talk) 18:12, 24 October 2022 (UTC)

Absolutely not. Per our policy, a citation is needed if some material is likely to be challenged, otherwise no reference is needed (although there are exceptions: BLP etc). Therefore, "cn" tag definitely means the material has been challenged. Indeed, when I see some statement that looks non-controversial, it would be ridiculous to put the tag. However, when I see some statement, and I have some doubts in its validity, I want to see the source that supports it. In that situation, I put the "cn" tag.--Paul Siebert (talk) 18:45, 24 October 2022 (UTC)

So we shouldn't use cn tags on any article that has no inline citations but does have a list of references? Valereee (talk) 17:16, 25 October 2022 (UTC)

That is a very good question. That depends on our interpretation of the term "verifiability". If we interpret it as "some statement that can be verified, because at least of published source supports it, and it is theoretically possible to find that source" then there is no need in any citations at all (BLP is an exception for legal reasons). If we follow this approach, everything that was published at least once is verifiable.

However, if we interpret "verifiability" as "some statement that any person with an average knowledge in a field can independently verify in a reasonable time", then everything depends on how we define the term "reasonable time". In that case, some short article that cites just one small book probably needs no inline citations. However, a huge article that cites a long list of books and has no online citations can hardly be considered verifiable in a reasonable time. Indeed, to verify some statement, a user must read, in full, the whole list of books cited in the bibliography section.

In addition, Wikipedia texts are being copy-edited, that is a continuous process, and even if the original text correctly reflected the sources cited in the list, we cannot guarantee that current version contains no unintentional (or even intentional) distortions. Normally, other editors may easily verify each statement that they find questionable just by going by the reference. However, if no inline citation is provided, that verification requires much more efforts.

One way or another, we need to decide what exactly should we call "verifiability". Paul Siebert (talk) 17:42, 25 October 2022 (UTC)

Maybe we need a tag for 'lacks inline citation' rather than 'citation needed'. And that tag would indicate that this is not a challenge to the material, simply a request for the addition of an inline citation when someone gets around to it. Valereee (talk) 17:49, 25 October 2022 (UTC)

If I remember it correctly, we already have that tag. It says something like "The article has a reference list, but it has no inline citations". I remember I saw it on a top of some articles. Check it. Paul Siebert (talk) 18:01, 25 October 2022 (UTC)

That's a banner rather than a tag, I think? Valereee (talk) 18:07, 25 October 2022 (UTC)

Yes. Paul Siebert (talk) 18:17, 25 October 2022 (UTC)

Valereee is right: you can, and people do, leave a {{fact}} tag in an article to encourage a source for content whose inclusion you support and would be sorry to see removed, or as a simple aide-mémoire for yourself.

The problem is that later editors, unless you added |reason=This is not a CHALLENGE. This is good content would have no idea what your intention was, and they are therefore free to interpret it as a proper CHALLENGE. One hopes that they'll use some common sense, and even that they'll know something about the subject matter, but if you add such a tag, there is a risk that it will be understood as a proper CHALLENGE even if you don't intend to do that. WhatamIdoing (talk) 06:32, 27 October 2022 (UTC)

I wonder if anyone thinks it is a valid challenge if you use the {{verification needed}} or {{Failed verification}} tags? Huggums537 (talk) 10:05, 31 October 2022 (UTC)
- Of course those are valid challenges... but so is a simple cn tag... and so is leaving a comment on the talk page... and so is removing with an edit summary... AND... so is simply removing without any edit summary or other explanation. They are all valid challenges. Sure, some are more helpful than others, but they are all "valid". Blueboar (talk) 10:55, 31 October 2022 (UTC)
  I added links to the templates so we can see exactly what they are supposed to be for. Any editors encountering these tags would naturally be expected to look up their actual meaning, and not just take on any old assumed meaning of what they *think* a tag is supposed to represent in their vivid imaginations. The same would apply for a {{Cn}} tag as well. Many of the arguments being talked about here are already settled in the instructions to editors at the templates themselves, such as the issues about time limits, and the good argument Valereee made about Cn tags being more of a request for sources than an issue of a challenge. For example, the Cn template specifically instructs editors the tag is a request for a citation to support the statement, and it goes on to say there isn't any time limit, asks specifically to, "please not delete just because nobody has ever cited in some arbitrary amount of time" [paraphrased], and the majority are willing to wait a minimum of one month for sources. It also says further up, The ^{[citation needed]} template is intended for use when there is a general question of the verifiability of a statement, or when an editor believes that a reference verifying the statement should be provided. Consider how this also applies to {{Verification needed}}, and BURDEN. If BURDEN is supposedly a challenge to content that doesn't have inline citations, then how in the world is that tag a "valid" BURDEN challenge to the referenced content it was intended to be used for? The instructions clearly state: Put the following after the reference - after and outside of the closing </ref> tag - to request that someone verify that the cited source supports the material in the passage: and the same goes for the {{Failed verification}} tag. The instructions clearly say: Use this tag only if: an inline citation to a source is given, So, exactly how is this tag a "valid" BURDEN challenge if the use of it demands that the inline source already be given? I think most editors are confusing the generalized idea of what constitutes the overarching premise of a basic dispute with the more specific idea of a BURDEN challenge. In other words, I think most editors are taking the specialized meaning of a BURDEN challenge, and mistaking it for the broader meaning of "anything I have disputed on Wikipedia". So, what makes these tags a "valid" challenge if those things I mentioned are true, and at the same time someone coming along to remove the so called tagged and challenged content is also said to have initiated a valid challenge? Who then initiated the "valid" challenge? Was it the tagger, or the remover? If you say it was the tagger, then the remover as a challenger becomes moot, and if you say it was the remover, then the tagger has not yet completed the removal portion of initiating a valid challenge. It can't be both. Once the material has been challenged with BURDEN, it can't be "re-challenged" with the exact same BURDEN again later on the talk page, and then again later with just a blank edit summary. That's just vandal blanking. At which of those three points do we name as the valid challenge point without invalidating the others? I posit that it must be the point on the talk page since it doesn't make any assumptions about what a tag or removal is supposed to mean in the imaginations of editors, but explains itself clearly, and adequately. Saying any of these other actions are "valid" challenges by just assumption alone without the requirement that any of the actions be accompanied by an explanation goes well beyond "unhelpful", and ventures too far into vandalistic territory as I've warned about before. Huggums537 (talk) 19:14, 31 October 2022 (UTC)
  I think you are looking for crystal-clear answers in a process that requires editors to guess at what someone else might have been thinking. We're telling you: When uncited material is removed, your safest bet is to assume that it's a valid CHALLENGE. You keep replying with "But what if it's not?!" And the answer is: Unless you have a strong reason to doubt it, then you should assume that it is. The harm of citing something is very low. WhatamIdoing (talk) 22:45, 3 November 2022 (UTC)
  But, that's just the thing, at least 3 reasons have been given to have a strong reason to doubt it in this thread. At least two from myself, and at least one from another. Many more can be provided by myself upon request, but how many more do you all need to doubt removal is a valid challenge? Huggums537 (talk) 06:50, 4 November 2022 (UTC)
Both of those are much more explicit challenges than {{citation needed}}. Each directs the editor to look for a source or check the listed source before placing the tag, making it a hopefully informed challenge of verifiability. CN only indicates "a general question of the verifiability of a statement, or when an editor believes that a reference verifying the statement should be provided." The later seems to be more a question than an outright challenge ("is it?" compared to "I don't believe it is."). — Carter (Tcr25) (talk) 17:13, 31 October 2022 (UTC)

Let's take a look at the problem from a different angle

Correct me if I am wrong, but the users who oppose to removal of the content labelled with the "cn" tag imply that any information added to Wikipedia is potentially valuable, at least, because the users who wrote it invested their time, Therefore, their labour must be respected.

However, from my experience, I know that it may be much easier to write some text than to verify the text written by somebody else. Meanwhile, our policy dramatically underestimates the latter aspect. Indeed, blanket removal of some text is considered as vandalism, and the editor who is engaged in such activity is considered a disruptive editor. That seems reasonable, because people who wrote the removed content invested their time and efforts, so their labour should be respected. However, a user who adds some statement that is not easy to verify (or that is not verifiable) is not considered a disruptive editor. In my opinion, that is a big mistake, because it may be much more difficult to find a reference for such a statement (or to prove that no sources can be found) than to write several paragraphs of unsourced text. In other words, users who write some text without providing references may (at least, sometimes) be even more disruptive, because they waste other user's time and efforts.

I fully realize that it would be a bad idea to propose any sanctions for those who provided no citations for the texts they write. However, our policy must become more strict in that aspect, for example, by making easier to remove unsourced texts. One possible approach is to specify that the texts labelled with the "cn" tags may be removed without further explanation if the tag stays, e.g., for 1 year. --Paul Siebert (talk) 18:45, 24 October 2022 (UTC)

I think this is even more of a very good example of why an explanation is needed to prevent bad actors from blanking or even adding content either way it goes. I fully disagree users adding text would be even more disruptive, but the requirement for them to somehow explain their actions would work both ways if Paul likes to imagine that is the case. I also disagree about our policy becoming more strict for the allowance of removing unsourced text. No good reason has been provided for it, whereas the very good reason of preventing bad actors (either way) has been provided for making users explain their actions. Huggums537 (talk) 01:21, 26 October 2022 (UTC)

Also, I think time limits are a bad idea since it is an indication of bad faith. It is good faith to assume the content was rightly put there in the first place, and then it is also good faith to assume the content/tag was not immediately removed for a good reason. Removal based on strict time limits is just a an automatic bad faith assumption. A terrible thing to write into policy. Huggums537 (talk) 01:30, 26 October 2022 (UTC)

Under my idea, the challenger merely has to say that they have a concern that the statement may not be verifiable. Nothing else required. Here's a example of the intended use before and after the proposed change. To simplify, this is just about a wp:ver question (not a wp:weight question):

Current

I put in a statement that Mary Smith received the XYZ award with a typical quality source (sufficiently strong but not bulletproof/wiki-lawyer proof). What is implicit is that a sky-is-blue statement does not require as bullet-proof of a source as a questionable claim does.
An editor who hates Mary Smith and wants to minimize positive content regarding her knocks it out on a wp:ver basis citing a technicality/imperfection in the source.

Proposed

I put in a statement that Mary Smith received the XYZ award with a typical quality source (good but not bulletproof/wiki-lawyer proof). What is implicit is that a sky-is-blue statement does not require as bullet-proof of a source as a questionable claim does.
In order to knock it out, the Mary Smith hater would have to say "I have a concern that it may not be verifiable that she recieved the XYZ award" Since they would look very stupid saying that about a sky-is-blue statement, they do not knock the material out.

Sincerely, North8000 (talk) 18:46, 24 October 2022 (UTC)

It seems to me that if tagged statements (which I believe is primarily what we're talking about here) are being removed after a "significant" (yes, I know that's subjective) amount of time has passed, and no edit summary has been provided, then perhaps the best course of action is to merely assume that the removing editor is doing so because the statement was tagged and they share the concern that the statement may not be verifiable. Of course, there's also nothing stopping you from asking the editor what their rationale was.

I think you're going to encounter a lot of resistance if you start pushing toward requiring edit summaries in any capacity (I've seen some of those discussions before), and may just encourage editors to lie in any case. As it is plenty of editors (particularly ill-meaning ones) use misleading edit summaries on a regular basis.

Put another way, if I was a bad-faith actor and the policy was changed in this manner, then I'd just start applying a generic edit summary of "may not be verifiable" when removing text.

I guess I just don't see what the net gain is here? DonIago (talk) 19:25, 24 October 2022 (UTC)

There are two benefits: 1) forcing bad actors to lie makes it more difficult for them to manage their deeds, and 2) it makes it easier for well meaning editors to track their actions and build a case to modify their actions or eject them from the playing field. Huggums537 (talk) 01:38, 26 October 2022 (UTC)

North8000, you are describing the situation that has no relation to what I say. I am describing the case when NO source was provided, and the "cn" tag was placed. In your example, the source was provided, and Mary Smith's hater points an some (real or perceived) imperfectness in the source provided.

Correct me if I am wrong, but the whole section (starting from Valereee's initial post) is devoted to removal of the content that is not supplemented by any citation and was labelled with the "cn" tag. Paul Siebert (talk) 19:43, 24 October 2022 (UTC)

True, but I think that I'm pointing out a flaw in such common conversations which causes them to be off target and their results to get misused. We talk as if we are dealing with zero source situations and then the results of the conversation get are being applied to are what IMO nearly all questionable wp:ver-based removals are which is this in tandem with nitpicking a provided source. North8000 (talk) 20:12, 24 October 2022 (UTC)

Wikipedia overflows with un-cited information. If there are cases where we're using less than stellar sources, remove that content, too. This encyclopedia doesn't need more content; it needs better content. Chris Troutman (talk) 20:16, 24 October 2022 (UTC)

I'm guessing that 90% of Wikipedia has sources/sourcing that are attackable as being imperfect. And IMO the majority of problematic removals are where it has sourcing is by common sense strong enough to support the statement. North8000 (talk) 20:41, 24 October 2022 (UTC)

Can you offer examples for your guess? We cannot be using blogs, Twitter, or rando websites even if we might suspect the facts asserted could be true or are at least not controversial. I don't see value-add in retaining content that is not well-sourced, true or not. Chris Troutman (talk) 20:44, 24 October 2022 (UTC)

I'll find some examples. BTW I would not advocate using/ keeping based on ultra-weak sourcing such as the examples that you gave. North8000 (talk) 20:54, 24 October 2022 (UTC)

OK, I'll do it by showing how a sourced sky is blue statement in some of my own work would be attack-able: Specifically the sky-is-blue statement that Justine Kerfoot wrote the book Woman of the Boundary Waters at Justine Kerfoot#Kerfoot as an author which is sourced to the book itself which is plenty-strong sourcing to support the statement that she wrote the book. If I was a Justine Kerfoot hater / POV warrior I'd knock it out saying "need an independent secondary source" North8000 (talk) 21:18, 24 October 2022 (UTC)

Of course, a book (generally) states who the author is. Unless other sources refute it, I don't think anyone would argue the book isn't good enough. Certainly we need a citation to state that she wrote it although I could see the point in claiming the work itself is the citation. I don't think we need a change in BURDEN for a case like this. Chris Troutman (talk) 21:40, 24 October 2022 (UTC)

WP:V doesn't require independent secondary sources for any individual claim. That's really only required by WP:N. For verifiability, the source has to be appropriate for supporting the claim. A book with the author's name printed on it is generally adequate for that purpose (unless contradicted by others, e.g., claims that a pseudonym was used). This is true even if the book is a self-published autobiographical e-book. WhatamIdoing (talk) 06:41, 27 October 2022 (UTC)

North8000, frankly, I really don't understand you. If you put in a statement that Mary Smith received the XYZ award and provided the source "A", then any user who believes that the source "A" is not good, or it does not say that she received the XYZ award is expected to demonstrate it, for example, in the talk page discussion. If you and that another editor are not able to come to a consensus on that matter, the standard way is to come to RSN and ask a simple question: "Is the source "A" reliable for the statement that Mary Smith received the XYZ award?" In a very unlikely case when this question has been ignored by a community, you may resort to DRN, RfC, and other tools that were specifically designed for this type cases.

In connection to that, I am sincerely unable to see any connection between the topic of our discussion and your question. I concede I had no time to read the previous discussion in full, and it may be possible that someone raised this question during it. However, that does not change the fact that the situation described by you is crystal clear, and our policy has all needed tools to resolve it. Paul Siebert (talk) 22:48, 24 October 2022 (UTC)

I think I understand North8000 and can give another exemple. Let's say I write that "Svante Pääbo won the medicine nobel prize in 2022". I will probably NOT source that. Because my expectation for a GOOD source will be exceedingly high : A secondary source, renowned in the field pertaining the article who so happen to name Svante Pääbo in a broader academic context. I will not just drop "www.nobelprize.org" as a source pretending it's appropriate. Because it's a waste of time for my reader. As long as I did the verification, no one should be allowed to swipe in and "contest" on a basis of verifiability just because the source isn't ajoined to the statement. They should either know who is the nobel prize themselves or be motivated enough to go check. If they meet neither of those criterion, they should just assume good faith and wait for someone interested in the topic to pass by.They could still contest on the basis on relevancy because it's a superbowl article, but that's a topic for another day. No one should assume that I WANT that statement to stay in the article or that I'm interested in having a debate in talk page about it. The policies should be clear, by their own, that this kind of unmotivated challenge is disruptive. "Verifiable" not "adjoined sources" is the core spirit of the pillars. And before this comes up, no it's not about how trivial the information is. The more the subject is obscur, the less urgent it is to fix it. The next user who find the information potentially useful WILL verify, if it take 12 months it's okay too. Iluvalar (talk) 03:53, 25 October 2022 (UTC)

As soon as we started ti talk about the spirit of the policy, it seems obvious, that, according to the policy's spirit, the content is deemed "verifiable" when it is relatively easy to verify. Thus, in your example, it is easy to verify that Svante Pääbo won the medicine Nobel prize in 2022, because that information can be found in almost every reputable mass media. Therefore, I doubt any citation is needed for that statement. However, many (if not majority) of other statements cannot be verified equally easily. Therefore, for sake of our readers (and fellow editors) convenience, it would be correct to supplement that information with a reference to the source.

I would say (and I just realized that that should be possibly become a part of our policy) that the rule of thumb should be:

1. If you add some statement that seems obvious to everybody, to citation is needed.
2. If you add some statement that you found in some easily accessible source, and you believe a major part of Wikipedia audience and the community may be familiar with that source, no citation, probably, is needed.
3. If you add some statement that, as you believe, is 100% verifiable, but the statement seems not obvious, and you don't remember where did you read it, then you should go to a library and try to find that source: it is quite possible that you remembered the information incorrectly, or that the source that, as you believe, does exist, is not possible to find. I am writing that based on my own experience: I know a story told me by my supervisor many years ago, it is a very interesting story that, according to some experts in the field looks 100% trustworthy and verifiable, but, despite my numerous efforts, I was not able to find any source that supports it. That means, although I, as well as some top experts in the field, sincerely believe that fact is verifiable, in reality it is not.
4. If you add some statement that is supported by some source that was not easy for you to find, then it is most likely that for other people it would be equally (or even more difficult) to find. Therefore, you must provide a citation, thereby showing respect to a reader and fellow editors.

Paul Siebert (talk) 16:41, 25 October 2022 (UTC)

@Paul Siebert, the thing is, this isn't really about adding content. It's mostly about older content created before our current requirements, and the content may have been added by someone no longer editing. So in many cases there's no one maintaining the article, which means the "you add" in the above four scenarios isn't what's happening. Valereee (talk) 16:48, 25 October 2022 (UTC)

And to underscore the point, this is content that may well be sourced in the article via a general reference or in another paragraph. It is verifiable. It may well be sourced. It's just primarily lacking an inline source in close proximity. — Carter (Tcr25) (talk) 16:59, 25 October 2022 (UTC)

That is a technicality, which is pretty resolvable by existing means. I can give you a concrete example. In the World War II article, there was a statement "Others follow the British historian A. J. P. Taylor, who held that the Sino-Japanese War and war in Europe and its colonies occurred simultaneously, and the two wars became World War II in 1941" that had no inline citation. I was pretty sure that the reference is still in the article, but the reference in this concrete sentence disappeared during copy editing. I added the "cn" tag and explained the problem. [1] I had no time to dig into sources, but I planned to do so by the end of the year. Fortunately, the problem was resolved, and the reference was added. What if I would fail to provide a reference, and somebody deleted the sentence on 1st of January, 2023? That would be a quite legitimate step, but that would urge me (or other users) to go to a library, to find the correct reference and to restore the removed text along with the reference. However, if nobody cares about some piece of text, maybe, it is not that valuable? Paul Siebert (talk) 17:14, 25 October 2022 (UTC)

Maybe it would urge you to, and on an article like WWII, there are 5000 watchers of whom almost 300 visited recent edits. But there are millions of much less-watched articles. I've seen articles with 3 watchers get maintenance visits by someone performing rote tasks. Sometimes those rote tasks include removing content that has been in the article for fifteen years and tagged for lack of inline citations for ten. Valereee (talk) 17:43, 25 October 2022 (UTC)

The concrete example I'm thinking of is the mess that happened with Radio recently where one editor decided to drop a large number of CN tags on the article, then a month later remove nearly 70,000 bytes of material because no-one had added inline citations. And then objected, per BURDEN, to being reverted as part of the adding inline citations process. Carter (Tcr25) (talk) 18:37, 25 October 2022 (UTC)

Hm...it looks like that editor does this frequently. This is really a problem...tagging non-dubious content, then just a few weeks later coming in and deleting everything because no one has hopped to it yet. On the upside, the article did get fixed, which is good, but honestly this behavior is kind of...not ideal. Valereee (talk) 18:52, 25 October 2022 (UTC)

Yes. However, it may be more harmful and dangerous to keep content that is (potentially) not supported by any source than to remove it. Therefore, the editors who have no time to dig into sources, but have some time to remove tagged statements that are staying non-fixed for years are doing pretty valuable job.

Again, the "cn" tag is aimed to draw other users attention to the problem. If noone has taken any action for a reasonable time (for example, a year, depending on how popular the article is), then the next step is removal. That action serves a dual goal. First, it removes a questionable (potentially unverifiavle) content. Second, it draws attention of other editors. In the latter case, the removed content will be restored, but with a citation. Paul Siebert (talk) 17:00, 25 October 2022 (UTC)

Maybe. Many editors who are expert at research are interested in creating and maintaining their own creations. If I come across an uncited article in my wheelhouse that isn't in too terrible shape, I might try to fix it. But if it's an entire article with no inline citations, just a list of references, even within my wheelhouse it's daunting. I just don't think we should be using deadlines for this. Just because I object to someone systematically deleting non-dubious content for being tagged as needing an inline citation for "too long" doesn't mean I want to spend my research time following along behind them on their maintenance journey cleaning up after them on articles I have no interest or expertise in.

Look, I absolutely support requiring any new addition have an inline citation. That should go without saying, and no one should be arguing about it. I'm talking about content that was created fifteen years ago with a list of references and was tagged ten years ago for needing inline citations. Valereee (talk) 17:09, 25 October 2022 (UTC)

Ok. Now I understand you, I didn't see the words " with a list of references". Now I agree, my bad. Paul Siebert (talk) 17:17, 25 October 2022 (UTC)

Even if there wasn't a list of references, a CN/dubious tag doesn't automatically mean "this material needs to be removed", or else the editor adding the tag would just remove it. The fact they are not removing it, but tagging it suggests it should be reviewed rather than outright removed. Waiting until a month later doesn't change much. All that means is that the removing editor is more guilty of not reviewing the content than anyone else since they are the ones who tagged it in the first place. They should be in big trouble for this kind of activity. saying, "oh well, I tagged it, and nobody cared" should not be an excuse for vandalistic blanking. If adding Cn/dubious tags to content is all we need to do to become legitimate blanking vandals, then I think something should be done to prevent that. Huggums537 (talk) 12:58, 26 October 2022 (UTC)

I do stand by what I wrote including it's usefulness and relevance. And I also agreed with you that it is off of the defined topic (although I tried to provide a justification for bringing it up in this conversation in essence that most controversial wp:ver based deletions are of cited material) so I think I should fade out of this thread. Sincerely, North8000 (talk) 15:53, 25 October 2022 (UTC)

North8000, that type disputes are pretty common in the topics that are a focus of my activity. Yes, some users (including myself) sometimes question some sourced content because the cited source does not support it. That is absolutely not unusual. However, from my experience, all disputes of that kind are pretty resolvable using existing rules, exactly as you described in your Proposed version. In reality, when you go to RSN and ask a question about reliability of the source "A", you literally write that some users have a concern that it may not be verifiable that Mary Smith received the XYZ award. If Mary Smith's hater joins the RSN discussion, they would look very stupid by endorsing their ridiculous claim about a sky-is-blue statement. And, it any event, the community will be on your side. Paul Siebert (talk) 16:52, 25 October 2022 (UTC)

@North8000, this doesn't touch on the original problem: content that was created fifteen years ago with a list of references and was tagged ten years ago for needing inline citations that is now being systematically removed for being "unsourced" when it is sourced -- but to a list of references. Valereee (talk) 17:11, 25 October 2022 (UTC)

Valereee, I think the things North8000 was touching on are more relevant than you think since you were just talking about articles with content which is sourced, but just doesn't have the inline citation. The lack of inline citation doesn't make it any less relevant to the ideas North is talking about, but the fact it is referenced does make it more relevant if you follow my meaning. In other words, I don't think age has any bearing on the ideas he was sharing. Huggums537 (talk) 02:02, 26 October 2022 (UTC)

@Huggums537, well, yes and no. Age has no relevance w/re whether something needs a source. It does have relevance to how we prefer that source be provided. Fifteen years ago we were happy to have a list of references. These days if someone writes a perfectly reasonable article with a list of references and no inline citations, it'll likely be moved to draft or a reviewer at AfC will reject it. But my point was that North is suggesting how we should handle someone now adding something unreferenced. I have no objection to anyone challenging that. What I'm objecting to is editors removing content that has been sourced to a list of references for fifteen years. Maybe I'm still misunderstanding you? Valereee (talk) 13:09, 26 October 2022 (UTC)

I think we both agree what you are talking about is a problem with the older stuff, we just maybe disagree it isn't still a problem with new stuff as well, but I understand that it isn't your issue here. Huggums537 (talk) 02:45, 27 October 2022 (UTC)

Here's an example: Ancient Greek art when it was promoted to FA. Zero inline citations. But it was a featured article, I'm sure that even without inline citations it didn't have any dubious material or it wouldn't have been promoted to FA. Now, it's Ancient Greek art, so obviously there were plenty of people over the years who were interested in maintaining, knew were to find sources, etc. But for less heavily-read/heavily-edited articles? If someone had gone in there in 2010 and started dropping cns like I did recently on Sunday roast, we could end up with someone doing maintenance chores simply removing the entire article. (Note that at Sunday Roast, I did remove something I found dubious; I left the stuff that sounded perfectly reasonable but simply needs an inline citation.) Valereee (talk) 17:28, 25 October 2022 (UTC)

Again, if consensus is achieved among editors that those tags are not needed (and I agree that the tags should be removed from Sunday roast), then they should be removed. It seems that is obvious, and if I am not wrong that would be in accordance with our policy. Paul Siebert (talk) 17:49, 25 October 2022 (UTC)

But all that content does need an inline citation. So how do we indicate information that is perfectly reasonable but needs an inline citation? Just use a banner for it? I kind of hate to do that when there's only one or two sentences that lack an inline citation, it seems like overkill. Valereee (talk) 18:02, 25 October 2022 (UTC)

Also how do we get people to change from dropping cns when they see content lacking an inline citation? I would think that would be an even more contentious discussion than this! :D Valereee (talk) 18:04, 25 October 2022 (UTC)

The main problem of this policy (and Wikipedia in general) is that it combines such articles as Sunday roast and Quantum entanglement, and we are trying to make the policy is equally applicable to all of them.

Of course, the former article should not be written using the same approach as the latter. If the article discusses cooking, and its bibliography has just four items, I am not sure anybody in clear mind would care which concrete source (out of four) supports the statement "Mashed potatoes are also a frequent accompaniment." (By the way, another reason to remove "cn" in this particular case is that no references are needed in a lede, so it should be removed per MoS.)

In contrast, almost every statement is expected to be supported by an inline citation is such article as The Holocaust, because the topic is very sensitive, and it is vulnerable to misinterpretations, edit warring and even direct falsifications.

I think we need to decide how can we make our policy more topic and case specific. Paul Siebert (talk) 18:12, 25 October 2022 (UTC)

Except that almost none of that stuff in the lead is cited in the sections, either. Which I guess is a different tag, though lol... Valereee (talk) 18:32, 25 October 2022 (UTC)

The structural policy basis for requiring a cite (via tagging or removal) is that the material has been challenged. Simply (re)define what is required to constitute a "challenge". North8000 (talk) 18:25, 25 October 2022 (UTC)

So basically...if we see something has been removed for having been tagged cn too long, maybe the answer is to restore without the tag and with an edit summary of "probably didn't require a tag", then just remove all the non-dubious cns and banner the article.

It still doesn't fix that problem of the fact that probably most of the time this is happening, it's happening on underwatched articles and no one is noticing. Valereee (talk) 18:34, 25 October 2022 (UTC)

For that to happen, BURDEN would need to be adjusted to make clear that 1) a CN tag in and of itself is not a challenge to verifiability and 2) that without a specific/valid challenge at the time of removal, content can be restored without an inline source. — Carter (Tcr25) (talk) 18:45, 25 October 2022 (UTC)

Yep. And there are a lot of people pushing back on that here, saying 'BURDEN is fine as it is' and that the way folks are interpreting old cns at articles where there is an existing list of references -- as a definite challenge to the content itself rather than a request for an inline citation -- is what's intended by it. Valereee (talk) 18:59, 25 October 2022 (UTC)

This would be to say what is required to count as a "challenge". That isn't a real change in burden. North8000 (talk) 19:40, 25 October 2022 (UTC)

How would you word a change? Literally I mean current wording > proposed wording. Valereee (talk) 19:47, 25 October 2022 (UTC)

Like maybe:

From: Any material that needs an inline citation but does not have one may be removed.

To:Any material that needs an inline citation but does not have one may be removed; however, the simple fact someone has tagged a statement for "citation needed" does not mean they are challenging the content, but may simply mean they're asking for a citation.

Or, otherwise:

To: Any material that needs an inline citation but does not have one may be removed. Statements that are tagged as "citation needed" are assumed to be challenged content and may be removed at any time.

Maybe that's what we need? A clarification that this is indeed what we have consensus for -- that a cn tag shouldn't be used unless you're challenging the verifiability of the content. Valereee (talk) 19:54, 25 October 2022 (UTC)

I would say:

"Any material that needs an inline citation but does not have one may be removed. Statements that are tagged as "citation needed" are assumed to be challenged content and may be removed if the tag has stayed for a reasonably long time, and no action has been taken to fix the problem."

Paul Siebert (talk) 22:37, 25 October 2022 (UTC)

I think one that needs to change is bullet point #3: all material that is likely to be challenged, because it is the odd man out that is causing a lot of trouble. bullet points 1, 3, and 4 are absolutely clear cut points of things that mandatorily need inline citations with bullet point #2 being absolutely clear that any material whose verifiability has been challenged must need an inline citation. Bullet point #3 should be consistent with this and say: all material whose verifiability is likely to be challenged. The way bullet point #3 currently reads makes it look like we might as well just interpret bullet #2 as saying: "any material that has been challenged" even though that isn't what it says at at all. Huggums537 (talk) 02:19, 26 October 2022 (UTC)

I'd support that clarification here. We'd need to also clarify at Template:Citation_needed#When_not_to_use_this_template that a cn tag can be interpreted by any other editor as a challenge to the content and an indication that the content can be removed by anyone after any reasonable period of time. Valereee (talk) 13:21, 26 October 2022 (UTC)

Requiring people to say they think the text is not "verifiable" is unnecessary bureaucracy and could be used to revert removals of false material. Most people are not aware of Wikipedia's idiosyncratic use of "verifiable". They might remove something unsourced saying it's "incorrect", but under this proposal they could be reverted by someone else who thinks the removed content is fine because it's been there so long.

Still not seeing any reason to change this in favor of unsourced material, and once again the discussion is becoming too long to keep up with and scaring off people. Crossroads ^-talk- 00:03, 26 October 2022 (UTC)

In your definitions here, is everything without an inline citation to be considered "unsourced material"? The problem in my mind is that material may lack a nearby inline citation, but it may be SKYISBLUE, supported by general sources, or sourced elsewhere on the page. Asking someone to state they think the text is not verifiable may be bureaucratic, but if means someone has to think about what they're looking to remove, I don't see that as a bad thing. So long as BURDEN leaves editors feeling they need not do more than look for a footnote before removing content, it isn't going to ensure that bad material is removed or that what remains is actually well sourced. — Carter (Tcr25) (talk) 01:45, 26 October 2022 (UTC)

"General sources" and sources elsewhere on the page are as good as sources not cited at all, because the wikitext is changeable and text-source integrity is easily destroyed. I don't really care about the "BLUESKY" essay, because (1) it applies to only an extremely tiny minority of material, (2) it is trivial to just source BLUESKY stuff anyway, and (3) BLUESKY gets invoked for claims that are contentious and nothing of the sort simply because someone else is convinced of their own beliefs. Of course BURDEN doesn't ensure all bad material is removed; it is a bare minimum, literally the least we can do. Crossroads ^-talk- 20:01, 26 October 2022 (UTC)

Except general sources were the standard in the past. I get that's not the case now, but what's the best way to bring things up to standard? Cut anything that doesn't appear sourced (hopefully rewriting the section/article to make sure what was excised doesn't impact the rest of the article) with the presumption that if the material cut was important or valuable someone will add it back in with a source? That doesn't seem collaborative or an effective way to build a better encyclopedia. Each of the cases that brought me to this point were instances where some basic work by the removing editor should have been able, if not to provide a proper source, at least determine that the information was correct even if a source speaking to each of the details wasn't quickly found. And in most of those cases, the content left after the cut was disjointed, made reference to the cut material, or otherwise left behind an article that was in worse shape than before. I see others worrying that any change to BURDEN might make it too hard to excise wrong or misleading information, but that's not the problem I've seen happening. If there's a better way to encourage editors to evaluate material they want to remove (excepting, of course, obvious bunk or BLP violations), just as they're required to evaluate material they add, I'd love to see it. — Carter (Tcr25) (talk) 20:34, 26 October 2022 (UTC)

Policy is that if challenged or likely to be challenged it needs a cite. "Likely to be challenged" is good guidance but moot regarding implementation. So regarding implementation it just for challenged material. This is just "thinking paper" not final wording but....

"Uncited challenged material may be tagged or removed. A simple statement of concern that the material may not be verifiable is sufficient to constitute a challenge" This softly says that tagging or removal requires a challenge, and sets (via example) a (low) bar on what is needed to be a challenge. North8000 (talk) 20:41, 25 October 2022 (UTC)

For me that doesn't clarify whether or not a 'citation needed' tag is a statement of concern. Valereee (talk) 20:44, 25 October 2022 (UTC)

The context here is that "Statement of concern" is the challenge. It's a bit soft/indirect, but it says that (from a wiki-structural standpoint) a cn tag requires a challenge/statement of concern, so the tag itself does not count as the challenge.North8000 (talk) 20:50, 25 October 2022 (UTC)

So this is a requirement for such a statement of concern via an edit summary or a statement on talk? Valereee (talk) 21:24, 25 October 2022 (UTC)

As it stands, BURDEN does state When tagging or removing material for lacking an inline citation, please state your concern that it may not be possible to find a published reliable source, and the material therefore may not be verifiable.^[5] If you think the material is verifiable, you are encouraged to provide an inline citation yourself before considering whether to remove or tag it. and part of that footnote #5 states Also, check to see whether the material is sourced to a citation elsewhere on the page. For all these reasons, it is advisable to clearly communicate that you have a considered reason to believe the material in question cannot be verified. Would this soft/indirect approach add any oomph to those existing statements, which are often ignored because BURDEN clearly gives permission ("may") to remove text that appears uncited? — Carter (Tcr25) (talk) 21:56, 25 October 2022 (UTC)

From what I've seen, no. People are reading "may". They stop there. Valereee (talk) 22:14, 25 October 2022 (UTC)

As I already said, "challenged" or "likely to be challenged" can be interpreted in two different ways:

"I think that is may me not true", and "I think others may decide that is dubious", accordingly, or
"I think that a reader may be interested to see the source that says so"

I prefer the second interpretation, for, IMO, the main value of Wikipedia is not the content per se, but the citations that one can find in its articles. Without references to RS, Wikipedia is not more valuable than any popular blog or Facebook page: it is just an interesting reading of totally unknown quality. Paul Siebert (talk) 22:53, 25 October 2022 (UTC)

I endorse this. Donald Albury 23:20, 25 October 2022 (UTC)

I'm absolutely against any wording of policy that forces sourcing on obvious BLUESKY content just because somebody else decided "they" thought maybe some reader "might" be interested in seeing a source that absolutely for sure says so or else "they" are taking your content out. There was a weak attempt to make adding unsourced content to seem as disruptive as vandals who blank content, but that is comparing apples to oranges because as North has pointed out before, if someone removes unsourced content with an edit summary that it was "disruptive", then they run an extremely high risk of looking very foolish, and getting into serious trouble for assuming bad faith or being incompetent if it turns out the content was verifiable or cited elsewhere in the article. OTH, if I revert someone for blanking with a rationale that it was disruptive, or vandalistic, then I have almost no risk of looking foolish or being accused of bad faith. Nobody is ever going to accuse anyone of being incompetent for trying to prevent vandalism. Huggums537 (talk) 03:33, 26 October 2022 (UTC)

In other words, you will be held accountable, and I wont. There's good reason for this. If I revert a blanking vandal, then others have argued that I should stop to consider the good faith chance that maybe the vandal had good reason for blanking even if they didn't leave an edit summary. The problem with this is that it fails to take into account the good faith possibility that the person who did the right thing adding the content, and left an edit summary, and added a source (or not as the case may be) also might have had a good reason for adding the content (like maybe the source is in another section). The question then becomes: whose good faith shall I consider over the other? I would not be faulted for whatever I decide. OTOH, if you remove unsourced content as "disruptive", you will be held accountable, and expected to explain why you showed bad faith to the adding editor because there is no other editor to answer to, explain your actions away from, or say you had to choose between. See what I mean? I can say I had to make a choice between giving good faith to this editor or that one, and you can't make that choice so you either have to answer for your bad faith or just assume good faith. The system is rigged for inclusionism. Some people would be happier if they just accept it. Huggums537 (talk) 04:34, 26 October 2022 (UTC)

Of course, this dynamic changes a little bit if a CN tag is involved. If I revert a blanking vandal, it still comes down to choosing between the good faith of other editors where two have correctly edited adding content and tags, and one has not correctly edited by providing an adequate summary, so I have to quickly decide if they are doing harm or not. I might not notice or even care that a CN tag is present in the blanked content if the vandal fails to leave a summary because that could mean many things other than "this content should be removed", so it would play little to no part whatsoever in the decision, and I would choose to consider the good faith of the editors who added the content, and the tag in the correct manner. That means the edit that was not done correctly automatically gets rejected by me, and the tag is really irrelevant since it could mean many things other than "this content needs to be removed". Now, when it comes to an editor blanking content tagged as CN with an edit summary of "disruptive"? Well, the dynamic changes a little bit for them too. They still have to be accountable for their bad faith to the adding editor, except now they will have to try to convince someone that the tag was there as evidence the content was just disruptive as opposed to the tag being there as evidence the content was verifiable, but just needed a citation. As N8000 might say, good luck trying not to look like a fool on that one. Huggums537 (talk) 09:38, 26 October 2022 (UTC)

So say this is the series of events:

1. Content was written years ago, sourced to a list of references but with no inline citations.

2. At some point, someone comes along and tags anything that doesn't have an inline citation, but not stating any actual concern that the material may not be verifiable.

3. Someone doing routine maintenance comes along, sees the tag, and removes the content.

Can someone else come back in, check the original placement of that tag, see that it doesn't have a statement of concern re:verifiability, disagree that the content is likely unverifiable, and restore the content sans tag as likely verifiable content that simply doesn't have an incline citation? Valereee (talk) 13:38, 26 October 2022 (UTC)

I don't see why not. Probably better to add the citation, but if you have a reason to restore with the tag, and explain it, then I think it would be just fine and dandy. Huggums537 (talk) 22:57, 27 October 2022 (UTC)

It's been noted that the situation that I have most in mind (POV warriors using this in tandem with nitpicking the source on sourced material) is off topic. Once we set that aside, I think that the common problem cases for controversial tagging/ removal of unsourced material are:

As a volley in a personal battle between editors
Someone being being overly pedantic, OCD or who doesn't understand the big fuzzy Wikipedia system and thinks that going overboard interpreting/"enforcing" a policy is doing the right thing.

I see a few things possibly alluded to above which I think are not a good idea:

Prescriptive statements about what should happen with the content based on only on one consideration. These do not take into account the other factors (variables) that are also taken into consideration.
Possibly make a different set of rules for "sky is blue" statements. IMO this would weaken wp:Ver too much

IMO the idea that I presented above (reiterating that tagging or removal requires an easy-to-do but explicit challenge) would help in all of the above situations. North8000 (talk) 16:56, 26 October 2022 (UTC)

In the cases that I've encountered, it seems to be much more #2 than #1. BURDEN is being understood as if there is no proximate inline source, then the content is fair game to remove as they wish (tagged or not). The explicit "the material may be removed" is being followed by some editors with the understanding the person who removes the material is under no obligation to even do a simple Google search. They don't need to do anything. I don't think that requiring an explicit challenge alone will fix that problem, in part because editors are making the assumption that the lack of an inline source is evidence of WP:OR or some other bad-faith insertion. If the explicit challenge includes some obligation to engage in a WP:BEFORE-like attempt to verify (as is currently suggested in BURDEN, but not required), then I would feel it would be more likely to help deal with the problem. I don't want to see WP:V weakened, but it's being abused to remove text on the basis that it's unsourced or unverifiable without actually checking first. — Carter (Tcr25) (talk) 17:54, 26 October 2022 (UTC)

Than goes a lot farther than I'm advocating. North8000 (talk) 18:33, 26 October 2022 (UTC)

That's not "being abused", that's the system working as intended. People should be citing their sources to support their assertions. Some material is grandfathered in very weakly, in that it exists, but it can be challenged easily at any time because it could be false or misleading. The whole point is that the WP:BURDEN of proof lies on the person making a claim - the person who wrote the text or those defending it later - not the person challenging it. That's not a Wikipedia thing, by the way - it's widely the case that the burden of proof for claims lies on those making them.

Any amount of shifting the burden of proof away from those supporting a claim is completely inappropriate and out of step with the rest of the world. Crossroads ^-talk- 19:52, 26 October 2022 (UTC)

Disagree. It's not working. If you claim something I've said is false, then yes I have an obligation to point to sources that back it up. But there's also a reasonable expectation that you can tell me why you think the claim is false. In a real-time conversation, that can all happen pretty quickly because we're both there ... which is similar to a new addition to an article that's provided without a source and then challenged. Presumably an editor who adds something today will be around later that week to deal with questions raised. But if you're reading something in an older book, there might not be a footnote next to something that strikes you as odd. So do you immediately decide it's false or misleading? Or do you look for more information before making that determination? That's more similar to the issue with older content that lacks a proximate inline source. There should be a reasonable expectation that someone is challenging content because they have determined it more likely than not is false or misleading, not because it could be false or misleading. It's not shifting the burden of proof; it's asking that someone explain why they're challenging things. That is the case in the real world and should be here too. — Carter (Tcr25) (talk) 20:48, 26 October 2022 (UTC)

+1 Valereee (talk) 21:29, 26 October 2022 (UTC)

+1 Except that it does not necessarily conflict with what Crossroads wrote. The latter post merely requires a good faith explicit expression of question or concern of the veracity (or verifiability) of the item as a prelude to the process that Crossroads describes. North8000 (talk) 22:07, 26 October 2022 (UTC)

Wikipedia isn't like a published book, it's a site where literally anyone can insert anything into a page at any time. It's also a site that's both enormous and underscrutinized. Therefore, Hitchens's razor applies in spades: "what can be asserted without evidence can also be dismissed without evidence." Which is of course a simple statement of the burden of proof.

Your statement that It's not shifting the burden of proof doesn't make it so. Putting any bureaucratic limitation whatsoever on challenging content is taking burden away from those asserting the content is good and placing some of it on the challenger.

Common real-world scenario: A reader is looking at an obscure article on a topic they have some familiarity with and they realize a claim in it with no citation that they know is false. They click "edit" and remove it, because 'anyone can edit', but don't leave an edit summary, being unfamiliar with them and not knowing the new rule that they have to specifically say they think no source on the planet supports it. A Recent Changes patroller who does know the rule sees an IP address and a bright red negative number, and reverts because they didn't say the magic words. Misinformation remains in Wikipedia.

This is not an improvement to the encyclopedia. This feels like an extension of the "more is better" fallacy that sometimes drives an overemphasis on number of articles over quality. Crossroads ^-talk- 22:20, 26 October 2022 (UTC)

+1 Donald Albury 23:15, 26 October 2022 (UTC)

@Crossroads, I can tell you that I myself have tagged hundreds of statements simply for not having an inline citation. Not because I thought they weren't verifiable, only because I wanted to call someone's attention to the fact we needed to cite inline. I won't be doing that any more, but there it is: for me, until this discussion, I did not realize other editors were perceiving this as a challenge to the content rather than a request for an inline citation. Valereee (talk) 23:43, 26 October 2022 (UTC)

I myself am wondering why Hitchens's razor is not a part of our policy. Indeed, everything that was added without a proof may be removed without explanation. +1. Paul Siebert (talk) 23:48, 26 October 2022 (UTC)

Because in many cases the person who first wrote the content is no longer around to explain where they got it from.

Because in many cases the source is there already. It's just not cited inline. Valereee (talk) 00:09, 27 October 2022 (UTC)

Exactly. And that is why, when I write some text, I always supplement it with an online citation to a good quality source, so everybody will be able to easily verify it by comparing the text with the source cited. And that makes my contribution verifiable.

I may disappear from Wikipedia (actually, I already did that), but my contributions will stay, because the users who look at texts written by me don't need to guess if those texts are verifiable. Instead, they can open a concrete source (the source I cited) and compare the text with the source.

And that is what we should call "verifiability". Paul Siebert (talk) 03:16, 27 October 2022 (UTC)

Another common real-world scenario: Someone blanks a few sentences from a paragraph because they think it is funny to do so. No edit summary explaining the removal. There's a footnote at the end of the paragraph that remains, but the text removed seems only tangentially connected. How does the Recent Changes patroller determine if the unexplained removal is vandalism or a valid challenge? If they decide it looks suspicious and should be reverted, do they need to add an additional inline citation? What if the reviewer is familiar with the topic, so they know it was vandalism when the content was correct. Can they revert without adding an inline source? What if the removal leaves a paragraph that is agrammatical or otherwise leaves the article disjointed or not making sense? As currently stated, BURDEN explicitly requires restoration with an inline source. Someone might make a I know [vandalism] when I see it argument, but it could quickly devolve into a revert war with BURDEN being cited on the site of removal.

Hitchens's razor may be a fine axiom for philosophical or theological debates, but it's about dismissing a statement that relies solely upon faith and can only be true if you believe in it. But it's not a blunt rule that applies in every situation. For example, a statement like AM (amplitude modulation) – in an AM transmitter, the amplitude (strength) of the radio carrier wave is varied by the modulation signal (removed with this edit) isn't about faith or even technical expertise; it's a practically definitional statement and it should be evident that it is verifiable even if only by a general reference down the page. This isn't about "more is better"; it's about encouraging better editing. — Carter (Tcr25) (talk) 02:45, 27 October 2022 (UTC)

Plus, another reason why Hitchens's razor is not part of policy is because it is an exceptional claim itself. Claiming that unsourced content is "an assertion without evidence" is an exceptional claim that requires the claimant to provide the burden of proof to show that there is "no evidence" for the assertion which is easy to do in Wikipedia terms since all you have to do is show there are sources for the assertion or not, so Hitchens's razor is just a stupid ass philosophical trick for blanking vandals to try to say unsourced content is an assertion "without any evidence" so they can have an excuse to "dismiss" any unsourced content they want without any evidence. Don't fall for it kids. Stay in school, and don't do drugs. Huggums537 (talk) 03:14, 27 October 2022 (UTC)

"Claiming that unsourced content is "an assertion without evidence" is an exceptional claim" Really?

A major part of activity of many established Wikipedia editors consists in removal of various assertions that are being added to Wikipedia without any evidence. That is a part of our work here. And such cases are by no means an exception, that is what is happening every day.

Remember, we all are considered amateurs, we have no obligations and no personal responsibility for what we are writing (the only risk we may face is to lose our account). Furthermore, Wikipedia policy allows us to use pretty lousy sources (which are unacceptable per scholarly and scientific standards). And in that situation you declare that the claim that some Wikipedian may put into Wikipedia an absolutely unsubstantiated assertion is "exceptional"? I would say, it is yours claim that is exceptional, and it needs an exceptional evidence. Paul Siebert (talk) 15:25, 27 October 2022 (UTC)

I think one of you is talking about "evidence at the end of the sentence in the Wikipedia article" and the other is not thinking so narrowly. There is evidence that the Sun is big, even when there is no little blue clicky number anywhere on the page. WhatamIdoing (talk) 16:24, 27 October 2022 (UTC)

See my responce to Blueboar in the below section. We are assuming majority of us apply common sense when editing Wikipedia, and that my claim is by no means exceptional :). Paul Siebert (talk) 16:39, 27 October 2022 (UTC)

Agree with Crossroads. Donald Albury 23:16, 26 October 2022 (UTC)

BURDEN - arbitrary break 2

Before we can resolve this, I think we need to reach a consensus on a very basic philosophical question: Do we (as a community) think the removal of uncited information is (at its core) “a GOOD thing”, or “a BAD thing”? Blueboar (talk) 13:58, 27 October 2022 (UTC)
As you've asked the question, it depends upon what the uncited information is. BLP violation, of course. Obvious bunk, yes. Something that is basic knowledge, no. Something that sounds odd, maybe, but I'd want to do some investigation first. There are times it's a GOOD thing and times when it's a BAD thing (or at least a less-than-ideal thing).

I'd rephrase the question to remove that ambiguity to focus on what I see as the core problem: Do we (as a community) think the removal of ~~uncited information~~ information primarily because it lacks a proximate inline citation to a reliable source that supports the statement is (at its core) “a GOOD thing”, or “a BAD thing”? — Carter (Tcr25) (talk) 14:16, 27 October 2022 (UTC)
Your rephrasing is incorrect, because the answer is obvious: it is a BAD thing.

Indeed, if the only reason for removal is WP:V, the user who removes the statement it is supposed to check if other sources in the same section support the information that is seemingly unsourced. If that user has no time (or no access to the cited sources), they place the "cn" tag to inform other users about a potential verifiability problem. If no action has been taken in a reasonable time, the tagged statement may (and, I would say, should) be removed. However, the action may include both replacement of the "cn" tag with an online citation and removal of the tag (if other users achieved a consensus that the tag was placed inappropriately).

Actually, during this discussion (see above) I already provided a real life example of correct usage of the "cn" tag. Paul Siebert (talk) 15:49, 27 October 2022 (UTC)
I agree that they're supposed to check. BURDEN suggests it, but, as Jayron32 noted earlier in all this: Notice, the person who removes the material is under no obligation to even do a simple Google search. They don't need to do anything. That is, in fact, the policy.. That's what I think is out of balance and why I tried to rephrase the question as I did. Some editors do see such removal as a good thing because it can act as a prod to increase inline citations. With new additions, I'd probably agree in a lot of instances even if I think investigating the question. But with older content and when removal is done on a largescale (e.g., what happened at Radio) and when the assumption is that a CN tag was added to challenge the statement, that I'm much less sure of. — Carter (Tcr25) (talk) 16:22, 27 October 2022 (UTC)
I agree that Jayron32's statement is out of balance. I would rephrase it as follows:
If the "cn" tag was placed after some statement, and nobody took any action, then, after some reasonably long period of time, anybody can remove it. The person who removes the material is under no obligation to even do a simple Google search.

I think that would be correct, because if the "cn" tag is ignored in some article, that demonstrates that the article is being ignored by currently active Wikipedians. That means that article may contain some information that was placed there many years ago, and that information may be not properly verified (just because nobody cares about that). That may undermine credibility of Wikipedia, so it would be better to remove that content than to allow it to stay. Paul Siebert (talk) 16:47, 27 October 2022 (UTC)

BTW, what happened at Radio? Paul Siebert (talk) 16:49, 27 October 2022 (UTC)
I'd be more supportive of your rephrasing, although what constitutes a "reasonably long period of time" is something that all like draw a good bit of disagreement. More guidance around WP:CITENEED might also be a good thing.

RE: Radio — Nearly 70,000 bytes of content removed en masse a month after the same editor placed CN tags. Removal was justified as per WP:V, WP:NOR, WP:CS, WP:NOR, WP:IRS, WP:PSTS, et al. based on a lack of proximate inline sources. Then there was back and forth reverts/removes while some editors tried to add sources based on BURDEN's ... should not be restored without an inline citation to a reliable source. It ended up at ANI with some support for a maximal interpretation of BURDEN both in terms of removal based on the lack of inline sources and restoration only allowed piece by piece and other support for a taking a more measured approach toward material lacking inline citations and allowing CN-flagged content to remain in main while sources are added. — Carter (Tcr25) (talk) 17:30, 27 October 2022 (UTC)

Blueboar, as soon as you asked a very general question, here is a general answer:
"We all are considered "Wikipedia editors", therefore, both removal and adding of information is what we are supposed to do. In general, both removal or adding of any information is neither bad or good, everything depends on context."

Actually, Wikipedia cannot exist without non-controversial unsourced statements. They serve as a glue that holds all pieces of each article together. If you remove them, almost any article will fall apart.
I propose not to continue this subsection, because it distracts us from much more narrow and concrete question: is removal of the content with the "cn" tag is generally legitimate, and what should we do when the tag has been used inappropriately? Paul Siebert (talk) 15:14, 27 October 2022 (UTC)
It sounds like frustration is building with editors who have a behavioral pattern of:
Tagging all uncited sentences/paragraphs

Waiting a little while

Blanking all tagged sentences/paragraphs (plus maybe some stuff that wasn't tagged, like pictures).
It sounds like the behavior we want to see is:
If you think that actually needs a citation, then WP:SOFIXIT already. Don't expect others to jump in to do work that you refuse to do yourself.

If, and only if, you actually can't find any barely passable source, after a non-trivial and good-faith effort, and it requires an inline citation, then tag it ("This is a type of ___.^[fact]").

If you tag more than one or two claims in an article, let someone else, ideally someone familiar with the subject area, decide whether that article is better off containing the content you tagged or better off containing nothing.

If you come across claims tagged by someone else, and you attempted to find a source for it, you can decide whether the article is better off containing the tagged content or better off containing nothing. You may not engage in quid-pro-quo article blanking, in which you tag all of Article A and Bob blanks it for you, and Bob tags all of Article B and you blank it for him.
WhatamIdoing (talk) 16:44, 27 October 2022 (UTC)
The behavioural pattern that you described is called WP:GAMING. It is already considered disruptive, so I see no reason to discuss it here. Paul Siebert (talk) 16:52, 27 October 2022 (UTC)
The people who engage in this do not agree with you. They do not agree that they are editing "to thwart the aims of Wikipedia", to quote that guideline. In fact, they seem to believe that they are editing to support the aims of Wikipedia – by getting rid of uncited text. WhatamIdoing (talk) 05:55, 31 October 2022 (UTC)

A few points:

A common problem which leads to endless discussions is attempting to be prescriptive / dictate what is supposed to happen based on only one consideration. In reality there are almost always widely varying considerations and multiple policies or guidelines influencing the result. Wikipedia:How editing decisions are made
We always talk as if WP:Gaming and WP:Wikilawyering are rare and serious occurrences. In fact, mild versions of them are both mild and common. For example, to use an unusually literal and categorical application of a policy or guideline to knock out material to tilt the article towards a POV. So we need to plan for these routinely occurring.
We are pretending that the common problematic application of this is for totally unsourced material, but in reality the most common problematical application of burden is in tandem with nitpicking a provided source.

So it's complicated. IMO my idea of requiring an actual challenge (and even an expression of concern about the verifiability of the material would be sufficient) vs an implicit challenge acknowledges all of these considerations. Or, more simply, since policy already in essence says that a challenge is required, we simply need to say that implicit challenges don't count. North8000 (talk) 19:17, 31 October 2022 (UTC)

BURDEN - arbitrary break 3

Throughout this discussion I've been pushing very hard to convince people about the concept that the mere act of removals and tagging in and of themselves alone can and should not be considered valid challenges. In the case of content removal, I've done this by repeatedly comparing content removal to vandalism, and making the argument that says if call mere content removal a "valid challenge", then that is the equivalent of saying vandals who blank content have made a "valid challenge" by merely removing content. This has not impressed anyone, especially the more experienced editors who seem to have heard this argument before. It also became apparent to me I was not the first one who came up with this concept when we started talking about MINREF in this thread, and I realized somebody had the idea before because I read this at MINREF: For example, section blanking may be considered vandalism, rather than a demand for inline citations. However, I have been doing some thinking, and I believe I have an idea about this concept that maybe nobody has thought about before, but they probably have and I just don't know - we shall see. To begin, we are going to have to ask ourselves some questions: What is a challenge? and, What makes it valid?

What is a challenge? Well, if it's defined as something similar to: "anything at all that I dispute" or "anything at all that I remove", then it becomes very easy to see why someone might believe the idea that even if you are a vandal who blanks material, and no matter if you were wrong or right, blocked or not, your removal counts as a valid challenge, because who knows, maybe the vandal had some unknown dispute (like a grudge), or maybe they thought they had some good reason to remove, but really didn't so just the removal counts even though it was suboptimal on the helpful level so vandals get a pass (at least in terms of what a challenge is). Same goes for tags. The second I put a tag on something it is disputed no matter what the reason is, so it is now "challenged", making it a "valid challenge".

But, what if we define a challenge as: "anything whose verifiability has been or is likely to be challenged"? This is actually how it is supposed to be defined under BURDEN. Suddenly, vandal blanking doesn't make much sense as a challenge any more, much less as a "valid challenge" because the act of blanking does not specifically dispute the verifiability of anything. The simple act of removing material does not dispute verifiability at all. For example, if someone accidentally types types the same same word twice twice, and I go in to remove the duplicates, shall we say that the mere act of removing these duplicates is a "valid challenge" simply because I removed them, therefore it shall be the equivalent to a dispute of verifiability? Of course not. The same goes for tags as it has been explained before. If someone sees an article with general references that needs inline citations, but doesn't question the verifiability, and tags it with a Cn tag, that doesn't mean a "valid challenge" has been issued simply because the act of placing a tag has occurred. The verifiability has not been disputed, and that is the difference between thinking that the actions themselves make the challenge (such as the removal itself or the tagging itself) or the dispute of verifiability being the thing that makes the challenge. That is whole key. I think many editors are very confused between the regular meaning of the word "challenge" as it applies to simple ordinary disputes, and the more specialized meaning of the word "challenge" as it applies to BURDEN verifiability disputes. We really should distinguish the BURDEN challenge from other average types of challenges. Otherwise, how will we know the difference? I mean there are lots of types of challenges to talk about on Wikipedia. Like, you could challenge a close, or challenge any kind of decision, so distinguishing the BURDEN challenge would be very helpful. Huggums537 (talk) 13:33, 3 November 2022 (UTC)

As I have said before, I don’t think any of this matters, because the remedy to a removal that you disagree with is to return with a citation. Even if the motivation behind the removal was to vandalize the article, the remedy is the same: return with citation. If you want to Wikilawyer about whether the removal was “valid” or not, fine… do so AFTER you return with citation. Blueboar (talk) 14:10, 3 November 2022 (UTC)
So the remedy to the removal of duplicate wording (I naturally disagree with duplicate wording) is to return with a citation? That doesn't even make any sense. I don't think you have been paying any attention at all. Huggums537 (talk) 15:56, 3 November 2022 (UTC)
Inappropriate duplication of words shouldn't be returned, so naturally a sensible editor wouldn't want to do that. But if those words belong, and have anything to do with content instead of grammar (e.g., all the people named John John), then slap a source on it when you put it back
If the duplication is purely a matter of grammar – What the trouble is is that you keep looking for simple answers; I had had a story to tell you – then you shouldn't be worrying about this policy at all. WhatamIdoing (talk) 22:52, 3 November 2022 (UTC)
Well, that's very convenient isn't it? Any removal of content is a valid challenge except when it is related to grammar. The only problem with that is that this imaginary "rule" about "valid challenges" doesn't appear to have any exceptions outlined in policy for it either. Huggums537 (talk) 07:07, 4 November 2022 (UTC)
"Fixing grammar" is not "removing material", even if fixing the grammar results in having fewer words on the page. These two sentences contain the same material:
Methodological observation of the sociometrical behavior tendencies of prematurated isolates indicates that a casual relationship exists between groundward tropism and lachrimatory, or 'crying,' behavior forms.

Children cry when they fall down.

If you replace the first with the second, you remove no material whatsoever. WhatamIdoing (talk) 21:11, 4 November 2022 (UTC)

If any of you are going to have the unmitigated gall to posit that any and all content removal is equivalent to a verifiability challenge without stopping for one second to think about the fact that maybe there are exceptions such as obvious vandal blanking, grammar/typo corrections, and untold numerous content removals during routine maintenance tasks such as the removal of external links, lint errors, and many others, then the best thing to do would be to embody the so called rule into a policy that restricts the application of it to verifiable content only so you don't have to name the plethora of exceptions that would be required for such a non existent rule. Huggums537 (talk) 07:26, 4 November 2022 (UTC)
A "citation needed" tag means "Prove it or remove it" (and placing one is inherently a "challenge"). If no one has done the first after a reasonable period of time, it comes time to do the second. And "reasonable period of time" is intentionally not defined—unreferenced material is subject to immediate removal; there is no obligation to do the "Citation needed" at all. If you don't want that to happen, cite a source to begin with. There is no problem with BURDEN; the problem is people not citing sources and then complaining when people challenge and remove the uncited material instead of just providing a citation. That also applies to people who like to shout "BLUESKY"—if someone is questioning it, clearly it's not as obvious as you think it is, and regardless, if it's that well known, just source it. I might think it's "BLUESKY" that water is composed of two hydrogen and one oxygen atoms, but if someone were to ever challenge that, it would be trivial to source that assertion. Seraphimblade ^{Talk to me} 14:20, 3 November 2022 (UTC)
Yes and no. From the CN tag's documentation: The {{Citation needed}} template is intended for use when there is a general question of the verifiability of a statement, or when an editor believes that a reference verifying the statement should be provided. Other templates are available for other or more specific issues; see the list of inline templates. For example, claims that you think are incorrect should be tagged with {{Dubious}}, and those which represent a non-neutral view should be tagged with {{POV statement}}. Being specific about the nature of the problem will help other editors correct it. Editors clearly interpret when to place a CN tag and what one's presence means in different ways, as Valereee has noted in this discussion. I generally see a CN tag as a request, not as a challenge or demand and certainly not as a threat or as a justifiable reason to remove something without some degree of research/investigation. — Carter (Tcr25) (talk) 14:48, 3 November 2022 (UTC)

Then perhaps we need to change the documentation language at the cn tag page - so that it better reflects how we actually use it (rather than how we wish it were used). Blueboar (talk) 14:57, 3 November 2022 (UTC)

Sure, if we can determine how we (as a community) actually use it, rather than how some of us use it. Reading through all this discussion, it's clear there are differing expectations and experiences as to when a CN tag should be placed, how it should be responded to, what level of investigation is expected, and how long they should linger without action. — Carter (Tcr25) (talk) 15:36, 3 November 2022 (UTC)

I think the documentation is correct. Not everybody uses that tag to say "Prove it or remove it". I don't always. WhatamIdoing (talk) 22:54, 3 November 2022 (UTC)

Yes, but some do interpret it that way (see the original post by me that started this whole convo), and they back themselves up with BURDEN. Valereee (talk) 01:49, 4 November 2022 (UTC)

I would say absolutely not. How does the Cn tag mean "Prove it or remove it" in the example I gave above? The answer is that it does not. If it does not in that example, then it does not always mean that. If it does not always mean that, then it can not "inherently" be a challenge if it also has the possibility of taking on other meanings such as simply being an indicator of improvements needed to the referencing format of the article. Nowhere does placing a Cn tag mean "prove it or remove it", and nowhere should it ever mean that either. If I place Cn tags in an article that say, "this article has verifiable general references, but they should be moved to inline citations for easier identification", then what part of the placement of that tag is "inherently" a challenge to "prove it" or remove it? I would really like to know because that is a rather extraordinary claim. Huggums537 (talk) 16:17, 3 November 2022 (UTC)

I don't think anybody has been paying any attention. I'm about to take my toys out of the sandbox and go home. Nobody is playing right. Huggums537 (talk) 16:40, 3 November 2022 (UTC)

I sympathize with your frustration, having made a fair number of suggestions in my lifetime that never garnered any real attention from others, but there also comes a point where if "nobody" is taking your side then you either need to WP:LETITGO or consider the uncomfortable possibility that it really is the case that everyone else is right and you're wrong. DonIago (talk) 17:12, 3 November 2022 (UTC)

Yeah, I guess so. It's just very difficult for me because I see the claim and concept immediately breakdown at the point where "inherently" a challenge has been altered to any version of: "inherently" a challenge apart from...; "inherently" a challenge except when...; or "inherently" a challenge not including...etc. It seems to me that if it is possible for the concept to have been altered to any of these versions, then it was never "inherently" a challenge in the first place. If you don't get what I'm saying, then I would like to tell you about this valuable "thing-a-ma-jig" I have. It will always be a valuable "thing-a-ma-jig", and I'd sure like to sell it to you. Now, I'm an honest person, so I do have to let you know there is a little caveat. Sometimes, at completely unpredictable moments it will be a valuable "thing-a-ma-jig" until suddenly it just won't be a valuable "thing-a-ma-jig" anymore, and the value of it will be completely worthless. You will be able to sell it to nobody, and will have to wait until it becomes a a valuable "thing-a-ma-jig" to sell it to somebody again. Wanna buy it? Huggums537 (talk) 22:21, 3 November 2022 (UTC)

Realistically, the only clarity we can give you is that it's safest to treat all removals of uncited material as real, valid challenges. WhatamIdoing (talk) 22:56, 3 November 2022 (UTC)

In addition, I think it's generally easy to tell. If someone who's never edited before comes along and blanks an entire article with no explanation, it is probably safe to assume that's vandalism. On the other hand, if an experienced editor removes some uncited content with the edit summary "Huh?", you should probably assume they are challenging it. Seraphimblade ^{Talk to me} 23:08, 3 November 2022 (UTC)

I don't think I've ever felt that I had significant trouble discerning a disruptive removal of unsourced material from a removal that was intended as a challenge, but if I did, then in a worst-case scenario I'd probably just start a Talk page thread to get other editors involved. If nobody proceeded to weigh in, then I'd likely either use my own best judgment (since someone else can always revert me) or forget about it and move on. Wikipedia isn't and hopefully never will be the most important thing I have going on in my life. DonIago (talk) 02:59, 4 November 2022 (UTC)

This brings us full circle to the problem I have been suggesting though doesn't it? I have been suggesting that this valid challenge rule does not exist, and is not supported by any policy, but only has some editors saying they think it is so, and using their own arbitrary judgement as to how it should be applied. And, so here we are with a few telling me "you should be able tell" and "we can't give you any clarity other than what we say" because there simply isn't any policy to point to with any exceptions, outlines, or guidance whatsoever. It appears to be made up of the thoughts of maybe some people who have been here a really long time who just thought that was the way it was for whatever reason... Huggums537 (talk) 07:43, 4 November 2022 (UTC)

Actually now that I think about it, I believe I agree with Doniago more than anybody. Wikipedia isn't and hopefully never will be the most important thing I have going on in my life It really is beyond time to pack up those toys and go home... Huggums537 (talk) 08:03, 4 November 2022 (UTC)

Before I pack up my toys and go, I'd like to make one last point because there is one other thing that brings us full circle here. I understand the idea you guys are trying to make is that we don't need any clarification about these exceptions in policy because they are very easy for any editor to use common sense and their own good judgement to identify, and supposedly anybody can discernably tell the difference between an ordinary content removal involving disruptive vandalism or routine maintenance like removing external links, and a removal that involves a verifiability challenge that requires a citation. However, this brings us all the way full circle back to the original post on this thread where the complaint was that there were editors who were actually apparently not able to tell the difference between a simple content removal, or a verifiability challenge, and were removing content based on one [verifiability/burden/content that must have inline citations] when it was actually the other [ordinary removal of problematic content]. So, it is clear that not all editors understand this difference or that it is really so easy to tell. Huggums537 (talk) 00:24, 5 November 2022 (UTC)

If we're going to treat a cn tag as a challenge, then we need to treat a cn tag that doesn't contain an explanatory edit summary as something the next editor along can remove for being unexplained. Valereee (talk) 11:50, 4 November 2022 (UTC)

No, we absolutely do not. A cn tag is self-explanatory: the person adding it thinks a citation is needed. You might disagree, but in that case it's on you to explain why one isn't needed. Nikkimaria (talk) 12:12, 4 November 2022 (UTC)

My concern is that (like me until this conversation) many people may be using the cn tag intending a simple 'hey, I notice this doesn't have an inline citation, does anyone know these sources well enough to know where to find one?' rather than intending to say, 'this doesn't seem right, cite it or remove it.' I won't be doing that any more (and am wishing there were an easy way for me to go back and remove them all), but we need to find some solution, as I can't be the only person who regularly did that pretty systematically for literally years. Valereee (talk) 12:23, 4 November 2022 (UTC)

Am I understanding correctly that you're saying you were applying cn tags not realizing that you were effectively challenging those statements and pushing them toward being subsequently removed? If so...I'm not really sure what to tell you. The documentation (and the name), clearly says "citation needed", not "citation would be helpful/nice", so to me that reads as a requirement that a citation be provided. That said, if I think a tagged and not-recently-added statement is likely verifiable/true, then regardless of how long ago it was tagged I'll likely leave it alone...but I can only speak for myself here. I suppose the documentation for the template could be updated to emphasize that yes, applying the tag is a challenge, but I don't know whether that would make any practical difference (though it also wouldn't IMO be a major change). DonIago (talk) 13:23, 4 November 2022 (UTC)

Yep. And, yes, I think we should make it clear.

Do most people really challenge stuff by dropping a cn? I challenge it by removing and opening a tp section. Challenging by dropping a cn...well, it could be decades before anyone bothers to respond. Valereee (talk) 19:57, 4 November 2022 (UTC)

I doubt I'd have any opposition to editing the documentation for the CN tag to clarify that, though I don't really feel that clarification should be needed. But I also don't see it causing any harm, so...

My general policy is that if unsourced info is newly-added and I feel it should be cited then I'll revert it and likely leave a notice at the editor's Talk page, on the assumption that the editor who added material is also best-positioned to source it. If it isn't newly-added, or for some reason I can't identify the editor who added it, then I'll drop a CN. That's to give editors time to provide a source...but depending on the nature of the material and how much I'm looking at the page, it may be months or years before I go ahead and remove it. There's WP:NODEADLINE. Typically it's someone else editing the article in a way that causes me to look it over again that will lead me to subsequently notice that there's now a CN tag that's been there for awhile. DonIago (talk) 20:39, 4 November 2022 (UTC)

I would have grave concerns about anyone removing a 'citation needed' tag on unsourced content in the body of an article without providing a citation, or a good argument that the content does not need a citation. - Donald Albury 13:34, 4 November 2022 (UTC)

And, what's to "explain"? It's pretty self-explanatory—a "citation needed" tag means it needs a citation (which it should have had to begin with anyway). Putting any more bureaucratic hoops to it than that is silly. Seraphimblade ^{Talk to me} 19:40, 4 November 2022 (UTC)

But, @Seraphimblade, it may very well have one. In the General References section. It may just need it inline.

By "explain" I mean an edit summary that says "Reads like OR" or whatever. Whatever makes you think the statement needs a source. Valereee (talk) 20:01, 4 November 2022 (UTC)

What makes me think the statement needs a source is that it is there. No more or less than that. So far as "general references", since we now have the "name" attribute for inline citations, there is absolutely no valid reason for anyone to ever be using "general references". If you want to use a reference more than once, set the "name" parameter and reuse it as many times as you like. Seraphimblade ^{Talk to me} 20:36, 4 November 2022 (UTC)

Well, we're all entitled to our personal opinions, and there are a (very) few editors who'd love to see a little blue clicky number at the end of every single sentence, but if you want to look at what the community overall endorses, then WP:When to cite is the requirement for FAs, which represent not just our best articles, but specifically our best practices for sourcing articles.

It doesn't endorse general references (though I believe that Wikipedia:WikiProject Mathematics does), but it does call out "Cited elsewhere in the article" as an indication that a given fact doesn't need to be cited again.

@Seraphimblade, looking through your contributions, it appears that you don't ever add any content or sources. Have I missed any examples of your additions, that would illustrate your ideas about how to improve articles? WhatamIdoing (talk) 21:23, 4 November 2022 (UTC)

WhatamIdoing, I must say, I'm impressed! You've managed to review tens of thousands of my contributions back as far as 2005 in that short a span of time? Would you care to share your methods with the rest of us? Seraphimblade ^{Talk to me} 22:28, 4 November 2022 (UTC)

Are you able to provide any counterexamples? WhatamIdoing (talk) 22:44, 6 November 2022 (UTC)

Sera, we're not talking about current additions, and I'm not talking about my own creations. Those are completely not what this is about; when I write, I cite. Always. (Literally I've occasionally inserted a cn while drafting when I know something is true but haven't found it in a source yet, just as a reminder to myself that I still need that source.) What we're talking about (from the beginning here, I'm the OP) is something that was created a decade or more ago with a list of references but no inline citations. Valereee (talk) 22:54, 4 November 2022 (UTC)

Valereee, do I understand it correctly that you want to discriminate three different situations:

The "cn" tag was added to some questionable statement that does need a strong support to stay in Wikipedia,
The "cn" tag was added to some statement that does not seem problematic, and can be possibly found in some of the sources listed in the "Literature" section, and the user who adds the tag does not question the statement, but just believes that addition of an online citation is desirable, but not necessary.
the "cn" tag was added to some pretty obvious statement that hardly needs any citation.

Would it be correct to say that it would be desirable to separate these three situations, and not to use the same tag? Paul Siebert (talk) 22:50, 5 November 2022 (UTC)

@Paul Siebert, yes, that's exactly what I'm getting at here.

The first situation, I'd support removal of the content even by the next editor along who thinks the information is at all dubious, maybe with an edit summary "rem dubious". If someone objects, the burden is on the objector to find the source.
The second situation is the one where I wouldn't like to see content removed systematically simply for being tagged. I'd support someone replacing the "cn" tag with some other useful tag, such as perhaps "unreferenced section" or "section needs more inline citations" with an edit summary of "more appropriate tag". If someone objects, discuss at talk. If the objector actually believes the content is dubious and needs to remain tagged as such, burden falls on the person who replaced the tag to find the source.
The third situation I'd support someone simply removing the tag with an edit summary of "bluesky" or something. If someone objects, discuss at talk. If it's really bluesky, others will agree.

Valereee (talk) 10:07, 6 November 2022 (UTC)

I would think the best response in all three scenarios would be: attempt to FIX THE PROBLEM, by adding in-line citations. That is, after all, the entire point of a cn tag. Blueboar (talk) 13:01, 6 November 2022 (UTC)

Sure. That's always going to be the best solution to any reasonably-placed tag. Valereee (talk) 14:15, 6 November 2022 (UTC)

Agreed. Where the difference in opinion seems to come in is when someone decides to "fix" the problem solely via deletion instead of adding in-line citations or examining the statement and determining it is not verifiable or otherwise should be removed. — Carter (Tcr25) (talk) 14:44, 6 November 2022 (UTC)

Here’s the thing… when someone deletes tagged material, they are not really trying to respond to the tag… they are taking things back a step further and trying to deal with the lack of citation in a different way.

We have three ways of dealing with uncited information… we can a) ignore it (ie do nothing) b) tag it or c) remove it. One person may choose b… but another person may come along later and choose c instead. Blueboar (talk) 15:26, 6 November 2022 (UTC)

Unless someone is engaging in a clear pattern of disruptively and systematically removing material, it seems to me that the best, and perhaps more importantly, most productive, option is still simply to find a source and restore the material with said source. Why not just assume good faith, figure the other editor did try to find a source and just couldn't find one for some reason, and find a source yourself? Why does it matter who is providing the source? How is this not preferable to starting a discussion that will likely turn confrontational, waste more time and energy, and fail to yield a more productive outcome? I mean, this whole conversation, to me, seems to be an example of this very problem, because I have my doubts that anything productive is going to come from it at this point (i.e. no consensus to change anything), so what's the goal here? DonIago (talk) 15:35, 6 November 2022 (UTC)

Well, for me the goal is to understand what our policy actually is and make sure others also understand it. I'm not trying to change policy. I'm trying to make sure everyone is clear on what the current policy is. Valereee (talk) 15:39, 6 November 2022 (UTC)

Consider the following statement:

"Paris is a capital of France. Its population is 2,165,423 residents. It is assumed that the first mention of Paris was made by Caesar in his Commentarvi de bello gallico: according to one Medieval translator, Caesar's "Lutetia" refers to a settlement where modern Paris is situated."

Now imagine I put "cn" tags, because I believe citations are missing, and now the text is:

"Paris is a capital of France^{[citation needed]}. Its population is 2,165,423 residents^{[citation needed]}. It is assumed that the first mention of Paris was made by Caesar in his Commentarvi de bello gallico: according to one Medieval translator, Caesar's "Lutetia" refers to a settlement where modern Paris is situated^{[citation needed]}."

What should we do with these tags? Obviously, the first tag is redundant, and anyone can remove it with an edit summary like "remove a redundant tag, the fact that Paris is a capital of France is too obvious to be supported by a citation".

The second tag has a different meaning: by placing this tag, I hardly question the statement: the number looks pretty reasonable, I just want to know what concrete source says that, but I have no time to dig into sources, although I don't think that would be hard.

The third tag is more problematic: it may require some considerable amount of time and efforts to identify the source that says so, and if it was not easy for me, it would be equally difficult for other editors/readers. Therefore, it would be correct to consider the third sentence unverifiable until the source has been provided, so if a "cn" tag is not replaced with a citation in a reasonable time, the sentence should be removed.

I think the core issue here is the statement "Readers must be able to check that any of the information within Wikipedia articles is not just made up." This "able" is too vague and too ambiguous. Imagine that the third sentence from my example above is supported by only one source, the book that was published 100 years ago, and the only survived copy can be found in Vatican library. Formally, a reader is able to find it and verify the sentence, but technically it is impossible: that source is something like a Russel's teapot. Therefore, all non-obvious statements that lack a citation must be considered non-verifiable: whereas, formally, it is possible to find some source that supports them, technically, it is not easier than to find Russel's teapot. Therefore, it would be more correct to rephrase our policy to something like:

"Readers must be able to easily check that any of the information within Wikipedia articles is not just made up, i.e.that all material in Wikipedia is attributable to reliable, published sources. To make it possible, quotations and any material challenged or likely to be challenged must be supported by inline citations.

Paul Siebert (talk) 16:52, 6 November 2022 (UTC)

If we add the word easily, someone will declare that it's not easy to verify offline sources, paywalled sources, non-English sources, and technical sources. WhatamIdoing (talk) 22:47, 6 November 2022 (UTC)

(od)The policy is quite simple:

Adding Material: An editor may freely add material - with or without an in-line citation, as they deem appropriate. However, there is a huge caveat to this… IF they don’t include an in-line citation, another editor can come along later and either tag OR remove the material (a “challenge”).
Challenging: When an editor comes across uncited material, they may deal with it in one of four ways: 1) they can ignore it and do nothing. 2) they can add a citation themselves 3) they can tag it in the hope that some future editor will add a citation. 4) they can remove the material. Which of these is “best” depends on specific context, and can only be determined by the challenging editor. It is a “judgement call” and not something that can be defined by policy. Different editors will make different judgements. Disagreements should be discussed on the talk page.
Responding to challenges: when material is challenged, those who want to retain the material (or remove the tag) are required to provide an in-line citation to do so.
A further note: sometimes there are additional issues with the material that are not resolved by providing an in-line citation. Those need to be discussed on the article talk page. However, providing an in-line citation is always a good “first step” towards resolving any issue. Blueboar (talk) 16:43, 6 November 2022 (UTC)

The policy is simple, but the use of the cn tag may not be clearly-enough communicated. For my own future editing I've got what I need. I'd still like to find a way to better communicate how to use a cn tag. Valereee (talk) 16:47, 6 November 2022 (UTC)

This long and winding thread is really the first time I've heard that there was any significant confusion about the usage of the cn tag, as I indicated in an earlier comment. That said, it seems to me that concerns specifically about the tag would better be addressed at the Talk page for said tag. DonIago (talk) 16:53, 6 November 2022 (UTC)

A cn tag means what it says: someone thinks that a citation is needed. What’s confusing about that? Blueboar (talk) 16:57, 6 November 2022 (UTC)

Some people believe that when someone thinks that a citation is needed, that does not necessarily mean that verifiability of the text labelled with "cn" is challenged. Thus, in the original Valereee's example, multiple "cn" tags were added to the article that contained a list of cited literature, that that article adequately reflected what those source said. Valereee argued that "cn" tags were added not because the text was challenged, but because there was a need in further technical improvement. Paul Siebert (talk) 17:48, 6 November 2022 (UTC)

And apparently someone got tired of waiting for a “technical improvement” that never came… and decided that removal was a better option. Tagging and removing are two different ways to deal with the same problem (ie, uncited - or unclearly cited - content). That’s not confusion over the use of the tag, that is a disagreement over which method (tagging vs removal) best resolves the fact that the material did not have in-line citations. Blueboar (talk) 18:24, 6 November 2022 (UTC)

Yes, that is a my standard approach: when I see some unsourced statement that is obviously odd, I remove it. When I see some plausible, although a controversial statement, and I cannot verify it by myself, I place the "cn" tag. If the tag has been ignored for a relatively long period of time, I either may remove the text by myself or do not object to its removal by others.

In general, when someone puts the "cn" tag, that means that user has a feeling that the text needs verification, and that verification is not easy (otherwise that user would have done that by themselves). Therefore, by default, the "cn" tag normally means that verifiability of the text has been challenged.

The problem is, however, that Wikipedia has many non-controversial articles of reasonably good quality that contains little or no inline citations and have only a literature list or cite no literature at all. Taking into account that Wikipedia has no deadline, what should we do with those articles if some users put "cn" after each sentence, and noone bothers to replace it with citations in a reasonable time?--Paul Siebert (talk) 20:50, 6 November 2022 (UTC)

I think you summed it up very well. Here is what I think is missing. If you see an uncontroversial statement that's been there for a long time on an article that lacks an inline citation, you should try to find a citation by checking the other references in the article, before you tag it with a cn tag. Going through and tagging statements which are uncontroversial and unlikely to be challenged on a verifiability basis but simply do not have citations, isn't really improving the encyclopedia, if you haven't done a good-faith BEFORE effort to cite those statements. Provided that the statement is otherwise thought to be uncontroversial and verifiable. There is no policy that states that every single statement in every article must have an inline reference, simply that we may challenge and remove those that do not, but it is implied that you should engage in good faith and collaborate on the process. Andre🚐 20:58, 6 November 2022 (UTC)

If the problem is just a lack of specificity in citations, but the verifiability of the content overall is not in doubt, we already have {{inline citations}} to call attention to that. (As well as WP:SOFIXIT, if someone actually has reviewed those "general" references and found that they really do support the article content. In that case, just add the citations, since you already reviewed those references and know which one supports what.) But, as with "BEFORE" in general, it is backwards. BEFORE you start an article, have the sources in hand that will demonstrate the subject's notability, and cite those sources. BEFORE adding material to an article, have the reference in hand that verifies what you want to add, and cite that reference. That is what "BEFORE" should actually be. The burden should be on the person who wants to add it, not the person who challenges it because someone failed to do that. Even if someone sources it after the fact, that's at best a substandard citation—we don't know if that's actually where the author got their information, we at best know it's where they might have. The author should be specifying where they actually did get it (and if the "source" is "I remember this as being so" or "I heard this someplace or another", refraining from adding it at all.) Seraphimblade ^{Talk to me} 22:13, 6 November 2022 (UTC)

Yes, but some articles are grandfathered and have existed for years and are in various states of being fixed. Andre🚐 22:31, 6 November 2022 (UTC)

Your take on BEFORE is very much today's standard, but the original question centered on older content added when an acceptable level of BEFORE was add it to the general references. The issue as I see it is there are some editors who are trying to enforce a "every single statement have an inline reference" standard and are using deletion as a tool to do that (either as a spur to get someone to add a reference or to simply reduce an article to material with an inline reference) without assessing if the material is uncontroversial or verifiable. — Carter (Tcr25) (talk) 22:36, 6 November 2022 (UTC)

Is it stated anywhere that old articles are “grandfathered” from changes to policy that occurred after they were written? I do understand giving editors some grace time to conform articles new standards when policy changes, but surely there comes a point when we can “you have had more than enough time to deal with this”. Blueboar (talk) 23:54, 6 November 2022 (UTC)

It's not that things are "grandfathered in;" it's a matter of how best to deal with the changed expectations. The point that some of the commenters here and in related discussions seem to be making is that people should have included an inline source when the content was originally added; my point is that this is a changed expectation, so we can't blame the original editor for failing to provide an inline citation. Once that's set aside, the question becomes how best to fix the issue.

In the actual articles I've been involved with, when an editor who holds the position "when I remove unsourced material, of course it's probably unverifiable—why else wouldn't it be sourced" and they are removing material based solely on that lack of a proximate footnote and they are under no obligation to even do a simple Google search, it has left text that is disjointed and worse than what was there before. It could be these cases are outliers, but my sense is that is not the case. One might argue that it's ultimately a good thing because the text was eventually fixed and sources added, but 1) what about the articles that don't have someone watching them and 2) the ends don't justify the means.

Again, this isn't about a new addition or BLP violation or something that is clearly bunk; it's about information that seems reasonable and verifiable even if it lacks a proximate footnote (or, as in the case of the situation that sparked this conversation, a statement that's been flagged with a {{CN}} tag for a while). What I think is a problem is that people are using BURDEN to cut without examining what's being cut. (And before I'm accused of making that statement in bad faith, it is based on comments such as the first one quoted above and this one: I cannot be expected to do this [look for sources] will [sic] all the uncited info found in all articles.) It's not a matter of "you've had more than enough time to deal with this," but a broader question of how do we bring these articles up to today's standards given WP:WIP, WP:NODEADLINE, and that most of these articles are not up for GA or FA review? From what I've seen, we can't delete our way out of this, so if people see policy as encouraging deletion without investigation, then it's worth asking the question if something is out of balance. — Carter (Tcr25) (talk) 02:49, 7 November 2022 (UTC)

I suppose this is where we disagree… I think we can “delete our way out of this” (although “delete” is the wrong word in this context). Sometimes, the best way to fix an article that does not meet our current citation expectations and standards is to use an axe… and to bluntly remove material. The community is thus encouraged to rebuild the article (by returning the information and adding in-line citations… thus ending up with an article that meets our modern expectations and standards). It’s all part of the normal editing process, just in slow motion.Blueboar (talk) 13:53, 7 November 2022 (UTC)

A drained lock chamber

@Blueboar, I want to agree with you, but I'm finding it a little difficult.

This image is an example of something that was removed, along with dozens of sentences and paragraphs, for being uncited. It was removed by one of the editors with a reputation for demanding citations, and if you don't add the citations, he blanks most of the article, claiming BURDEN as evidence that it's his right.

Looking at this, and thinking about our actual-in-practice standards for images and captions, do you think he carefully examined what he was cutting? Or do you think that removing this might have been a bit ham-fisted and unjustified? WhatamIdoing (talk) 23:25, 8 November 2022 (UTC)

I would certainly agree that it was ham fisted. But not necessarily that it was unjustified (I would need to see the state of the article prior to the removal). You are correct that we have a different standards for images as opposed to text. Blueboar (talk) 13:33, 9 November 2022 (UTC)

It sounds to me as though you feel this editor was being disruptive. Have you approached this matter from that perspective, and if so, what was the outcome? That one or even "many" editors might follow policies while still behaving disruptively doesn't necessarily indicate an issue with the policies themselves, just the editors' interpretations thereof. As an example, things tend not to go well for editors who claim that they were not violating the edit-warring policies because they did not explicitly violate WP:3RR.

I'm not commenting as to whether or not I feel this editing was disruptive or whether or not our verifiability policies could be better, merely pointing out that potentially disruptive conduct that nevertheless falls within the policies doesn't inherently mean the policies need to be changed. I'm admittedly having some difficulty believing that this example is indicative of a widespread, multi-editor issue. DonIago (talk) 04:12, 9 November 2022 (UTC)

DonIago raised an important point… it is possible to be “in the right” policy-wise, and yet still be disruptive in how you follow the policy. But… that is a behavioral problem with a specific editor, not a flaw with the policy. Blueboar (talk) 13:33, 9 November 2022 (UTC)

It is a behavioral problem with a specific editor, but in the case of the en masse removals at Radio, for example, when the behavior was taken to ANI, the bulk of the discussion either skipped over the behavior because it was allowed under BURDEN or outright approved of it based on BURDEN. If the policy is being used to excuse (if not actually approve of) problematic behavior, then it's fair to ask whether or not there is a flaw in the policy. — Carter (Tcr25) (talk) 14:34, 9 November 2022 (UTC)

It is also necessary to consider, at that point, whether you are in fact correct in calling it "problematic" to begin with. If the community said it was not, that may just be a case where your view on it didn't carry the day, and we've all been there. Seraphimblade ^{Talk to me} 21:31, 9 November 2022 (UTC)

I'd be curious to read the ANI discussion, but ANI has also been known to draw conclusions that I disagreed with on occasion. DonIago (talk) 21:46, 9 November 2022 (UTC)

I Took a look at the ANI discussion about Radio (you can find it in the archives from early September). I note that some of the people involve in that discussion are also involved in this one… and opinions haven’t changed much: we are split, with some supporting the removal, and others feeling that it was disruptive.

What the ANI discussion does not note (but I will) is what happened back at the Radio article itself… after some initial revert warring, (while everyone was off at the drama board arguing about “the rules”)… a very dedicated editor finally took the time to insert the requested in-line citations.

In other words, after years of editors not paying attention to uncited material and cn tags… the problem was finally addressed and fixed. Why? Because someone decided that they were willing to be the asshole… and insist that proper in-line citations be provided. Someone decided that since tagging wasn’t working, it was time to change tactics, up the anti, and actually remove content. And look what the end result was… we have a much better article. So … to me, that removal was a success! Blueboar (talk) 22:54, 9 November 2022 (UTC)

That's not how wikipedia is supposed to work. WP:POINT. Wikipedia is about collegiality and ease of use and friendliness, that's core policy and high guideline. Yes, an improvement was wrested from the grasp of the radio article, but also, that doesn't improve the encyclopedia and the project if we strongarm people into finding references by cutting a wide swath through material that isn't prima facie unverifiable without making an attempt to verify it. Andre🚐 22:58, 9 November 2022 (UTC)

um… no. Wikipedia is about presenting information that is well researched, verifiable by citation to reliable sources, presented with a neutral point of view, and without original research. Collegiality and friendliness do (usually) help in achieving those goals, but they are not the goals themselves. And if (occasionally) being unfriendly will improve an article, I have NO problem with unfriendliness. Blueboar (talk) 23:51, 9 November 2022 (UTC)

WP:CIVILITY is required. Andre🚐 00:21, 10 November 2022 (UTC)

Sure, but in examining that situation, I did not see anything that violated WP:CIVILIY… certainly removing uncited information is not a breach of civility. Indeed, I would say that hauling someone in front of ANI for doing so is rather uncivil. Surely the most CIVIL thing to do is to avoid the drama, and quietly (if grudgingly) return the removed information with citations. Blueboar (talk) 00:45, 10 November 2022 (UTC)

+1 Donald Albury 01:47, 10 November 2022 (UTC)

I do think that if someone asks politely, the important thing is to grant reasonable requests and not to insist that the material must be removed like it's some kind of absolute right. AGF and the benefit of the doubt within reason. It's not exactly incivil not to depending on the tone. We expect a good bedside manner and explanation of decisions that take into account the consensus and the will of the community. Andre🚐 02:14, 10 November 2022 (UTC)

+1 Carter (Tcr25) (talk) 03:37, 10 November 2022 (UTC)

I'd also note that in the radio article, checking it found at least one actual error, specifically that "omnidirectional" antennas receive signal from all directions, while omitting that this is only true in one plane. So, some people probably thought that it was "obvious" that "omnidirectional" means "all directions", but, well—it doesn't really. And that's the exact reason for insisting that people write from actual research in sources, and not from personal knowledge or "blue sky" stuff. It is actually an AGF to use a "fact" tag at all; there is nothing wrong with just removing unreferenced material flat out. Using a "fact" tag is the extension of AGF to give some time to fix it, but if after some period of time that hasn't happened, removal is the next step, and then it doesn't come back without a reference, period. We don't need to set some particular waiting period, since that waiting period can be zero to begin with. Seraphimblade ^{Talk to me} 07:27, 10 November 2022 (UTC)

I haven't read through this ridiculous long thread, I'm not going to. Thankfully Seraphimblade states everything I would have said. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 10:43, 10 November 2022 (UTC)

I wouldn't classify that as an "error", I'd say it would be technical jargon that doesn't really need to be in the article for getting the overall idea across to general readers the main difference between directional and omnidirectional. It is fairly obvious almost nobody will be using either type of antenna while it is laying on its side, or at least it should be assumed people are using them in an ordinary manner, and I don't think it is the job of the article to say if omnidirectional types are "best" for space and aviation communications or not. That isn't really a case of something needing to be "fixed" so just removing "flat out" doesn't make a good faith assumption either, but really denies another good faith assumption, while leaving the material in does not in any way deny the good faith assumption that the article could be improved with a better explanation. That is a big difference in faith assumptions. Huggums537 (talk) 11:28, 10 November 2022 (UTC)

It's not an error, it's a lack of precision. An omnidirectional antenna only works on one plane. Well, yeah, and we only are human beings in this universe. Sure, it's an improvement to explain a bit more how an omnidirectional antenna works. But how did that end up in the article? Someone wrote that in thinking it was true but didn't explain it in detail. And it is true, but a qualification or a caveat will make it truer. Andre🚐 23:28, 11 November 2022 (UTC)

Blueboar, that removal was a success from a content perspective (many others have not had the same result), but it's not a success from a collegial or community-building perspective. In fact, it feels like an abusive or bullying interaction – like an angry father declaring that if you haven't cleaned up your room by the time he gets home from work, he'll shovel everything into the garbage. Sure, he let you pick some things back out of the garbage, but it shouldn't have reached this point in the first place. This kind of power play might result in getting the room cleaned (or at least emptied), but it does not result in editors feeling like we're all working together towards a positive goal. We need editors, and articles need sources, but I don't think we should upset editors long-term (and I do mean very long term: This kind of thing is remembered for many years. I'm sure I'm not the only editor who remembers that the blanker in question was desysopped years ago for blocking someone who re-added material that he'd just blanked as being unsourced) in an effort to manipulate people into providing sources today. WhatamIdoing (talk) 20:33, 10 November 2022 (UTC)

+1, Wikipedia has goals of accuracy, community-building, and being a place where abusiveness isn't tolerated. Mistakes are ~~punished~~addressed but there is also forgiveness. Reasonable requests should reasonably be granted. "It's time to remove this" says who? WP:NODEADLINE, eventualism holds that eventally all the things will be fixed. If you're impatient, WP:SOFIXIT. Fixing it would mean actually making some attempt to source the material, not to do it en masse if people are objecting and wanting to go provide some sources. The Wiki way to do it is to put some tags, and start a discussion, and engage constructively and productively when someone objects to it. It's not improving the project to trim out things that haven't been sourced yet, but probably could be and will be easily, and people who can tell the difference should be the ones removing that material. Andre🚐 01:30, 11 November 2022 (UTC)

Andrevan. No. Wikipedia's goals are accuracy, fact checking, neutrality and avoidance of original research. That is our core content policy. Contrary to your belief, Wikipedia's goal is NOT "community-building, and being a place where abusiveness isn't tolerated". Yes, we are trying to create conditions where editors can work collaboratively and without abuse, but that is not our primary goal: our primary goal is creation of content that complies with NOR, V, and NPOV, and everything else is needed to create comfortable conditions for the editors, provided, but only provided that they are working in accordance with WP:V, WP:NOR, and WP:NPOV.

If some user do not understand what WP:V means and repeatedly adds non-verifiable information, it would be quite correct to let them know that this user should not edit Wikipedia. It is quite possible that that user will see that as abuse, but that is our policy. And so on, and so forth.

WhatamIdoing Yes, shoveling everything into the garbage may look too radical, but everything depends on the ratio of valuable things and real garbage in what is being shoveled.

In connection to that, let me remind you that high quality content is created mostly by removal of everything that does not meet quality standards, and no by non-critical adding new texts. Removal of the content that does not meet our standards may be more important than adding new content, especially in the encyclopedia that is being written by amateurs.

In other words, out policy should facilitate removal of questionable content, so addition of new content should be more difficult than removal of poor quality content. That is the only way to improve quality of Wikipedia. Yes, that may lead to disappearance of some relatively good texts, but, as you know, almost nothing disappears from Wikipedia's history: everything can be restored if someone finds needed sources. Paul Siebert (talk) 04:04, 11 November 2022 (UTC)

I was referring to a situation where content is likely to be verifiable, but hasn't yet been verified. Most likely, older content that wasn't given an inline citation. So I disagree. Wikipedia absolutely has a goal of "community-building, and being a place where abusiveness isn't tolerated." Your straw man about removing unverifiable content is entirely irrelevant and an uncharitable interpretation of what I said. I am referring to content that is verifiable but hasn't been attempted to be verified being removed and then users not wanting to restore it on good faith that a citation will be provided. Andre🚐 04:20, 11 November 2022 (UTC)

Again, we are not Facebook or any other social network. Community-building is not our goal, it is just one condition that makes creation of good content more efficient.

With regard to verifiability, what exactly does it mean? If I wrote some content keeping some source in mind, but I have not bothered to provide a reference to it, is this content verifiable? If everybody can google my source, then it would be correct to consider it verifiable. However, if my source is some unique book that currently exists in a couple of copies, and it is virtually impossible to figure out that I used that book for creation of my content, is the text added by me really verifiable? I would say no.

We have a policy, and it says that burden of proof is on those who adds/restores new material.

With regard to your "that is verifiable but hasn't been attempted to be verified", how exactly do you see that? If I remove some text that stayed with the "cn" tag for one year, should I provide any proof that I tried to verify it? How long should I be googling to prove that I took sufficient amount of efforts to fine the reference? Do you realize that by writing that you shift WP:BURDEN from those who adds texts on those who remove them?

I already provided a reference to Russel's teapot, and I reiterate: if I want my contribution to stay in Wikipedia, I do my best to make its verification as easy as possible (and the only way to do that is to provide a citation). If someone hasn't bothered to provide a citation, then their contributions hardly deserve to be kept. Paul Siebert (talk) 04:53, 11 November 2022 (UTC)

And, as soon as you want to speak in terms of abusiveness and community building, adding some information that is hard to verify without a citation is a form of disrespect: by doing that, the user literally says: "I do not care how much efforts it may take for you to verify my information, I do not value your time and do not respect your efforts." In contrast, the user who provides a correct citation with a page number makes it easy for other users to check that information. This user shoes respect to others. The first type behaviour is abusive, the second one is respectful and collaborative. The first one should be discouraged, the second one should be encouraged. The easiest way to do that in Wikipedia is to remove contributions by the first type users and to keep contributions by the second type users. And that is the best way to create collaborative community devoid of abusiveness. Paul Siebert (talk) 05:04, 11 November 2022 (UTC)

I'm guessing that English isn't your native tongue, so forgive me if I'm misunderstanding, but once again your mischaracterization of my argument is specious, I never said Wikipedia was a social network like Facebook, I said Wikipedia has a goal of community-building. There absolutely is a goal of having a community of editors and a project of civil, collegial individuals who treat each other respectfully whilst engaging in a project of encyclopedia construction. The meta aspects of the Wikipedia community have long rested on explicit policies of civility and good faith. These are foundational. WP:5P. Respect. I'm specifically talking about, and have been talking about, long-standing content that is non-controversial, which has stood in an article, that is being removed as though some cosmic timer is up. Your account was created in 2008 but back in 2004 when I became active, we didn't have a content guideline that every piece of content had to have an inline citation in that sentence. So your idea that this is disrespectful to WP:PRESERVE and defend featured articles having massive deletions against consensus... well... it's quite beyond the pale in my view. You say that the best way to create community is to treat users as abusive if they had added content that helped build that encyclopedia? You clearly have a very definition of abuse and respect. And as far as WP:BURDEN I think you're demonstrating the argument why a change or clarification to WP:BURDEN might be worth while: because it should not be used as justification for such logic as this. Andre🚐 05:26, 11 November 2022 (UTC) P.S. WP:ETTIQUETTE

WRT yours "So your idea that this is disrespectful to WP:PRESERVE and defend featured articles having massive deletions against consensus...", in which universe can an article devoid of online citation be considered WP:FA?! Paul Siebert (talk) 19:09, 11 November 2022 (UTC)

Well the example was radio, it's not an FA. But there absolutely are older FAs that don't have inline cites for every sentence. Andre🚐 19:36, 11 November 2022 (UTC)

If some paragraph has just one inline citation, we assume that all what it says was taken from that source. Your reductio ad absurdum attempt is absurd. Paul Siebert (talk) 19:47, 11 November 2022 (UTC)

Maybe that's what you do. But that is kind of the point: you should not indiscriminately tag things cn or remove content that is tagged without checking the other nearby sources or perhaps the ones in the notes/references/literature section. Andre🚐 21:04, 11 November 2022 (UTC)

Please see my 23:17, 11 November 2022 (UTC) post where I discuss this issue. Paul Siebert (talk) 23:18, 11 November 2022 (UTC)

"We" might assume this, but the history of Radio suggests that at least one editor does not agree with you (⌘F on "fairly limited" to see one example in that diff). WhatamIdoing (talk) 21:20, 11 November 2022 (UTC)

Yes. I have long felt that adding content to Wikipedia without citing reliable sources is unfair to readers and other editors. We are here to build an encyclopedia. Maintaining a collegial community is important because it supports building the encyclopedia. The community is not the goal, the encyclopedia is. - Donald Albury 15:12, 11 November 2022 (UTC)

But I was talking about content that was added years ago before the policy requiring such inline citations. Andre🚐 19:36, 11 November 2022 (UTC)

I think that ref tags were introduced in 2005. We certainly have content that predates the existence of ref tags. WhatamIdoing (talk) 21:23, 11 November 2022 (UTC)

(BURDEN) Time to wrap this up?

It is becoming obvious that we have some fundamental disagreements about the removal of content (even beyond long tagged content). I doubt we will reach a consensus. Indeed, it seems that the more we talk, the further from consensus we get. So… perhaps it is time to accept that further discussion isn’t going to change anyone’s mind (and so is pointless), agree to disagree, and end this?

In any case, my opinion has been stated and isn’t going to change… so I’m done. “Have fun storming the castle”. Blueboar (talk) 16:40, 11 November 2022 (UTC)

No objection here. While I've raised relevant points where I thought I could, I feel the discussion's generally been going in circles for some time now. DonIago (talk) 16:54, 11 November 2022 (UTC)

Per Aumann's agreement theorem, when honest rational thinkers have a dispute about some subject that belongs to their common knowledge domain, they cannot agree to disagree. Therefore, if you state that we are moving further from consensus, you thereby imply that a significant number of participants are either incapable of thinking logically or they have some information in mind that they do not want to share. My impression is that that is not the case, so we are doomed to achieve consensus.

I propose to take a look at the problem from a little bit different perspective. If we agree that removal of information labelled with "cn" is always allowed, that may have some negative consequences, concretely, some well written and non-controversial texts may disappear from Wikipedia (although they remain in history, so they are not completely lost). However, that also has some positive consequences: the fraction of well sourced texts in Wikipedia will increase, which may lead to improvement of the overall quality of Wikipedia, which may positively affect its reputation (which is currently not too high).

If we agree that "cn" tag does not necessarily a reason for deletion, and the user who deletes such a text should demonstrate that they did their best to fix the problem first by attempting to find a source, and that "cn" tag can be easily removed after some time, that may preserve some good texts in Wikipedia, but the price that we pay for that is a further decrease of the quality of Wikipedia in general.

Which option does our community prefer? Maybe, it is a time to start an RfC? Paul Siebert (talk) 17:58, 11 November 2022 (UTC)

I think you're presenting a false dichotomy here. A CN tag may sit on good information or bad information. Similarly a reasonable-looking reference may sit on good or bad information. In some cases, a source can sit on a paragraph verifying part of the text, but not all of it. An edit may move the proper source away from a statement making it appear that a CN tag is needed. In either case the quality of the information hasn't been assessed; it's simply been approved or questioned due to the presence or lack of a ^[#]. The negative consequences of unassessed removal of information isn't just that some well written and non-controversial texts may disappear from Wikipedia, it's that what remains may be disjointed, poorly written, or otherwise problematic, decreasing the overall quality of Wikipedia and harming its reputation.

You've made the argument that Removal of the text is another way to draw attention of other users, which is fair, but vandalism, edit warring, and other behaviors generally accepted as non-constructive to the common goal of building a better encyclopedia can also draw the attention of other people. I know we all have to edit in our own way to our own standards and can't require others to do the work for us, but let's agree that an editor should do some degree of examination of the material and then check what's left behind if it is removed to make sure the resulting edit is an improvement. — Carter (Tcr25) (talk) 19:04, 11 November 2022 (UTC)

I think you are seeing a false dichotomy where it is not present.

Not only I agree that a "cn" tag may sit on good information or bad information, I myself say that. However, the question is what should we do with that: should we remove the information labelled by the "cn" tag if no source has been provided during a reasonably long period of time (which increases a probability of disappearance of both good and bad information from Wikipedia), or we are allowed to ignore the "cn" tag and even remove it (and, as a result, challenged texts, both good and bad, are preserved)?

The answer to this question may have a profound effect on quality of Wikipedia. That is up to the community to decide, but we need some definite answer. Paul Siebert (talk) 19:43, 11 November 2022 (UTC)

It's not just should we remove information or let it stand; it's also what should guide the decision to remove information. I think we all would agree the best possible action one can take is to improve the content be that by adding a source, reworking the material, or removing it if necessary. However, where the difference in opinions comes in seems to be whether or not removal is, on the whole, an improvement that should be taken if one isn't willing or able to take the other approach. BURDEN as written advises caution in making removals, but does give permission to remove something that appears uncited with no obligation to even do a simple Google search. That's the bit missing in your question and where I think the bulk of the problem is. It's not just should we remove material flagged with a CN tag (presumably with some consensus-based considerations around how long is long enough), but if it's removed what are the obligations on the remover to make sure they aren't leaving behind an article that is in worse shape. Right now, the easy thing to do is to delete even up to the point of WP:TNT without consideration of what one leaves behind, but I don't see how that builds a better encyclopedia. — Carter (Tcr25) (talk) 20:30, 11 November 2022 (UTC)

As soon as you started to speak about obligations of those who removes tagged texts, may be we should remember that the person who placed the tag also has some obligations? When I see the "cn" tag, I assume that the person who placed it

(i) had read the text and found it non-obvious;
(ii) had tried to find a source, and realised that simple googling gives no results;

Therefore, when I see the "cn" tag that stays in Wikipedia for several months or years I assume that other people either failed to find any good source or this information is not important, so nobody cares about it. In both cases, removal of the tagged text is quite justified.

Maybe, instead of discussing what should we do with tags, let's talk about correct usage of this tag? Paul Siebert (talk) 23:17, 11 November 2022 (UTC)

I agree, that you should not place a citation needed tag if the text is obvious, but that is not what many editors do. They will just say it "needs citations" without having done i, and ii above. If we had that as a recommendation or a guideline, it would resolve my objection to WP:BURDEN. Andre🚐 23:35, 11 November 2022 (UTC)

We are assuming good faith, and therefore we assume that if some user puts the "cn" tag, they sees the text potentially non-verifiable, and they attempted to find the source by themselves, and these attempts were unsuccessful. That is what we assume. However, it is always possible that some user misuses the tag. If the tag was used incorrectly, it can be removed, provided, but only provided that other users achieved consensus about that.

Another case when the tag is being used is as follows (I myself do that sometimes): it is a question addressed to other editors: "Guys, this information looks odd, where did you take it? I would like to see your source, because my sources do not confirm it." If there is no responce in a reasonable time, it would be quite correct to remove it.

One way or the another, if the original reason for that discussion was the fate of old texts, maybe, we can make a special reservation about that and specify that old article are, to some extent, exempt? Paul Siebert (talk) 00:45, 12 November 2022 (UTC)

I suppose that would work as well. BURDEN as written also doesn't, as far as I can tell, actually advise that users attempt to find sources before places tags or removing material. Andre🚐 01:10, 12 November 2022 (UTC)

Paul, I can see how you came to this view, and I think it's quite logical, but it doesn't align with the facts. You assume that "the person who placed it...had tried to find a source". I am telling you that your assumption is wrong because I have personally placed such tags without attempting to find any source whatsoever.

Sometimes I do this because I'm in a hurry. Sometimes I do this because I don't like the text for some other reason. Sometimes I do this because I see some content that we always want citations for (e.g., statistics), but don't have any concerns about.

So given this fact, how does that affect your confidence in your assumptions? WhatamIdoing (talk) 18:55, 12 November 2022 (UTC)

I also agree there's a false dichotomy here. In response, I wanted to raise what is not the main issue here, but something to keep in mind nevertheless, namely the bad actors. There are those, who if they can game the system, they will (i.e., vandals and trolls). Whatever decision is reached, some will try to take advantage of whatever the (new) wording is to justify their bad acts. If a change is forthcoming, please take care not to give them a bigger foothold than they now have. Mathglot (talk) 00:17, 13 November 2022 (UTC)

I fear any change could also have the opposite issue. IP and new editors can already find editing troublesome. Many of my edits when I was an IP were reverted by editors patrolling for vandaldism. Generally this was becaise I was fixing technical issues and they didn't understand them, and were easily sorted by discussion. An IP editor removing text from an article with no edit summary could be an expert in the field. The last thing Wikipedia needs is such an edit being reverted with a link to some new piece of word salad. Note I don't mean to criticise any particular side in this argument when talking of word salad, I'm generally of the opinion that we need less automated templating and ingroup language. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 00:55, 13 November 2022 (UTC)

It's nobody's fault. An effort to prescribe categorically what should happen based on only 1 or 2 criteria (which is most of what this discussions is about) is doomed to fail because in Wikipedia, such decisions are made by considering many more criteria. Wikipedia:How editing decisions are made Sincerely, North8000 (talk) 21:08, 11 November 2022 (UTC)

I don't think so. In this case, we have a few editors "successfully" editing based on a single criteria. The criteria is: if the paragraph doesn't end with a little blue clicky number, it's fair game for deletion. Points like whether I can be expected to know that water is wet at typical room temperatures are unimportant, and I'm definitely not required to spend 60 seconds with my favorite search engine. These editors cause a lot of drama, but, hey, not everyone thinks that annoying other editors is a barrier to success. That's because some editors believe that "success" looks like having a little blue clicky number at the end of every sentence, rather than having relevant, verifiABLE, if sometimes unCITED, information in the article. IMO this is similar to saying we should block everyone, because that will stop vandalism, but hey – different values, different results.

That, by the way, is where Paul's invocation of Aumann goes astray. The dispute doesn't center on common knowledge. It doesn't even center on whether you're allowed to remove one uncited fact (or a small number of them) from an article. The question is whether it's better to mass-blank an article (e.g., Radio was cut from 11,750 words to 2,000 words) without trying to find sources, have any true fact-based doubts about the contents, or mentioning your dissatisfaction on the talk page. Some editors think that in cases like this, it is better for Radio to mention what "AM" and "FM" stand for, with links to the relevant articles, than to have an article with a high percentage of little blue clicky numbers.

I suspect that many editors would say that if you want to remove one or two uncited facts from an article, that's fine. Readers will lose some good content, but they'll also be spared some bad content. We don't even mind if editors address promotionalism and other unwanted content by WP:Stubbing it down to a couple of bland introductory sentences.

But, as a behavioral matter, if you want to blank nearly 10,000 words from a long-established article solely because there are no little blue clicky numbers (and you think it's unreasonable for anyone to expect you to look for sources yourself, or to recognize that FM radio was not just something made up one day by newbie editors), then we are not impressed. We tend to think that mass-blanking of uncited content that many other editors appear to accept is a violation of WP:PRESERVE. We tend to think that mass-blanking content because it wasn't cited on your unspoken schedule is WP:POINTY. We probably think that you shouldn't do this, and we probably think that, at minimum, you really ought to start (and maybe even advertise) a discussion about your desire to remove huge amounts of content on the talk page first. WhatamIdoing (talk) 21:59, 11 November 2022 (UTC)

I know I said I was done, but WAID raises a point I do want to address… in the Radio example, the disruption was not caused by the removal of uncited information, but rather by the removal of too much uncited information all at the same time! Had the challenger gone step by step - removing a few statements at a time, and given others a chance to fix those before moving to the next batch - I think few would have been upset.

I DO think there is a valid complaint when it comes to ANY mass-editing - even when that editing is 100% in line with policy. Simple volume can overwhelm, and that is disruptive. That, however, isn’t a flaw in policy… it’s a flaw in how zealously someone enforces policy. It’s a behavior problem, not a policy problem. Blueboar (talk) 22:29, 11 November 2022 (UTC)

While I broadly agree with WAID I do think this is a pertinent and constructive point here from Blueboar. Indeed, had the challenger gone step by step, he would have probably been stopped in his tracks as soon as he tried to remove what AM and FM stand for. Because everyone knows what they stand for and yeah, you can tag them cn or even remove them and get reverted. But if someone reverts you because it's WP:BLUESKY to a certain audience and they feel it could be easily cited, the disruptor is now the person citing WP:BURDEN for an obvious fact. FM stands for frequency modulation. If you know the most basic facts about radio and wireless communication, you know this. So removing it, and then insisting that it remain removed because of WP:BURDEN, is disruptive and not simply due to the volume. While I do agree fundamentally, that good behavior can fix bad policy, that doesn't mean the policy is good. Andre🚐 23:08, 11 November 2022 (UTC)

In other words, we still fundamentally disagree. Got it. Blueboar (talk) 01:10, 12 November 2022 (UTC)

I suppose; we agree inasmuch as one should not enforce policy with zealous and rigid enforcement, in general, since Wikipedia: The rules are principles Andre🚐 01:11, 12 November 2022 (UTC)

If such removals were done in small stages (e.g., a paragraph a week), then other editors would either decide that they don't care about the removed paragraph, or they would decide to WP:Let the Wookiee win and spam in a source when reverting it back in. It wouldn't be a huge lift. It wouldn't produce drama. It wouldn't result in readers missing three-quarters of the article while editors are scrambling to find sources with no warning. It also wouldn't result in certain zealots irritating everyone else with their claims that they are so utterly incompetent that they can't even guess whether any material might be verifiable unless there is a little blue clicky number at the end of the sentence.

I'm not sure what the best "rule" would be, but perhaps something like "Mass-blanking of uncited material can be disruptive. If you want to remove more than a couple of paragraphs from an article solely because the material is uncited, you should post a note on the talk page first. This rule does not restrict anyone from removing material that is non-neutral, promotional, or that would otherwise be inappropriate for the article even if it were already well sourced." WhatamIdoing (talk) 19:10, 12 November 2022 (UTC)

+1 Carter (Tcr25) (talk) 00:52, 12 November 2022 (UTC)

"...the disruption was not caused by the removal of uncited information, but rather by the removal of too much uncited information all at the same time!"

And what if I say that, whereas the presence of a small piece of test is a small problem, huge texts without a single reference is a huge problem. If removal of a small piece of uncited text is allowed and encouraged by our policy, removal of big pieces of texts without references should be the more encouraged.

It seems that we are too focused on one specific tag, "cn". In reality our policy clearly and unequivocally says that it should be used to request a citation for an isolated unsourced statement. When a user puts "cn" after each sentence of some section and then blanks that section, that is a violation of the policy. But the violation consists in incorrect usage of the tag, for the policy says

"Other templates exist for tagging sections or entire articles here."

That means if a user sees one sentence that has no inline citation and seems potentially non-verifiable, that user cam place "cn" there, and a possible outcome will be deletion of the sentence. However, when a user sees a whole section that has no online citation, a correct way would be to place an {unreferenced section} template. What should we do with the section that contains that template? Are we allowed to delete it completely if no action has been taken, and/or if good faith googling yielded no sources? Why not? Again, if a small pieces of unreferenced text pose a small problem, big pieces of unreferenced text are a pig problem. How can that removal be considered vandalism?

We can speak about vandalism only if the text that has been removed is obviously verifiable, or (which is pretty much the same) if missing sources are really easy to find. That type behaviour relates to WP:VANDAL or WP:CIR, but this type issues should be discussed at the WP:DE talk page. Paul Siebert (talk) 22:11, 12 November 2022 (UTC)

It's another straw man, I never saw anyone compare the removals of text to vandalism. Andre🚐 23:49, 12 November 2022 (UTC)

Paul, please consider these two approaches:

Article has a lot of uncited information. I don't have any specific reason to doubt the accuracy or verifiability, and I don't choose to SOFIXIT myself.
Surprise! I just blanked half the article, and I'm willing to edit war to keep it out unless and until you do all the work that I don't want to do myself.

versus:

Article has a lot of uncited information. I don't have any specific reason to doubt the accuracy or verifiability, and I don't choose to SOFIXIT myself.
I post a quick note on the talk page that says "Hey, it looks bad to have so much uncited text. If nobody objects, I'm going to blank all the uncited stuff next week/month."
Nobody replied, so I blank half the article.

I'm looking at an article you've edited more than 50 times. There are about 50 paragraphs that I could blank under BURDEN if I apply the "remove any paragraph that doesn't end with a little blue clicky number" approach. Which approach do you want me to take with your article? WhatamIdoing (talk) 02:41, 13 November 2022 (UTC)

Y'all have reinforced what I said. A few of the other considerations are whether or not it is sky is blue, how universal the "sky is blue" situation is amongst readers, whether or not the person appears to be engaged in mass deletions, whether or not the person appears to have a genuine concern about verifiability, whether or not the person discussed it, whether or not the deletion appears to be a part of a pissing war. Sincerely, North8000 (talk) 15:22, 12 November 2022 (UTC)

Blueboar, yes, I've long grown tired of reading this, it is full of BLUDGEON and going in circles. Crossroads ^-talk- 23:40, 12 November 2022 (UTC)

The only bludgeoning or circular reasoning I've seen is the supporters of BURDEN repeatedly saying "JUST PUT A SOURCE THERE! DON'T ADD WITHOUT A SOURCE, AND JUST RESTORE IT WITH A SOURCE!" when that is not at all what is being discussed. Andre🚐 23:48, 12 November 2022 (UTC)

I agree with you but for a different reason. The vast majority of controversial wp:burden deletions are not for unsourced material, they are for sourced material, in tandem with questioning or nitpicking the source. So the "just provide a source" retort in this debate is not really relevant to most controversial deletions. Sincerely, North8000 (talk) 01:07, 13 November 2022 (UTC)

If a source is present, then the BURDEN is satisfied, and you can't CHALLENGE it. You have to use normal dispute resolution methods, which generally look like blanking the bad (in your opinion, of course) material and complaining about the source, but ideally not claiming that BURDEN applies, because it doesn't. WhatamIdoing (talk) 02:30, 13 November 2022 (UTC)

North, I am not sure there is any appreciable difference between texts that have inline citations to sources that do not support them and texts that have no online citations. "Citation needed" and "failed verification" tags have the same effect: both texts may be removed, because they both are de facto unsourced..

I would say, removal of the latter is even more easy: we already have a reference to some concrete source that can be directly compared with the corresponding Wikipedia text. If the former says "A", but the latter says "B", that means that the ostensibly sourced Wikipedia text is not supported by the source cited, which means it is unsourced. But in any case WP:FAIL is applicable here. Paul Siebert (talk) 03:40, 13 November 2022 (UTC)

I disagree with your first paragraph. Burden doesn't apply just to unsourced material, it it clearly apples to sourced material including winning the nitpicking battle battle if somebody is nitpicking the source or how the source does/doesn't supports it. Further, IMHO the majority of controversial applications of burden are on sourced material. Sincerely, North8000 (talk) 14:36, 13 November 2022 (UTC)

Again, it seems you incorrectly understand the term "sourced". If you provide a wrong citation, the material still is unsourced, although the inline citation is present.

But a wrong citation is not a citation.

In other words, the arguments "You provided no citation, so I remove your text", and the "The source provided by you does not support your text, so I remove it" are essentially the same argument.

And our policy says that absolutely clearly:

"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^[1] the contribution. "

If a source does not support the text directly, then BURDEN is not satisfied, and the text is unsourced. Paul Siebert (talk) 17:57, 13 November 2022 (UTC)

@Paul Siebert, you accidentally forgot the footnote at the end of that sentence:

Once an editor has provided any source they believe, in good faith, to be sufficient, then any editor who later removes the material must articulate specific problems that would justify its exclusion from Wikipedia (e.g. why the source is unreliable; the source does not support the claim; undue emphasis; unencyclopedic content; etc.). If necessary, all editors are then expected to help achieve consensus, and any problems with the text or sourcing should be fixed before the material is added back.

We wrote this to stop the "bring me a rock" problem. We do not want editors to be able to declare unilaterally that those sources aren't convincing, so that cited content can be blanked without any explanation or justification and can't be restored per BURDEN. Based on this, the correct interpretation is closer to "The burden...is satisfied by providing an inline citation that the person who is adding the citation believes is a reliable source that directly supports the contribution even if another editor (or a POV pusher) disagrees."

BURDEN (specifically/narrowly) doesn't give you an unlimited right to blank already sourced text. WhatamIdoing (talk) 22:08, 13 November 2022 (UTC)

References

^ A source "directly supports" a given piece of material if the information is present explicitly in the source so that using this source to support the material is not a violation of Wikipedia:No original research. The location of any citation—including whether one is present in the article at all—is unrelated to whether a source directly supports the material. For questions about where and how to place citations, see Wikipedia:Citing sources, Wikipedia:Manual of Style/Lead section § Citations, etc.

I think the difference there is a dispute over the suitability of the source, versus it containing what it's claimed to contain. If someone sources a statement to Joe's Blog, and that entry on Joe's Blog does indeed support what the article text says, people can object that it's not reliable or otherwise unsuitable, but at that point it is not specifically a BURDEN issue. The same would be true if people disagree over how to properly interpret and weight the reference cited. If, on the other hand, the cited reference doesn't verify the article text at all (if, say, someone cites a source in the article about apples that is about dog grooming and never mentions apples at all), then the text is still fundamentally unverified, and that is still BURDEN at that point. Seraphimblade ^{Talk to me} 22:55, 13 November 2022 (UTC)

I wish that you were right. Most sources do not rigorously 100% meet Wikipedia criteria so it quite common to exclude the material while using a quibble about the source to delete the material and using burden to create a high burden to reintroduction North8000 (talk) 01:06, 14 November 2022 (UTC)

In Seraphimblade's example (content about apples vs source about dog grooming), we would likely have a question about whether "they believe, in good faith" that the source is sufficient. It could be a simple mistake (e.g., pasted the wrong URL), but if it's not, then a WP:NOTHERE or WP:CIR discussion might be in that editor's near future.

As North notes, true instances of obviously unsupported content are rare. If you really do encounter such a situation, BURDEN is unnecessary, and BURDEN's footnote tells you exactly how to get it excluded anyway: "articulate specific problems that would justify its exclusion from Wikipedia (e.g. why the source is unreliable; the source does not support the claim...)". The goal is to "achieve consensus", rather than for me to announce that I don't have to lift a finger to help with sourcing, because I can blank anything I want and you can't add it back until you produce a source that satisfies me – and I warn you, there is no source that will ever be reliable enough to write that kind of POV about those Greebertarians (or those Republocrats, or those Demicans) in Wikipedia.

We had to put some limits on BURDEN to stop this kind of abuse. The rare instance of a completely mismatched citation can be (and is, every single day of the year) fixed without BURDEN's support. WhatamIdoing (talk) 03:09, 14 November 2022 (UTC)

Well, if I understood you correctly, the situation described by you is artificial.

Yes, BURDEN allows us to delete every unsourced text, and it cannot be restored until a source has been provided.

Your reference to achieving consensus are not correct: consensus should be achieved for adding, not for removal. If I say that some source does not support the text added by you, it is your job to convince me and other users that I am not right. If you failed to do that, the text cannot be re-added. In contrast, I don't have to convince anybody: I am just saying, e.g.: "I compared the source X with the Wikipedia text Y, and I conclude that the source does not support that statement, so I am removing it." If you (or anybody else) cannot prove that my statement in not justified, I may delete the statement X. That directly follows from BURDEN's letter and spirit.

But BURDEN has some natural limits. The first limit is common sense: thus, we hardly can achieve a consensus that the statement "Paris is a capital of France" needs a source: every minimally educated person knows that, and there is no (and can be no) controversy about that.

The second natural limit is RSN: if I remove some text about Republocrates because the source (the "Gotham globe" article) ostensibly does not support it (or that "Gotham globe" is not reliable), you can always go to RSN and ask: "Is "Gotham globe" reliable for the statement about Republocrates?" If the answer is "Yes", no references to BURDEN can prevent you from restoration of this text.

I am coming to a conclusion that there is absolutely no need to change anything in this policy, and the policy needs no further clarification: if someone challenged some text, it should be removed unless a source has been provided. However, if removal of some text was done in such a way that it rises legitimate doubts in the editor's good faith or competence, that user can be reported at relevant noticeboards and potentially sanctioned. Paul Siebert (talk) 04:38, 14 November 2022 (UTC)

In re consensus should be achieved for adding, not for removal: No. By the time you get this far into the dispute, consensus should be achieved for what to do. Editors (not just one of them) should achieve consensus regardless of whether any individual wants to add, remove, or change it. WP:ONUS (which is not BURDEN) says that it's duty of the person who wants to add the content to push for a consensus to include it.

The exact scenario that we want to avoid in BURDEN is If I say that some source does not support the text added by you, it is your job to convince me, because there are individual editors who cannot be convinced, no matter what source is provided. BURDEN is not meant to be a gift to POV pushers, nor is it mean to encourage to the occasional person who thinks that it's a policy violation to Wikipedia:Use our own words. Once someone's added a source that they think supports the sentence, please move on to the entire rest of the policy. Don't try to make BURDEN do all the work by itself. WhatamIdoing (talk) 20:31, 14 November 2022 (UTC)

You contradict to yourself. You correctly note that WP:ONUS says that it's duty of the person who wants to add the content to push for a consensus to include it, but ONUS says nothing about removal. Similarly, BURDEN says about the editor who adds or restores material, but it says nothing about those who remove them.

And that is a crucial factor for creation of a good quality content: that can be achieved only when addition is more difficult than removal.

BURDEN can never be a gift for POV-pushers, because they are engaged mostly in addition of questionable materials, not removal. And, BURDEN clearly outlines conditions that must be met. It seems you incorrectly interpreted by "convince others" as "obtain their permission". No, I didn't mean that. "Convince" means to demonstrate that "your 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 WP:NOR." As soon as such a source has been provided, further opposition may be reported at relevant noticeboard. However, you must provide a proof (vide supra) that allows any good faith user to see that your text is well sourced. The words "good faith" are important here, for this policy keeps in mind that category of users. Paul Siebert (talk) 22:15, 14 November 2022 (UTC)

POV-pushers, because they are engaged mostly in addition of questionable materials, not removal: This may be technically true (in the sense that 51% is "mostly"), but it is not sufficiently true for the problem to be negligible. You don't edit WP:AP2 subjects, right?

The policy actually says, verbatim, that "Once an editor has provided any source they believe, in good faith, to be sufficient," then you can't just blank it with no explanation. If they've already cited it, that step has been achieved. WhatamIdoing (talk) 06:18, 15 November 2022 (UTC)

I agree with @Seraphimblade. I think he meant that in an ideal world that is how it is supposed to work, and I was going to say something similar because a source could also partially verify a statement as well. It isn't always black and white as being either a source supports a statement, or is unsourced if it doesn't since it might support one thing, but not another. However, in the real world there is a practical difference like North pointed out. Huggums537 (talk) 01:14, 14 November 2022 (UTC)

Actually, the policy already contains something that would solve most of the noted problems except that it is written in such a weak and lengthy manner that nobody follows it. It says "please state your concern that it may not be possible to find a published reliable source, and the material therefore may not be verifiable." Maybe shortening and strengthening this provision would be a good thing. Sincerely, North8000 (talk) 20:53, 14 November 2022 (UTC)

I think that clause could be strengthened to advise against not doing that. Andre🚐 21:38, 14 November 2022 (UTC)

As I already said, it would be desirable to specify that if no citation is provided, and it is not easy to find it in a reasonable amount of time, the unreferenced statement is deemed unverifiable. If a reference is provided, but the source is not easy to find (for example, it is some non-free scholarly journal), the text cannot be considered non-verifiable: it is still possible to go to a library or to pay some reasonable money to get it, and it is much, much easier than goodling for some information without any clue on where exactly it was taken from by a user who added it. Paul Siebert (talk) 22:21, 14 November 2022 (UTC)

I do think a distinction should be drawn there. If a reference is provided, but the reader can't get to it at the moment, that's not unverified. That can be done via a library or the like (and I've done exactly that, and generally it has checked out). BURDEN is about unreferenced material, not "I don't have instantaneous free access to the reference cited". Seraphimblade ^{Talk to me} 01:21, 15 November 2022 (UTC)

[directly_supports-1] A source "directly supports" a given piece of material if the information is present explicitly in the source so that using this source to support the material is not a violation of Wikipedia:No original research. The location of any citation—including whether one is present in the article at all—is unrelated to whether a source directly supports the material. For questions about where and how to place citations, see Wikipedia:Citing sources, Wikipedia:Manual of Style/Lead section § Citations, etc.

[1]