Wikipedia talk:Editing policy#Edit summaries for reverts

This is a bit of a mistake I've made in my Huggle sessions. Most people who add unsourced material that sounds true are completely new to Wikipedia. Usually, it's not only true, but the source is out there. Looking for a reliable source (and knowing which RS guideline applies) will take effort most people are unwilling to put in. People who add unsourced material are generally here to build an encyclopedia, they're just bold. (Of course, there's the defamation issue, but although you should revert unsourced claims on BLPs you should paste them somewhere first so you can research them.) T3h 1337 b0y 02:44, 21 September 2024 (UTC)

:I typically revert newly added unsourced edits where I'm not 100% sure they're true and leave a note for the editor who added the information, on the grounds that if they added the information to begin with, then they're probably most qualified to know from where they found the information. DonIago (talk) 05:34, 21 September 2024 (UTC)

:In my experience, most instances of editors making totally unsourced additions to articles will fall under one of the following scenarios:

:# The addition boils down to a series of claims with dubious verifiability

:# The added material is very likely to be undue within the article.

:# After being reverted, the material is readded in a constructive manner that includes viable inline citations

:I think there's a very real difference in net editor-work required to improve the encyclopedia between adopting a GIGO stance that insists on new iterative additions being adequately cited, and going out of one's way to remove existing uncited but plausibly WP:V material of unknown provenance—which often makes it harder for later editors to fill in the gaps of substance that are often created. Remsense ‥ 论 07:44, 21 September 2024 (UTC)

::Repeating my wish for a tool that would display each of the major additions throughout an article's history that have since been removed. I would make tough choices with much less dread if I knew it was easier for others to see and evaluate what is no longer in an article if they so choose. Remsense ‥ 论 08:04, 21 September 2024 (UTC)

:My approach is similar to Doniago's. If it seems like a plausible and useful addition, I'll do a quick search to see if I can add a source myself (and sometimes succeed). I don't know about Huggle, but Twinkle lets you select Reverted good-faith edits, which is a bit gentler. Schazjmd (talk) 13:32, 21 September 2024 (UTC)

:Right. One alternative at WP:FIXTHEPROBLEM is tagging. - Butwhatdoiknow (talk) 15:52, 21 September 2024 (UTC)

::I'll tag if the content isn't newly added (i.e. if it's a stable part of the article), but in the majority of instances where I've tagged, the upshot is that a couple of months later I'll remove the still-unsourced content. That said, I've been editing long enough that 'majority of instances' allows for a fair number of instances where citations were subsequently added. DonIago (talk) 23:36, 21 September 2024 (UTC)

:::The post that started this conversation talks about edits that are "not only true, but the source is out there." Why remove the still-unsourced content? - Butwhatdoiknow (talk) 02:43, 22 September 2024 (UTC)

::::Is "new editors should be strongly encouraged to always cite their contributions, as to not to inherently create more work someone else has to do" an acceptable answer? Pragmatically, I think it is. (While tagging may be encouragement, reversion is clearly stronger.) Secondly, in the grand scheme of things the class of clearly V statements is actually pretty small, and I generally don't trust myself to produce verifiable prose if I'm not looking at a source. Plus, there's also the DUE issue. Remsense ‥ 论 02:50, 22 September 2024 (UTC)

:::::Wikipedia:Please do not bite the newcomers. - Butwhatdoiknow (talk) 05:40, 22 September 2024 (UTC)

::::::I firmly reject treating reversion like a "bite" in itself: I am putting the page into a better state than it was presently in, and there is nothing to apologize for in that as long as my reasons are adequately communicated—in this case adequately for an editor still very much getting the hang of things. Remsense ‥ 论 05:43, 22 September 2024 (UTC)

:::::::Reject all you want, but it's example 1 at Wikipedia:Please do not bite the newcomers#How to avoid being a "biter". - Butwhatdoiknow (talk) 05:51, 22 September 2024 (UTC)

::::::::What provides more benefit to a new editor, though? Fixing their edit for them, which they may not even notice, or reverting their edit and bringing Wikipedia's sourcing policies to their attention so that they (ideally) won't contine to add unsourced material and possibly be reverted in a less gracious manner? DonIago (talk) 06:00, 22 September 2024 (UTC)

:::::::::These are not the only two options. - Butwhatdoiknow (talk) 15:23, 22 September 2024 (UTC)

::::::::::I didn't say they were...but if you're going to note that there's other options, perhaps you could bring them up? DonIago (talk) 00:37, 23 September 2024 (UTC)

:::::::::::Example 1 at BITE says:

:::::::::::::Improve, don't remove. If something doesn't meet Wikipedia's standards, try to fix the problem rather than just remove what's broken. (Nothing stops new contributors and regulars from coming back like having all their hard work end up in the bit bucket.)

:::::::::::If you click on the "fix the problem" link you'll find a list. - Butwhatdoiknow (talk) 05:37, 23 September 2024 (UTC)

::::::::::::None of that is clearly intended to apply to new edits though. As I've said, I treat unsourced content differently if it isn't new to the article. DonIago (talk) 03:53, 24 September 2024 (UTC)

::::As I said in my initial message in this thread, if something is 100% true and I'm 100% sure a source is out there then I might leave it alone. But that's a rare occurrence. I'd also note that it's almost always more expedient to just add a source when content is challenged than it is to argue about whether a source is needed, and very often when editors do question the need for a source I find myself questioning whether they're making that argument because they don't themselves believe that a source exists. Put simply, if the editor arguing for inclusion of content isn't willing or able to provide a source (or at least demonstrate that they want to collaborate to find one), then I feel that punches a large hole in any arguments that it's appropriate to include the content. DonIago (talk) 03:11, 22 September 2024 (UTC)

:::::Right, an editor who is challenged has the obligation to produce an RS. But how should you challenge? Reverting from the get-go is easy, fun, and likely to cause hurt feelings on the part of the reverted (especially if they're a newbie). All I'm asking is that editors consider alternatives first. - Butwhatdoiknow (talk) 05:45, 22 September 2024 (UTC)

::::::Honestly, if you're new and you're going to have your feelings hurt because someone reverted you while telling you relatively gently that you need to provide a source (and again, this shouldn't be a significant burden), then I think you're in for a rude awakening if you continue to edit in any meaningful capacity. There's a lot of policies and guidelines on Wikipedia that it's easy to fall afoul of (frankly I'd rather have a single sentence reverted because I didn't provide a source than have the five hundred words I added to a film's plot summary reverted per WP:FILMPLOT), and I think new editors should expect that they might accidentally run into some of those P&Gs. Similarly, anyone who sticks around long enough is, unfortunately, probably going to run into editors who will be a lot more rude than an editor who gives someone a Level 1 advisory for having added unsourced content. DonIago (talk) 05:58, 22 September 2024 (UTC)

:::::::Yes, a first revert that is accompanied by a relatively gentle explanation is less likely to scare off a newbie than one that is a lot more rude. But I believe the point of the BITE essay is that someone whose very first edit is reverted is not likely to stick around at all regardless of the tone of the explanation. You may disagree - or you may feel that an editor who cannot handle a "gentle" revert of their very first edit is too thin-skinned and it is best to scare them off from the get-go. - Butwhatdoiknow (talk) 15:41, 22 September 2024 (UTC)

::::::::I've heard that there's been research on this point, and the answer appears to be:

::::::::* If you ignore it: Newcomer will probably not add a source.

::::::::* If you tag it: Newcomer might add a source, usually the next day.

::::::::* If you revert it: Newcomer is unlikely to ever edit Wikipedia again.

::::::::The mechanism seems to be that promising newcomers check back the next day to see whether their contribution has been accepted. They're happy if it's accepted, and if there's a tag indicating a problem, they're motivated to see if they can address the problem. The inline {{tl|citation needed}} are particularly effective in this circumstance, as it's something that people have usually seen before and (more or less) understand what's being asked for. Small copyedits are also effective, because the general category is "someone saw my contribution and didn't reject it".

::::::::On the flipside, if you revert it, they assume that they did something wrong, and they give up. Blanking a first contribution, even though it's sometimes necessary, and even if it's accompanied by a nice note (which they might not see, and which often aren't all that nice anyway), is one of the most effective ways to get new editors to stop editing. WhatamIdoing (talk) 18:15, 28 March 2025 (UTC)

With a total of four editors in favor and zero against, the editors at wp:Consensus agree that wp:NOCON should move from that page to wp:EP (discussion at wt:Consensus#Moving NOCON to Editing Policy). What say the editors at this page? - Butwhatdoiknow (talk) 00:05, 6 January 2025 (UTC)

• Support. - Butwhatdoiknow (talk) 00:05, 6 January 2025 (UTC)
• I do not see that there is a consensus for it, and you'd need to advertise a centralized discussion to get a sufficient WP:CONLEVEL. Andre🚐 00:08, 6 January 2025 (UTC)
• :I would say the discussion at wt:Consensus can be better summarized as “no one expressed any opposition to moving it”. However, the number of people responding to that RFC was quite small… so, yeah, hopefully it will gain more participants here. Blueboar (talk) 01:53, 6 January 2025 (UTC)
• :CONLEVEL applies to a change that would override a community consensus. Is there a community consensus somewhere that NOCON must be in Consensus? - Butwhatdoiknow (talk) 03:15, 6 January 2025 (UTC)
• ::There’s not. Someone put it there, unilaterally, not knowing where better to put it. SmokeyJoe (talk) 04:18, 6 January 2025 (UTC)
• :::Well, perhaps not unilaterally, since it was discussed multiple times on multiple pages. But the specific line enshrining WP:QUO could perhaps be described as a unilateral-over-objections addition. WhatamIdoing (talk) 05:07, 6 January 2025 (UTC)
• ::::Whatever issue there may be about "enshrining" QUO exists whether NOCON is in Consensus or Editing Policy (or Dispute Resolution). In short, that issue isn't relevant to THIS discussion. - Butwhatdoiknow (talk) 06:31, 6 January 2025 (UTC)
• :::::I agree. WhatamIdoing (talk) 06:35, 6 January 2025 (UTC)
• I don't think it fits well on the current version of this page. The only mentions of consensus is a tangential one on mass creation of articles, and a sentence saying {{tq|Bold editing does not excuse edits against existing consensus...}}. Without discussion of consensus as a decision-making mechanism, there would be a lack of context. As the last sentence of {{section link|Wikipedia:Editing policy|If you need help}} directs editors to Wikipedia:Dispute resolution if they {{tq|need help reaching an agreement with other editors}}, perhaps that page would be a better fit. isaacl (talk) 03:12, 6 January 2025 (UTC)
• :I agree, but is it more so out of place where it is. When I looking at improving it at WP:CONSENSUS, I can only see myself cutting it back, severely. The same may happen here, but here, the focus on “how to edit” matches better, rather than there theoretical/philosophical/idealistic stance there. SmokeyJoe (talk) 04:14, 6 January 2025 (UTC)
• :Think how silly it would be if the last sentence of Wikipedia:Editing policy § If you need help were to direct editors to WP:CONSENSUS. SmokeyJoe (talk) 04:17, 6 January 2025 (UTC)
• Support. It’s simple logic. NOCON is about who to do (how to edit) in the absence of an explicit consensus case of a declared explicit “no consensus”. It looks in the direction of what to now in the meantime. As it is not advice on how to develop consensus, it does not belong under the title “Consensus”. This page is a great location because: (i) this page deals with practical pragmatisms; (ii) this page has the same “Policy” taggery, and altering the weight of what is written under “NOCON” is no one’s stated intent. SmokeyJoe (talk) 04:10, 6 January 2025 (UTC)
• I support moving it out of WP:CONSENSUS. Where exactly it ends up is less important to me. I think it could make a fine {{tl|information page}}, for example. It is important to me that the {{tl|info}} message come with it. Editors need to understand that this is a handy summary of pre-existing rules, and not a separate set of rules. WhatamIdoing (talk) 05:11, 6 January 2025 (UTC)
• :Agree. I am more supportive that WP:Editing policy is the right page, and agree with that it is a handy summary of old accepted practices. I think it is important that WP:CONSENSUS forever point to this documentation, without the implication that it is mandated by the policy WP:CONSENSUS. SmokeyJoe (talk) 06:36, 6 January 2025 (UTC)
• I think the discussion needs to be better publicized: through a listing on a village pump or WP:CENT. Maybe put an RFC template on it so it gets feedback requests. Andre🚐 09:37, 6 January 2025 (UTC)
• :Sure. Who knows how to do that? SmokeyJoe (talk) 10:30, 6 January 2025 (UTC)
• ::Can do both, treat this convo as an RFCbefore, open an RFC (here?) and then add it to CENT. Selfstudier (talk) 10:48, 6 January 2025 (UTC)
• :::You can just add an RFC template to the top of this discussion. There's no need to re-start it. WhatamIdoing (talk) 18:53, 6 January 2025 (UTC)
• ::::[https://en.wikipedia.org/w/index.php?title=Wikipedia%3AConsensus&diff=1267685974&oldid=1265370047 Meantime] Selfstudier (talk) 19:10, 6 January 2025 (UTC)
• Oppose moving the entire No consensus after discussion subsection (permanent link): most of its items are not relevant to WP:Editing policy. They already link to their specific policies, guidelines, or other pages. I would be more open to copying only {{tq|When discussions of proposals to add, modify, or remove material in articles end without consensus, the common result is to retain the version of the article as it was prior to the proposal or bold edit.}} I recommend refining the proposal before publicizing it. I skimmed through the previous discussion at WT:Consensus#WP:NOCONSENSUS, but I did not follow the reasoning. Flatscan (talk) 05:21, 7 January 2025 (UTC)
• :Complying with your point is on the mark, and is not a reason to not move what is relevant. Note that none of it is relevant where it is. SmokeyJoe (talk) 09:11, 7 January 2025 (UTC)

:{{nayc}} Oppose. I don't think a lack of consensus is related to the editing policy. Also, per what Flatscan said, maybe put only the part about discussions of proposals end without consensus? The Master of Hedgehogs (talk) (contributions) (Sign my guestbook!) 21:08, 26 February 2025 (UTC)

• Comment: If this is upgraded to editing policy, then the text should be updated to address edits outside articlespace other than deletion. For example, this RfC on edits to talk pages ended in no consensus, and the current text of WP:NOCON does not really account for this scenario, nor does it address vandalism. Gnomingstuff (talk) 20:54, 3 April 2025 (UTC)
• :I don't think that moving something from the Wikipedia:Consensus policy to the Wikipedia:Editing policy is "upgrading" it.
• :It should be easy enough to form a consensus against actual vandalism, no? Have you ever seen a "no consensus" result involving a vandalism dispute? WhatamIdoing (talk) 22:38, 3 April 2025 (UTC)
• ::Nope, would be interesting if one did end that way though! I don't think that no consensus from a vandalism dispute would happen. Most vandalism cases I have seen are fairly straightforward except for a few, but they got worked out eventually. Sheriff U3 01:01, 4 April 2025 (UTC)

:* Support I think that this would fit better in the editing policy. Also if it needed to be expanded to include more scenarios then that can be done after the fact. It is much easier to split a small section then a big one. Plus we could keep a summary at WP:Consensus if we wanted to and have the main part at WP:Editing policy, just like what we do with articles in the main space.

:Sheriff U3 01:10, 4 April 2025 (UTC)

• Oppose If I were looking for information on how to handle a lack of consent, I'd be looking under pages associated with consent. It would never occur to me to look for it under editing policy. (Is there a grand index? There should be.)— Preceding unsigned comment added by Ghost writer's cat (talk • contribs) 02:58, 13 April 2025 (UTC)

Where NOCON is the balancing policy to WP:ONUS, I'm concerned that moving NOCON would somehow lead to longstanding content being removed more easily? Kolya Butternut (talk) 13:02, 13 April 2025 (UTC)

:Why are you concerned about this?

:It's currently on a page that says "policy" at the top.

:If it's moved here, it will still be on a page that says "policy" at the top.

:Why would that result in any change to editors' behavior? WhatamIdoing (talk) 22:35, 14 April 2025 (UTC)

::Do you want to move WP:ONUS out of V? Kolya Butternut (talk) 00:26, 15 April 2025 (UTC)

:::I want the community to decide whether they actually believe ONUS is the right rule, and make all the policies and guidelines align with their decision. WhatamIdoing (talk) 00:54, 15 April 2025 (UTC)

::::why not answer the question? I can't really answer yours except that consensus seems like the stronger policy than the editing policy. Kolya Butternut (talk) 02:25, 15 April 2025 (UTC)

:::::I don't care whether the ONUS sentence is in WP:V or in some other policy page. The exact location is unimportant. For example:

:::::* The ONUS sentence and shortcut could be kept in the {{tl|policy}} page Wikipedia:Verifiability, and it would still be part of our written policies, and editors would still invoke it when they wanted to remove content.

:::::* The ONUS sentence and shortcut could be moved to the {{tl|policy}} page Wikipedia:Consensus, and it would still be part of our written policies, and editors would still invoke it when they wanted to remove content.

:::::* The ONUS sentence and shortcut could be moved to the {{tl|policy}} page Wikipedia:Editing policy, and it would still be part of our written policies, and editors would still invoke it when they wanted to remove content.

:::::* The ONUS sentence and shortcut could be moved to the {{tl|policy}} page Wikipedia:What Wikipedia is not, and it would still be part of our written policies, and editors would still invoke it when they wanted to remove content.

:::::Its location does not matter. What I want to know is: Does the community actually, fully, completely want this to be the rule, with not just the advantage of selectively invoking it to win a dispute, but also its obvious disadvantages (e.g., losing decent content because you can't get enough people to join a discussion about whether there's consensus for the decent content)? WhatamIdoing (talk) 17:14, 15 April 2025 (UTC)

::::::I disagree with you on the meaning of ONUS. I think any editor who wants to make a change to consensus has the onus. I think in that context include meant add. Kolya Butternut (talk) 18:14, 15 April 2025 (UTC)

:::::::The ONUS sentence says:

:::::::* "The responsibility for achieving consensus for inclusion is on those seeking to include disputed content."

:::::::Are you interpreting it as meaning "The responsibility for achieving consensus for removal is on those seeking to remove disputed content"? While I think that's a reasonable interpretation of the Wikipedia:Editing policy (specifically the WP:EPTALK section, "If someone indicates disagreement with your bold edit...") and Wikipedia:Consensus in general, I don't think that's a reasonable interpretation of the ONUS sentence itself. WhatamIdoing (talk) 19:57, 15 April 2025 (UTC)

::::::::I think while ONUS articulates that, our policies also indicate:

::::::::*"The responsibility for achieving consensus for exclusion is on those seeking to exclude disputed content."

::::::::There is no thumb on the scale against inclusion. ONUS is in V just to indicate that if an editor wants to add something, verifiability isn't enough, they need consensus. Kolya Butternut (talk) 06:09, 16 April 2025 (UTC)

:::::::::"ONUS is in V just to indicate that if an editor wants to add something, verifiability isn't enough . . ." That was likely the original intent of the wp:ONUS sentence. But the plain text has since taken on a life of its own, free from the context provided by the sentences above it. - Butwhatdoiknow (talk) 15:43, 16 April 2025 (UTC)

::::::::::^ This. For better or worse, we haven't been able to get experienced editors to agree that ONUS or any other policy specifically treats removal of disputed (long-standing or otherwise) material as something that requires the would-be remover to prove consensus for. WhatamIdoing (talk) 16:02, 16 April 2025 (UTC)

:::::::::::{{tq|or any other policy specifically}}; huh? Any change requires determining consensus. Kolya Butternut (talk) 16:19, 16 April 2025 (UTC)

::::::::::::But not every change requires the person making the change to be responsible for achieving consensus. Every change requires someone to achieve consensus; an ONUS addition requires the person making the addition to achieve consensus. WhatamIdoing (talk) 16:24, 16 April 2025 (UTC)

:::::::::::::In the case of an addition to restore consensus, the onus is on the person seeking to exclude. Kolya Butternut (talk) 21:02, 16 April 2025 (UTC)

::::::::::::::I'm not sure what that means. Do you mean:

::::::::::::::* We have a major discussion that decides the article should say "Most people think coffee tastes good".

::::::::::::::* Soon afterwards, someone removes that (i.e., is definitely editing against consensus).

::::::::::::::* Someone reverts the removal, pointing to the discussion on the talk page.

::::::::::::::* Since the removal was definitely made against consensus, the person editing against consensus needs to demonstrate that consensus has changed.

::::::::::::::Or do you mean:

::::::::::::::* Someone boldly dumps something bad in an article.

::::::::::::::* Nobody notices for years.

::::::::::::::* You eventually notice that it's bad and remove it.

::::::::::::::* Someone else reverts you and claims you have to get proof of consensus to remove it, but they don't have to do anything, because if a random bit of content persists long enough, then there's "consensus" for it, and you have to prove that consensus has changed before you get to remove disputed content, but they don't have to do anything to re-insert disputed content, even though the ONUS policy says that the person seeking to re-insert the disputed content is the one who needs to demonstrate that consensus supports re-inserting it.

::::::::::::::WhatamIdoing (talk) 22:15, 16 April 2025 (UTC)

:::::::::::::::Sort of yes to both. But where consensus is not clear everyone has to do the work to determine consensus. And this is false: {{tq|ONUS policy says that the person seeking to re-insert the disputed content}}. Kolya Butternut (talk) 22:46, 16 April 2025 (UTC)

::::::::::::::::Imagine that Alice wants to remove something from an article, and Bob wants to put it back.

::::::::::::::::When you read the literal words of the ONUS sentence, as written, do you think ONUS says that the responsibility for achieving consensus is on the editor who is seeking to include the disputed content, or on the editor who is seeking to exclude the disputed content? WhatamIdoing (talk) 01:27, 17 April 2025 (UTC)

:::::::::::::::::Shouldn't the policy explain that it depends and what it depends on? Roughly, it depends on what the strength and age of the consensus is for the content and whether it has any reason such as BLP that limits it. Or substitute that with something more accurate if I am mis-stating that. Andre🚐 01:53, 17 April 2025 (UTC)

::::::::::::::::::If we wanted a rule that says the person responsible for finding consensus depends on various circumstances, then the policy should say that. But the ONUS policy actually says quite the opposite: It lays out one (1) single point, that is not the strength of past consensus, the age of past consensus, or whether there is any reason such as BLP.

::::::::::::::::::ONUS says that the only thing that matters is: If Alice wants to exclude and Bob wants to include, then ONUS says the job of achieving consensus is Bob's duty, full stop. WhatamIdoing (talk) 02:30, 17 April 2025 (UTC)

:::::::::::::::::::In practice, that is not how it works. In practice that would be no consensus for any change to the status quo. Andre🚐 02:31, 17 April 2025 (UTC)

::::::::::::::::::::The gap between practice and written policy is one of the things that concerns me. We say ONUS but do QUO. We when ask editors to correct the written policy to match actual practice, they refuse because they want ONUS as the rule ...when they find that convenient. WhatamIdoing (talk) 23:56, 17 April 2025 (UTC)

:::::::::::::::::::::Agreed. That is why I think ONUS should be amended to explain when and how it is meant to apply. Andre🚐 23:58, 17 April 2025 (UTC)

::::::::::::::::::::::I'm pretty sure that if we had an RFC next month saying "Shall we amend ONUS to explicitly say that it applies even to someone who is reverting to a long-standing but now-disputed version?", people would answer yes.

::::::::::::::::::::::And, although I'm less confident about this prediction, I would not be surprised if we asked the opposite (after a long enough delay, or in an alternate universe), people would also answer yes. WhatamIdoing (talk) 01:14, 18 April 2025 (UTC)

:::::::::::::::::::ONUS does not say that "the only thing that matters". You're making inferences. I could also infer that ONUS refers to new additions, and simply doesn't speak to anything else. Kolya Butternut (talk) 02:49, 17 April 2025 (UTC)

::::::::::::::::::::ONUS lists nothing else that matters. WhatamIdoing (talk) 23:56, 17 April 2025 (UTC)

:::::::::::::::::It depends. But ONUS is not a conduct policy; it just reiterates part of existing conduct PAGs. Kolya Butternut (talk) 02:12, 17 April 2025 (UTC)

::::::::::::::::::What "part of existing conduct PAGs" does ONUS actually reiterate? Not "well, it says something about consensus"; what policy or guideline actually says the same thing as ONUS, namely that the responsibility for finding consensus, in the case of an inclusion/exclusion dispute, is on the person who favors inclusion and not on the person who favors exclusion?

::::::::::::::::::Wikipedia:Consensus, for example, says nothing like that AFAICT. WhatamIdoing (talk) 02:26, 17 April 2025 (UTC)

:::::::::::::::::::{{tq|If your first edit is reverted, try to think of a compromise edit that addresses the other editor's concerns. If you can't, or if you do and your second edit is reverted, create a new section on the associated talk page to discuss the dispute.}} This section of WP:CON sounds like the onus is on the editor making a new change. Kolya Butternut (talk) 02:54, 17 April 2025 (UTC)

:::::::::::::::::::{{tq|is on the person who favors inclusion and not on the person who favors exclusion?}} I'm saying ONUS doesn't say this. Kolya Butternut (talk) 02:56, 17 April 2025 (UTC)

::::::::::::::::::::(Meaning it doesn't put the thumbs on the scale for inclusion or exclusion without further context.) Kolya Butternut (talk) 10:29, 17 April 2025 (UTC)

No one knows what ONUS means. Perhaps we should return to the subject of this section: Where, if anywhere, to move NOCON. - Butwhatdoiknow (talk) 16:43, 17 April 2025 (UTC)

:Anywhere as long as ONUS and NOCON are in the same place Kolya Butternut (talk) 17:48, 17 April 2025 (UTC)

:What they said (policy page). Selfstudier (talk) 18:15, 17 April 2025 (UTC)

::Ah, but which policy page? - Butwhatdoiknow (talk) 18:35, 17 April 2025 (UTC)

:::Well my proposal was to move the ONUS shortcut to a new section at Consensus (and leave the text at V the same), so both ONUS and NOCON would be there. This would be the simplest solution. Kolya Butternut (talk) 02:04, 18 April 2025 (UTC)

::::While I do like the idea of moving the shortcut somewhere else like NOCON, I think wherever it is moved should also contain an expanded explanation of ONUS to help alleviate the confusion it causes today. And if that's not possible, then strongly consider doing away with it altogether. I've seen some of Wikipedia's most experienced, well-meaning editors over the years share a variety of interpretations that, on many occasions, radically diverge. Having followed many of these discussions, I'm convinced it's possible to Wikilawyer your way out of any situation in which you want ONUS to conform to your position.{{pb}}If we can't reach a common understanding over its meaning or proper use, how can we truly expect anyone else will? -- GoneIn60 (talk) 02:29, 18 April 2025 (UTC)

::::What would the "new section" say? - Butwhatdoiknow (talk) 02:56, 18 April 2025 (UTC)

:::::It's in my proposal, {{tq|The onus to achieve consensus for changes to consensus is on those seeking the change.}} Kolya Butternut (talk) 04:37, 18 April 2025 (UTC)

::::::I'm wondering if a better solution would be to move away from language saying whether content should be kept or remove entirely (e.g. get rid of both NOCON and ONUS). Instead wrote a new section discussing what to do in such situations that's neutral about which one is right; don't revert without good cause, don't edit war, if multiple editors revert you then you need to find consensus to (keep/remove) the content. Have main details here or in CONSENSUS (whichever is more appropriate) and link to it from whatever page currently has wording it would replace.
The most contentious use of either appears to be when it's a WP:1AM situation. In those cases the 'majority' should determine whether the content stays or goes, and anyone wanting to change that needs to find consenus for doing so (obvious exceptions would apply e.g. BLP, obvious vandalism, BURDEN, etc). -- LCU ActivelyDisinterested «@» °∆t° 11:39, 18 April 2025 (UTC)

:::::::{{tq|move away from language saying whether content should be kept or remove entirely}}; my suggestion does this? Kolya Butternut (talk) 12:02, 18 April 2025 (UTC)

::::::::Your version seems to imply that editors need permission to edit, or at least I can see editors trying to use it that way. I'd rather something with more guidance, and that it be the only wording on the subject instead of the current situation where editors use both ONUS and NOCON/QUO. -- LCU ActivelyDisinterested «@» °∆t° 12:55, 18 April 2025 (UTC)

:::::::::Maybe ideally, but I think my proposal is a low barrier way to move things around without actually changing anything, and they could be edited from there? Kolya Butternut (talk) 13:22, 18 April 2025 (UTC)

::::::::::I would oppose it as written, editors would use the new wording to say articles can't be edited without first gaining consensus. It's not like ownership and stonewalling are any less any issue than misusing ONUS. -- LCU ActivelyDisinterested «@» °∆t° 14:31, 18 April 2025 (UTC)

:::::::::::What about saying that the onus to provide support for a proposed change to consensus is on those seeking the change? Or some other tweak that just clarifies editors need to provide a rationale for their proposals, not that they have the only responsibility to achieve consensus either way. Kolya Butternut (talk) 14:42, 18 April 2025 (UTC)

:::::::This ventures down the right path in this hypothetical, but we need to consider the guidance that already exists to avoid unnecessary duplication.

:::::::*"Don't revert without good cause" – already covered in EDITCON

:::::::*"Don't edit war" – also covered in EDITCON

:::::::*"If multiple editors revert you..." – EDITCON covers this too, whether it's the same editor reverting you or multiple doesn't matter.

:::::::*"The majority should determine whether the content stays or goes" – This intent is described in DISCUSSCONSENSUS as an ideal outcome. However, when the "consensus-forming process" stalls or results in no agreement (i.e. there is no majority), the removal of NOCON would create a gap here.

:::::::--GoneIn60 (talk) 18:10, 18 April 2025 (UTC)

::::::Given that presumed consensus arises whenever an edit is made, what is the difference between "changes" and "changes from consensus"? - Butwhatdoiknow (talk) 16:11, 18 April 2025 (UTC)

:::::::I suppose none, but saying that editors have the onus to achieve consensus for "changes" may be interpreted to make silent consensus equal to consensus. Kolya Butternut (talk) 16:40, 18 April 2025 (UTC)

::::::I'd like to remove all implicit references to "who should start the discussion on the talk page", or replace them with something like "If there is a dispute, and you wonder who is supposed to start the discussion to determine consensus, then it is always your job to start the discussion. We have established this rule because of the number of people who revert changes or remove content but think it is too onerous for them to write a short comment on the talk page. This means you, and if you make a pest of yourself by saying that it's the other editor's duty to start the discussion instead of yours, then we might decide to block you for being disruptive, tendentious, and uncollegial."

::::::If we could free ourselves from the bickering over why the other guy is supposed to start the discussion instead of me, then we could probably re-write ONUS to something like "Even when the content is verifiable and cited to a reliable source, editors may not necessarily agree to include it in that (or any) article. Other policies and guidelines, such as WP:NOT, WP:CON, WP:DUE, also apply".

::::::I think Kolya's text has two problems:

::::::# That's not what ONUS says, in that ONUS is narrowly concerned about "inclusion", not "change". ONUS does not assign responsibility for (e.g.) a change that rearranges the order of the sections on a page. ONUS only applies if someone wants to put some verifiable content into the article, over someone else's objections.

::::::# That's not what ONUS says, in that ONUS assigns a responsibility to the person who adds content (or wants to), not to someone who "changes" content. If Alice adds content, Bob rearranges it, and Chris removes it, those are all "changes". ONUS only applies to Alice's action. Kolya's proposal applies to all of them. And it opens the doors for disputes about which version is "the" status quo version, with the result that all of the editors can claim that their version is "not change": Chris reverted, which takes it back to status quo; Bob rearranged, which didn't add anything, so it's "not really a change", and Alice says that two other people have edited since her, and that makes hers be the status quo and theirs (especially Chris's) be "the" change, so Chris has to start the discussion. In other words, when you make something be everyone's job, it ends up being nobody's job.

::::::WhatamIdoing (talk) 18:00, 18 April 2025 (UTC)

:::::::That is an acceptable rewrite of ONUS, possibly even great, essentially wiping out the need for a dedicated ONUS shortcut at WP:V. The rewrite would become an extension of VNOT, allowing the section to stay on message and not stray unnecessarily or too far into other aspects that are best covered in more detail at EP or CON.{{pb}}As for updating guidance elsewhere, such as CON, I would hope that most already agree that we should be less focused in policy to assign responsibility, except to say that it is everyone's equal responsibility to start the discussion. --GoneIn60 (talk) 18:51, 18 April 2025 (UTC)

::::::::Maybe "Other policies and guidelines, such as WP:NOT, WP:CON, WP:DUE, may result in the exclusion of material even if it is verifiable and cited to a reliable source". WhatamIdoing (talk) 20:24, 18 April 2025 (UTC)

:::::::::Thought about this a little more... VNOT essentially includes all this already (aside from mentioning policy examples), right? So then it becomes, do we want to insert those examples (NOT and DUE) into VNOT or leave it alone? What happens to ONUS after it is removed?{{pb}}You've put forth a good argument on why we should consider abandoning ONUS and the like altogether, but do we still need to retain a figurative "finger on the scale" that tips the balance slightly in favor of exclusion? If not, then how do we ensure a consistent outcome in disputes of "no consensus"? NOCON says "the proposal or bold edit" is undone, but it is often unclear which edit (the addition, change, or removal) falls under this classification.{{pb}}As soon as we reach this point in every discussion, opinions split. -- GoneIn60 (talk) 16:15, 19 April 2025 (UTC)

::::::::::I'm not sure what a "dispute of 'no consensus'" is. NOCON clearly applies only after a good faith discussion ends in no consensus. ONUS is not so clear and some editors use it during discussion to say "you have to convince me." - Butwhatdoiknow (talk) 16:53, 19 April 2025 (UTC)

:::::::::::Meaning a dispute or discussion that ends in no consensus. And yes, ONUS is misused beyond its original intent, which is why I think most of us agree something needs to happen. We just can't seem to agree on what that something looks like. -- GoneIn60 (talk) 17:29, 19 April 2025 (UTC)

::::::::::::{{tq|but it is often unclear which edit}}; yes, but maybe we can't do anything to solve that problem, because people will just have to figure out what the last consensus version was. Kolya Butternut (talk) 17:42, 19 April 2025 (UTC)

:::::::::::::I suggest to you that if the discussion ends in 'no consensus', editors will be unable to agree on which version of the article was the one with "the last consensus". And some of us, including me, believe that identifying such a version is unimportant, because what we know at the end of a no-consensus discussion is that there isn't any consensus, and that includes no consensus for reverting to that old version. After all, if there was a consensus to revert to that old version, then the discussion would have had a 'consensus' result instead of a 'no-consensus' result. WhatamIdoing (talk) 19:15, 19 April 2025 (UTC)

::::::::::::::And what's the solution? Kolya Butternut (talk) 23:13, 19 April 2025 (UTC)

:::::::::::::::If there were a good, fast, cheap solution, we'd have implemented it years ago.

:::::::::::::::But Good, fast and cheap basically never exists in practice – certainly not for complex or disputed subjects – so we have to ask: Which one of these desirable goals are we willing to give up? Shall we have bad content, so long as it's fast and cheap to make the decision? Shall we have lengthy processes, so long as the result is good and doesn't require too many editors? Shall we invest huge amounts of editors' time, so long as we get a good result from a large number of editors? WhatamIdoing (talk) 01:14, 20 April 2025 (UTC)

::::::::::::::::I was asking you what you want, and you put the question back onto me? Kolya Butternut (talk) 08:17, 20 April 2025 (UTC)

:::::::::::::::::I have already told you repeatedly what I want. I want the community to decide whether they want ONUS or not, and I want them to live with whatever decision they make.

:::::::::::::::::The options are:

:::::::::::::::::# If there's a dispute about including something, and we can't reach agreement to include it, then it's out.

:::::::::::::::::# If there's a dispute about including something, and we can't reach agreement to include it, then it's in.

:::::::::::::::::# If there's a dispute about including something, and we can't reach agreement to include it, then some other factor decides whether it's in or out (e.g., editor with the oldest account wins, older version of the article wins, we write a random-number generator to flip a coin...).

:::::::::::::::::I don't care what the choice is, but they need to pick something, and it needs to be the same rule for everyone. WhatamIdoing (talk) 20:23, 20 April 2025 (UTC)

::::::::::::::::::I do not think that is realistic. You need a test with more conditions and stipulations. Andre🚐 20:27, 20 April 2025 (UTC)

::::::::::A few thoughts:

::::::::::* There's already a list of such examples in Wikipedia:Verifiability#cite note-3, in BURDEN: "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.)". We could make that more visible.

::::::::::* I'm not sure that we need to "ensure a consistent outcome" if a discussion ends in 'no consensus'. My main concern is that some editors will see the factual statistical statement in NOCON and think that it favors inclusion of long-standing but disputed content, in opposition to ONUS favoring removal of disputed information, long-standing or otherwise. I don't want policies to contradict each other, or even to be capable, if a proficient wikilawyer is involved, to appear to contradict each other.

::::::::::* Because, you see, it is not true that "NOCON says "the proposal or bold edit" is undone". It is only true that "NOCON says that's what usually happens". NOCON doesn't say that no consensus should end in a status quo result; it only says that it's the statistically "common" outcome, assuming none of several common situations apply.

::::::::::* The problem of "which edit (the addition, change, or removal)" is the One True™ Status Quo Version is one of the reasons that I dislike the WP:QUO approach. People who feel strongly about Wikipedia:The Truth will never agree that any of the non-True versions is "the last consensus version".

::::::::::* If you haven't actually read the QUO section of the Wikipedia:Reverting essay during the last few years, it is probably worth your time to do so. The gap between what many editors have been told that WP:UPPERCASE says and what it actually says is dramatic.

::::::::::WhatamIdoing (talk) 19:13, 19 April 2025 (UTC)

:::::::::::I do not think a reasonable resolution would be to erase the defense of the status quo, as it is important. Otherwise, a brigade can come by and overturn anything by asking for a consensus of editors right then and there when one may have strongly existed at one time, but people are not obliged to satisfy you to turn out for a new head count quorum whenever some trolls might decide to call the roll. Focus on the arguments versus a tally would improve things if people were reasonable. Because then you can say, no, invoking ONUS is not sufficient here because there was a consensus that such and such an argument was valid and no new consensus has found otherwise. The idea that it takes a consensus to include something but that anytime anyone demands it to present itself that the old consensus is now null and void is non workable in my humble opinion. Andre🚐 02:44, 20 April 2025 (UTC)

::::::::::::In my experience, if you can actually point to written evidence of consensus in the past – a talk page discussion directly on the point, for example – then people will usually accept that (exceptions for very poor discussions and moderately determined POV pushers).

::::::::::::Usually, though, people invoke QUO when they know that no such discussion exists (e.g., because there has never been any discussion on the talk page at all, which appears to be the case for about 70% of the Talk: namespace, based on the number of pages that contain "UTC" [2.5 million] and the number that don't [5.7 million]).

::::::::::::Usually, these QUO invokers are hoping that you will take the fact that you were the first person to object to the content as proof that everyone else supports it. To which the only rational response is "See also Wikipedia:List of hoaxes on Wikipedia", because blatant vandalism didn't stay in the high-traffic Diabetic neuropathy for over four years and 192 intervening edits because there was "consensus" for it; it stayed there because nobody who knew anything about diabetes noticed it. WhatamIdoing (talk) 04:31, 20 April 2025 (UTC)

:::::::::::::In my view these different kinds of consensuses are indeed different. And if it should be the case that some past consensus that was explicit has some kind of protection by convention or de facto practice, that would be an example of where ONUS should not be a blank check to remove things, and a meaningful carveout or clarification. Moderately determined POV pushers are not a rarity after all. In the case that something was added, never discussed, and is removed without discussion, and reverted once, I can see ONUS being a benefit if it justifies removal or a 2nd revert and shifting the BURDEN on the reverter. But in the case where something had been discussed in the past, that discussion should be some protection against a drive-by assertion of the BURDEN. If someone is defending a well-attended, well-advertised discussion that led to a consensus, that is a situation where I would side with QUO/NOCON and say that the 2nd revert is misplaced. Andre🚐 06:34, 20 April 2025 (UTC)

::::::::::::::I think you've switched shortcuts. BURDEN applies only to uncited content.

::::::::::::::If something was established by {{xt|a well-attended, well-advertised discussion that led to a consensus}} (and no obvious reason to believe things have actually changed since then), or even something half that good, then you shouldn't be siding with QUO, whose real name is Wikipedia:Reverting#Avoid reverting during discussion and therefore inapplicable except "during discussion"; you should be siding with Wikipedia:Consensus, because you have every reason to believe that there is a consensus for whatever that was.

::::::::::::::The situation you describe ought to go like this:

::::::::::::::* A: I'm removing this cited, verifiable text because I think it has The Wrong™ POV.

::::::::::::::* B: I'm reverting it back in because we have a consensus to include this. See {{fake link|Talk:Article#Great Huge RFC}}, which was summarized with the words "Clear and convincing consensus" to include those exact words and those two sources.

::::::::::::::* A: Oh. [Insert optional rant about how they're right and everyone else is wrong.]

::::::::::::::It should not go like this:

::::::::::::::* A: I'm removing this cited, verifiable text because I think it has The Wrong™ POV.

::::::::::::::* B: I'm reverting it back in because of the essay QUO, which only applies during a discussion [which we're not having], and because reverting it back in is the statistically most likely outcome in the rare event that the discussion [which we're not having] doesn't come to a consensus either way.

::::::::::::::WhatamIdoing (talk) 06:53, 20 April 2025 (UTC)

:::::::::::::::Sorry for confusing matters by substituting BURDEN for the actual concept of who has the burden of proof or whose responsibility it is. My point is that in the 2nd case you gave, reverting it back is not because of QUO but because of the past consensus. Then the question is what happens when

:::::::::::::::*C: Revert of revert, removes the item citing ONUS

:::::::::::::::Andre🚐 08:10, 20 April 2025 (UTC)

::::::::::::::::Then you point "C" to the existing evidence of consensus. If "C" accepts that, then you're done. And if "C" doesn't, you decide how you want to pursue this. For example, another discussion? Ping the person who wrote the closing summary? Chat up admins about "editing against consensus"? WhatamIdoing (talk) 20:17, 20 April 2025 (UTC)

:::::::::::::::::But ONUS currently as you said has no concept of existing consensus. As written, any inclusion is now the responsibility of whoever is restoring it today. Andre🚐 20:22, 20 April 2025 (UTC)

::::::::::::::::::ONUS requires "whoever is restoring it today" to take responsibility for achieving a consensus to include it. Pointing at last month's "Great Huge RFC" is "achieving a consensus". ONUS doesn't say that you can't get written permission in advance. It only says you have to get written permission. WhatamIdoing (talk) 20:26, 20 April 2025 (UTC)

:::::::::::::::::::That certainly is not clear from how it is written. Andre🚐 20:27, 20 April 2025 (UTC)

::::::::::::::::::::Clear enough? If you impose a "must be done after the dispute", then you end up with the nonsensical result that ONUS allegedly supports:

::::::::::::::::::::# Hold a Great Huge RFC, which closes on Monday with clear evidence of a strong consensus to include ____.

::::::::::::::::::::# Add ____ on Tuesday.

::::::::::::::::::::# Someone reverts on Wednesday, and says "I cry ONUS. You have to do another RFC, because ONUS magically resets the clock on consensus, and any discussion or evidence of consensus that predates my revert doesn't count."

::::::::::::::::::::# And if you do, another opponent can force a third RFC, etc.

::::::::::::::::::::The community won't stand for that. WhatamIdoing (talk) 20:32, 20 April 2025 (UTC)

:::::::::::::::::::::That is not too far from situations that have happened before. You aren't supposed to relitigate things if nothing substantive has changed, but I know of situations where they just keep restarting the discussion until people are tired and the result has changed due to the draw of who was there at the end. No consensus = status quo is key to how Wikipedia works. Many situations where the arguments are equally strong and the attendance is roughly equal should lead to a status quo or a compromise. But sometimes this is not what happens. ONUS should be updated to reflect that consensus does not vanish simply because those people are not there today. Andre🚐 20:35, 20 April 2025 (UTC)

:::::::::::{{tq|ONUS favoring removal of disputed information, long-standing or otherwise}}; you keep saying this, but there is no consensus that ONUS means that. Kolya Butternut (talk) 08:21, 20 April 2025 (UTC)

::::::::::::Maybe not a consensus but in practice, it can work like that, it is a bit like a finger on the scale. Still think maybe first try to find an agreed policy page to park these things and go from there, rather than trying to cross every t and dot every i beforehand. If that's possible. Selfstudier (talk) 09:36, 20 April 2025 (UTC)

:::::::::::::Actually, when we ask editors to tell us what ONUS says, they agree that ONUS doesn't provide an exemption for "long-standing" content.

:::::::::::::I have at least once had to ask them to pretend it was a Reading comprehension test to get them to read the words that were in front of them, but I've never yet had anyone say that {{xt|The responsibility for achieving consensus for inclusion is on those seeking to include disputed content}} contains any words about "long-standing" content. WhatamIdoing (talk) 20:19, 20 April 2025 (UTC)

::::::::::::::That is a conflict between advice and practice. As we know, practice trumps the written policy and it should be updated to reflect that. However, as you say, an RFC might likely reach no consensus to do so. Andre🚐 20:26, 20 April 2025 (UTC)

:::::::::::::::I find that editors are perfectly capable of understanding ONUS exactly as it is written—as long as they're trying to get something excluded. When they want to include information, they appear to develop temporary amnesia about ONUS' existence. WhatamIdoing (talk) 20:29, 20 April 2025 (UTC)

{{outdent}}

If the word "include" means "add" then "longstanding" is implied if longstanding indicates a certain level of consensus. Kolya Butternut (talk) 23:34, 20 April 2025 (UTC)

:Include clearly does not mean add. Add means add. Include means include. Sorry, not to be tautological, but the act of adding something that was not there before versus the act of removing something is what is at question here specifically. Andre🚐 23:37, 20 April 2025 (UTC)

::Kolya, the reason that "add" is a problem is basically that we create new articles.

::I wrote Cake mixing techniques yesterday. So far, there are three edits, all by me. All three versions contain the full list of techniques.

::If you disputed the inclusion of one of the methods listed in it, then what "long-standing" or "status quo" version could you revert to, that wouldn't include the disputed method?

::An insistence on "add", rather than "include", implies that the article has past versions that are potentially acceptable. This is not true for many articles. It also creates a First-mover advantage for the article creator: I will never have to defend any of the content I included or take responsibility for achieving a consensus for Wikipedia to include this information, because mine is always the most long-standing version. After all, nobody's preferred version could possibly be more long-standing than the original creation. WhatamIdoing (talk) 01:19, 21 April 2025 (UTC)

:::That is an edge case. The case that is problematic involves include where inclusion is an ongoing negotiation between different editors and very often in controversial articles, sensitive and delicate agreements are brokered between opposing camps over a lengthy period of time. Andre🚐 01:27, 21 April 2025 (UTC)

::::I don't think that's an edge case. Single-author articles are common. Have a look at a few of mine:

::::* Holland–Frei Cancer Medicine: Me and two bots.

::::* Mattering: Me, two bots, and one script-wielding AWB editor.

::::* National Council on Severe Autism: Me and a couple of editors, one of whom has such strong strong feelings about the organization's existence in the real world that he [https://en.wikipedia.org/w/index.php?title=National_Council_on_Severe_Autism&diff=prev&oldid=1283246772 removed] a link to severe autism and replaced it with a scare quotes version – and yet it's still the case that more than 90% of the words on the page were written by me.

::::* Reversible poem: Me, a bot, five other editors, and every single word was written by me.

::::* Second-chance hiring: Me, a bot, five other editors, and yet almost 95% of the words were written by me.

::::* Disperse blue dye: Me, a bot, and two script-wielding editors.

::::* Shall We All Commit Suicide?: Me, a bot, four other editors, and every word was written by me except the caption for the image.

::::* Grinnell 14: Me, two bots, three humans, and every word on the page except "Iowa US" was written by me.

::::* List of fires in Egypt: Me, a bot, and four script-wielding editors. Every single word in the article was written by me.

::::I could go on, but it's kind of discouraging. People just don't collaborate much. So if you show up at one of these with a desire to remove something, there realistically isn't a "long-standing" version that doesn't contain whatever I wanted to include. WhatamIdoing (talk) 02:10, 21 April 2025 (UTC)

:::::Yes, I understand and I also have written some articles alone. But concerns about revert wars, POV pushers, and the misuse of ONUS are generally going to be about big vital articles with many editors. So it is an edge case pertaining to ONUS use because nobody is likely revert warring or citing ONUS on articles with one author. Andre🚐 02:13, 21 April 2025 (UTC)

::::::It's not "some" articles. That's a consecutive list of every one I looked at. WhatamIdoing (talk) 06:06, 21 April 2025 (UTC)

:::::::I think that is a function of what you choose to be interested in to create or edit. But consider an article such as a major political figure, language, ethnic group, municipality, and so on. Andre🚐 06:30, 21 April 2025 (UTC)

::::::::Yes, consider high-traffic, high-conflict articles.

::::::::But also consider that >95% of Wikipedia's articles aren't those. Reversible poem is in the top ~6% of page views (7,852 page views during the last year [https://pageviews.wmcloud.org/?project=en.wikipedia.org&platform=all-access&agent=user&redirects=0&start=2024-04&end=2025-03&pages=Reversible_poem]), and yet there isn't a single word on that page that wasn't written by me.

::::::::We need the rules to work for all the articles, not just the high-traffic, high-conflict ones. WhatamIdoing (talk) 18:18, 21 April 2025 (UTC)

:::::::::In my view ONUS is not about all articles. ONUS is about disputed content. Many articles never have a dispute. Within the subset of article text that is disputed, ONUS has to do with how to handle a dispute of that material and the responsibility or proper role of the editors involved in the dispute. Consensus is part of it and verifiability has to do with it. But, in the end, it is not about all articles. Most of the articles that you wrote solo about lesser-known facts will never need an ONUS because nobody will ever challenge or question the text. And if they do, it will be pretty cut and dry. No need for an RFC or a 3rd opinion. Andre🚐 01:52, 22 April 2025 (UTC)

:::::I thought you didn't have a preference for what ONUS should mean? Kolya Butternut (talk) 02:20, 21 April 2025 (UTC)

::::::I prefer to have rules written clearly. If we mean "add", then it should say "add". It doesn't. It says "include", and we have (a) no reason to believe that "add" was intended and (b) good reasons to believe that "add" and "include" are not synonymous in very practical ways. WhatamIdoing (talk) 06:07, 21 April 2025 (UTC)

:::::::But we don't have good reason to believe "include" was intended over "add" when in context it is a policy intended to be about V and not a policy about editing and consensus. Kolya Butternut (talk) 12:08, 21 April 2025 (UTC)

::::::::* The first good reason to believe that "include" was intended over "add" (in the sense of "make a change") is that the editor chose the word "include".

::::::::* The second good reason to believe that "include" was intended over "add" is that it's in the section saying that getting your content to stick on the page requires more than verifiability.

::::::::* The third good reason to believe that "include" was intended over "add" is that the editor who added it [https://en.wikipedia.org/w/index.php?title=Wikipedia:Verifiability&diff=prev&oldid=621111849 said] that he was trying to stop "POV pushers", not "changes".

::::::::* The fourth reason is that editor has never contradicted the interpretation of ONUS as being intended to facilitate the removal of bad content regardless of when it was added.

::::::::* The fifth reason is that editor has also spent a lot of time removing information that other editors wanted to include. 85% of his recent mainspace contributions reduced the size of the page. Very few of them were undo/reverts.

::::::::So, yes, I think we do have good reasons to think that "include, regardless of when it was added" is the intended meeting. WhatamIdoing (talk) 18:03, 21 April 2025 (UTC)

:::::::::That's not convincing. In ten years he's never actually articulated his intentions? Kolya Butternut (talk) 18:14, 21 April 2025 (UTC)

::::::::::I'm not sure that it will help, but we can do this:

::::::::::@JzG, please tell us whether you think that WP:ONUS should apply to only to "additions" to an article, or if you think it should apply to any content in the mainspace that has ever been "included", or that someone wants to have included.

::::::::::Please be clear, specific, detailed, and direct, so that your words cannot be misunderstood or twisted.

::::::::::We may find it helpful if you make specific statements about:

::::::::::* whether you believe that "long-standing content" should be exempt from an ONUS-based removal, and

::::::::::* whether you believe that content present from the first version of an article should be exempt from ONUS because it hasn't been merely "added" but is the most long-standing of "inclusions".

::::::::::WhatamIdoing (talk) 18:24, 21 April 2025 (UTC)

:::::::::::That's a loaded question. And "longstanding" is not a PAG based term is it? There are only varying degrees of consensus. Kolya Butternut (talk) 19:56, 21 April 2025 (UTC)

::::::::::::As far as I know, there are no actual policies or guidelines that favor retention of "long-standing" content merely because it is long-standing. It's also not in QUO. But QUO-promoting editors regularly use descriptions like "the long-standing version" when explaining why Wikipedia:Reverting#Avoid reverting during discussion supposedly means that they are supposed to revert to their preferred version during discussion.

::::::::::::"Long-standing" is theoretically not a measure of consensus. It is a measure of purely factual age. We can have "long-standing hoaxes" and "long-standing vandalism" and "long-standing copyright violations". WhatamIdoing (talk) 20:01, 21 April 2025 (UTC)

:::::::::::::That was a rhetorical question to point out that your question was loaded. Kolya Butternut (talk) 20:27, 21 April 2025 (UTC)

::::::::::::::I have confidence that JzG can and will see through how I've written the question, and that he tell us directly if he disagrees with me. WhatamIdoing (talk) 20:45, 21 April 2025 (UTC)

:::::::::::In my view, the onus is always-on the editor seeking or defending inclusion. The number of editors adding stuff, sometimes in overt defiance of policy (especially NOTDIR), vastly exceeds the number of gnomes and others cleaning it up. Policy is policy. You have to demonstrate that content you like, meets policy. Guy (help! - typo?) 22:32, 21 April 2025 (UTC)

::::::::::::Does "include" mean "add", or does it refer to, for example, restoring content with potential silent consensus? Kolya Butternut (talk) 23:38, 21 April 2025 (UTC)

Consensus. Let's put them all at Consensus. Kolya Butternut (talk) 14:18, 20 April 2025 (UTC)

:Np with that. Eventually, that's what all the onus/quo/blah discussions are about, even if they are "stuck" for some period of time (bit like this one). Selfstudier (talk) 13:45, 21 April 2025 (UTC)

• Could we satisfy both sides by transcluding NOCON to this page, without moving it?—S Marshall T/C 13:51, 21 April 2025 (UTC)
• :Can you draft that up? Kolya Butternut (talk) 18:17, 21 April 2025 (UTC)
• ::Why would we want to violate Wikipedia:Policies and guidelines#­Content, which opposes duplication and redundancy?
• ::More to the point, the reason for this proposal was to get NOCON out of CON. Transcluding it here would not meet the goal of removing it from there. WhatamIdoing (talk) 18:26, 21 April 2025 (UTC)
• :::Not sure it's a goal at this point, a subject of discussion maybe, at any rate I am less concerned with where individual things are, more whether they are together, wherever they are. Selfstudier (talk) 18:37, 21 April 2025 (UTC)
• :::The policy does not oppose duplication and redundancy, it says {{tq|minimize redundancy}}. But I agree with Selfstudier; let's get them together on one page. Consensus seems like the easiest place. Kolya Butternut (talk) 14:28, 22 April 2025 (UTC)
• ::::"Minimize" is opposition. It is not absolute opposition ("ban"), but it is the opposite of support and therefore opposition. WhatamIdoing (talk) 17:20, 22 April 2025 (UTC)
• :I like the doubling up idea. You could have two sentences slightly different. One about NOCON as pertaining to consensus, with a link to editing policy, one about NOCON as pertaining to edits, with a link to consensus. Other policies and guidelines follow this reciprocal lens link idea. It is still minimizing redundancy because we made it less redundant by having each one reflect its own position. Andre🚐 17:24, 22 April 2025 (UTC)
• ::Can you propose something specific? Kolya Butternut (talk) 22:40, 22 April 2025 (UTC)

WP:PRESERVE is, in my experience, one of the most frequently poorly-cited parts of policy; I generally see it cited by people who make WP:BOLD edits and are then upset to see them reverted via WP:BRD. The basic idea is not bad - people whose objection to a new addition is fixable should consider fixing it themselves rather than reverting. The problem is that it is far too stridently-worded, in a way that implicitly places the burden of fixing deeply-flawed material on people who object to it rather than people who want to retain it, and which encourages people in a dispute over a BOLD new addition to derail the discussion into futilely arguing that their additions should remain in place automatically rather than trying to fix it themselves or provide context-specific reasons why their additions are valuable. Suggested possibilities:

• Add "consider" near the start, making it clear that WP:PRESERVE is just one consideration and not an automatic entitlement. {{tq|Rather than removing imperfect content outright, consider fixing problems if you can...}}
• Remove the "might" from DONTPRESERVE, which implicitly indicates that it is weaker than PRESERVE. The opposite is true! There are no situations where PRESERVE is strictly, unequivocally required; but there are, in fact, situations (mostly WP:BLP) where DONTPRESERVE is non-negotiable and where it is a "must" rather than a "might." I would simply tweak it to: {{tq|Several of our core policies discuss when to remove information from an article rather than preserve it.}}
• PRESERVE also needs to reference WP:BURDEN (and perhaps WP:ONUS and WP:BRD). When the problem is sourcing, it is inappropriate to cite PRESERVE and demand that other people produce sourcing; the ultimate burden to produce sourcing for new additions is the people who want to add or retain it, not the people who want to retain it. And per ONUS, inclusion is ultimately determined by consensus. I would add a paragraph to the end of PRESERVE along the lines of {{tq|Preserving material is only one consideration out of many; if something you added was reverted, especially if it was a bold addition, do not immediately cite preserve to argue that it should be kept. Instead, try to gather more information about why it was removed if it isn't clear, and see if you can address those problems yourself. Remember that the ultimate burden to provide sources is on people who wish to add or retain material, not on people who believe it should be removed, and that the inclusion of material ultimately depends on consensus.}}

As it is worded now, PRESERVE gives people the impression that as long as they believe material is fixable, they can point to it and demand that other people do the work of fixing it instead of removing it. This isn't, I think, what it's about - it's a gentle encouragement to try and fix rather than removing. Like all such encouragements, if someone is egregiously ignoring it and removing stuff that is obviously both essential to the article and trivial to fix, that might be a problem, but the vast majority of disputes over content are about things that could easily be included or excluded from the final article, where it ultimately comes down to editorial calls based on consensus. And PRESERVE sits at the intersection of a huge number of other policies for dealing with content disputes, most of which (WP:BRD, WP:ONUS, WP:BURDEN, WP:QUO when dealing with new additions) truthfully point in the other direction. The current strident wording of PRESERVE is a poor fit for those disputes. Having a policy that points out that it's worth trying to preserve something is good; having something that sounds like people can point to it to argue their bold inclusion must be kept and that other people are somehow in violation if they remove it is not good. --Aquillion (talk) 18:33, 21 April 2025 (UTC)

:I don't think this is necessarily helpful. PRESERVE is widely flouted, especially by self-selected "reviewers" and "patrollers". Softening would make this desirable behavior seem even more optional.

:For example, consider your suggestion "Rather than removing imperfect content outright, consider fixing problems if you can...":

:There is already a tendency among some RecentChanges patrollers to treat Wikipedia like a game of Mother, May I?: If your contribution is imperfect, I'll just remove it. I feel no obligation to fix problems, even when I can do so easily. If you didn't do it perfectly on the first try, then it's gone, and you're welcome to try again, if you think you can do better and don't assume that there was something more fundamentally wrong with your effort.

:If we want to build the encyclopedia, then we really should be fixing problems when we can. This has two positive effects:

:* When you collaborate with newcomers, they feel like their contribution was wanted, even if the extent of your collaboration was merely to slap a {{tl|fact}} tag on it.

:* When you fix the fixable problems (e.g., moving a paragraph to a more relevant article), then readers get what they want most, which is decent content.

:WhatamIdoing (talk) 19:09, 21 April 2025 (UTC)

::I mean, my counterpoint is that you're conceding that WP:PRESERVE doesn't reflect current practice. Sure, I do think that it makes good gentle advice - I'm suggesting we reword it to make it clear that it is merely general advice; I don't want to delete it entirely - but it's not a hard requirement and never will be. The wording, here, that implies that is a hard requirement is wrong, and because all experienced editors know that it is wrong it will only frustrate people who try to cite it as if it is. If you make a BOLD edit and your response to it being reverted is to cite PRESERVE rather than to try and talk things out and fix it yourself, you're going to have a bad time, and you're going to waste the time of everyone involved, because there simply is not a requirement that other people fix seriously-flawed text themselves instead of removing it. If you then tried to go to WP:ANI or WP:AN or WP:AE with the complaint that "yes, my text was deeply flawed and poor-quality and riddled with spelling errors and was stridently POV and totally uncited, but all of that could have been fixed, and they deleted it instead, and PRESERVE said they're not allowed to do that outside of these narrow cases" you'd be laughed out of the room and be lucky to avoid a WP:BOOMERANG if you refused to back down. Don't get me wrong, repairing text instead of deleting it is often a nice thing to do, but there's a balance-point to it; not everything that could be in the final article is required to be in the final article, and having nothing is usually safer than having something that is seriously-flawed. Preserving all potentially salvageable text in all cases, as PRESERVE currently instructs people to do, is not and has never been required, nor is it desirable, reasonable, or even helpful advice.

::Whether to fix vs. remove vs. tag problematic text is a nuanced call that depends on numerous factors, including how essential the basic concept is; how serious the flaws are; the risk of harm posed by the text's flaws; editorial leeway over things whose inclusion is plausible but not essential; etc. It is absolutely not an outside of these narrow cases, you must do this no matter what thing. Even with my proposed tweaks, compared to other policies, {{tq|consider}} is not weak wording (eg. BLP even uses {{tq|strongly consider}} when it comes to implying criminality for people not convicted of crimes.) I don't think that PRESERVE meets the standard for more aggressive wording than that. You could still, if someone egregiously and repeatedly deletes easily-salvageable text rather than repairing it, drag them to WP:AN with the argument that they're clearly not even {{tq|considering}} salvaging it. But I don't think that you can really impose more of a requirement on people than that - the reviewers and patrollers you're upset at are broadly accepted by the community; their removals are valid content disputes, not violations of policy. Regardless of the wording of PRESERVE I doubt you'd have any success bringing them to AN or anywhere else unless the problem with their removals was so severe that they clearly weren't even considering repairs, because there's just too many other policies and practices and realities pushing us towards the removal of problematic text as a serious option that we need to frequently consider as well. It is not the only option, and WP:PRESERVE serves a useful purpose in reminding us of that; but it doesn't, in practice, go any further than that, and its wording needs to reflect that reality. --Aquillion (talk) 01:22, 22 April 2025 (UTC)

:::Sure, people wrongly invoke PRESERVE. Every now and again, Ned Newbie will dump garbage in a page and claim that PRESERVE says I'm not a WP:VOLUNTEER and I'm supposed to clean up his mess. But I can easily reply that WP:DONTPRESERVE is also policy and is much more relevant when confronted with garbage. (And I've certainly seen some serious garbage over the years. Aside from the ordinary vandalism and POV pushing, we had one editor who claimed that his Repetitive strain injury prevented him from using the space bar on his keyboard, so he claimed we were required by US law(!) to accept his space-free typing and then clean it up afterwards so that his articlesandtalkpagecommentswouldn'tallbewrittenlikethis. I never could decide whether it was some sort of intentional trolling/performance art, or if he actually believed his claims.)

:::The fact is that I don't want people, when they see the kind of problem [https://en.wikipedia.org/w/index.php?title=Walford_Davies&diff=prev&oldid=34175871 in this edit], to only "consider" fixing the problem and then decide that it would be much easier to just revert it or blank everything. I want them to actually fix the problem, because this is a collaborative project, and fixing other people's mistakes is not just "nice to do". It's why we're here, instead of having our own blogs. RecentChanges patrollers who don't want to fix a specific problem should just leave it alone, and let someone else do it. They should not blank everything over minor problems.

:::Has anybody fixed a mistake you've made? Tagged something to alert you to a problem you hadn't noticed? Do you want more or that, or would you rather that every time you had a typo, someone reverted all your work? PRESERVE has a lot of the Golden Rule behind it. WhatamIdoing (talk) 01:37, 22 April 2025 (UTC)

::::Again, it feels like you're agreeing with me but would prefer to keep a version of the text that you disagree with because you feel that falsely stating that PRESERVE is required makes it more likely that people will do it. WP:DONTPRESERVE doesn't currently contain the exception for quality you imply; we would have to tweak it to add that. And WP:PRESERVE, as currently worded, is required, while DONTPRESERVE is just a "might". Obviously you recognize that this doesn't reflect actual practice (you just admitted to sometimes not following PRESERVE yourself.) Yes, of course I think that it's nice when people do fix things I added instead of removing them - but it would be absurd to suggest that we could require that they do so; and I have had far more problems from well-intentioned editors misled by the current inaccurate wording than I have from any other policy; when someone invokes it is almost always a sign that a content dispute is going to go drastically awry, wasting huge amounts of time and energy by falsely giving them the impression that they can demand that other people put in the work to repair things they've added. You concede that people sometimes misinterpret PRESERVE in that way; but that misinterpretation is in the text, to the point where even you, in order to defend the current inaccurate wording, had to read something into it that is not there. Clearly there is no requirement to repair poorly-cited or poorly-written text instead of deleting it - there are clearly points where it must be deleted. But PRESERVE, as written, does not allow for that, and DONTPRESERVE's extremely narrow exceptions where it grudgingly says that we might allow deletion do not include the ones you read into it. You're an experienced editor, so when you read PRESERVE your brain is inserting a bunch of "but actually" and "really this is just something you should consider doing" that isn't actually there. Forget what you know about actual practice and read it, not as an essay trying to encourage people to consider preserving when possible by using grandiose language (which seems to be how you're interpreting it) but as actual you-will-be-taken-to-AN-if-you-violate-this policy whose precise details are part of Wikipedia's rules. Its wording is wildly inaccurate and completely out of line with current policy. --Aquillion (talk) 11:33, 23 April 2025 (UTC)

:::*Also, I should add - would you at least agree, at the bare minimum, to the addition of the paragraph I suggested about how preserving material is only one consideration among many (and that people should not cite PRESERVE in actual content-dispute discussions), and the addition of bullet-points to DONTPRESERVE about things that are too low-quality to serve as the basis for a better version? You implicitly indicated that you believe that the current version is all right because you thought that such an exception is present, but it currently isn't, which means we ought to add it, surely? I think that the crux of the other part of the paragraph I want to add is that leaping straight to WP:PRESERVE is assuming bad faith - editors should assume that a removal is based on a valid dispute; PRESERVE should only come up in discussions when someone unambiguously indicates that they removed something solely for reasonably fixable (ie. comparatively minor) spelling and grammar issues. --Aquillion (talk) 11:35, 23 April 2025 (UTC)

:Agree strongly. Why do we need to "preserve" poorly-cited content? It's not like we're struggling to fill the servers or anything. I've had pushback on this basis when removing content sourced to the most ludicrous of sources. It is absolutely not my job to fix the lack of sources supporting Randy in Boise and his assertions of sword-wielding skeletons. Guy (help! - typo?) 22:33, 21 April 2025 (UTC)

::The middle of WP:PRESERVE has a bullet list of 11 items. It might be "not my job" to "improve grammar" (=first bullet point in PRESERVE), but it's also not okay to remove a large amount of well-sourced material just because it's got a fixable grammar problem. Doing so would violate Wikipedia:Gaming the system#REMOVELARGE as well as PRESERVE.

::PRESERVE is mostly not about uncited and poorly cited content. It's mostly about verifiable content that isn't perfect. [https://en.wikipedia.org/w/index.php?title=Walford_Davies&diff=prev&oldid=34175871 This edit] from your early days is what PRESERVE is trying to encourage: Fix the minor problem, instead of blanking the whole thing because it wasn't perfect. WhatamIdoing (talk) 01:25, 22 April 2025 (UTC)

:::I tend to agree with WAID that PRESERVE is poorly-understood and poorly-followed. The point of it, in my view, is about incrementality as WAID says. Wikipedia works because in theory at least, it engenders aggregation. In other words, we benefit from other people's flawed contributions, SOFIXIT and try to fix problems instead of deleting stuff because it is not up to par. Andre🚐 01:49, 22 April 2025 (UTC)

::::Sure, but WP:SOFIXIT is a loose guideline. My problem with PRESERVE is that it is currently written as if deleting low-quality contributions that could have been fixed is a violation of policy, which is clearly not the case. It is "poorly-understood" because the wording is actively misleading and does not reflect actual policy or practice. --Aquillion (talk) 11:47, 23 April 2025 (UTC)

::*PRESERVE mentions sourcing three times across its eleven bullet-points. If you agree that it is {{tq|mostly not about uncited and poorly cited content}}, would you agree to removing those mentions of sourcing so that people can no longer cite it to demand other people find sources for their own additions? Such a demand already violates WP:BURDEN in any case. And perhaps even adding "verifiable" to it, eg. {{tq|Instead of removing verifiable content...}} --Aquillion (talk) 11:42, 23 April 2025 (UTC)

• It ought to say "try to preserve appropriate content", and it used to. Who changed that, and why?—S Marshall T/C 22:44, 21 April 2025 (UTC)
• :That happened in [https://en.wikipedia.org/w/index.php?title=Wikipedia:Editing_policy&diff=prev&oldid=1073717138 this edit] by Butwhatdoiknow three years ago. I see no obviously relevant discussion in the archives. WhatamIdoing (talk) 01:49, 22 April 2025 (UTC)
• ::Thank you.—S Marshall T/C 06:03, 22 April 2025 (UTC)
• ::For what it is worth (not much), it looks like my intent was to reorganize without changing the meaning. - Butwhatdoiknow (talk) 16:00, 22 April 2025 (UTC)

• Restore the longstanding wording "Preserve appropriate content" as the second sentence of WP: PRESERVE. I believe this will succinctly clarify its intent.—S Marshall T/C 06:07, 22 April 2025 (UTC)
• :Support Andre🚐 17:04, 22 April 2025 (UTC)
• :By "restore," do you mean add the text as a "new" three word sentence, replace the current second sentence with the text, replace "Rather than remove imperfect content outright" in the second sentence with the text, or something else? - Butwhatdoiknow (talk) 16:42, 23 April 2025 (UTC)