This is an archive of past discussions about Wikipedia:Verifiability. 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.
"Get consensus" objections to text citing to Consensus policy
As a preliminary matter, I note that "get consensus" is not a valid objection to a bold edit on a policy page per wp:PGBOLD ("you should not remove any change solely on the grounds that there was no formal discussion indicating consensus for the change before it was made").
These two edits [1][2] reverted the addition of the following text at wp:ONUS:
CompareWP:NOCON (lack of consensus commonly results in retaining the version of the article as it was prior to the proposal or bold edit).
I get the part about add rather than include, I agree they are not the same. For my edification, what is the purpose of the reverted addition? Selfstudier (talk) 17:16, 9 May 2022 (UTC)
I'm not asking about add/include here (but will later). With regard to the WP:NOCON sentence, the purpose is to alert editors to another policy dealing the same issue. - Butwhatdoiknow (talk) 17:57, 9 May 2022 (UTC)
It waters down the policy here by excessively privileging pre-existing text. "Commonly results" is not the same thing as saying who has the burden of getting consensus to retain text. Material which is dubious is unlikely to gain consensus to include but may end up without consensus either way; such material should be left out. Changing the policy implies there must a consensus to exclude before it can go. Crossroads-talk-03:30, 10 May 2022 (UTC)
Do you genuinely believe that the current default for longstanding text, when discussions deadlock and fail to reach a consensus, is to remove it? I feel like we've been in several disputes where you supported longstanding text and restored it when there was clearly no affirmative consensus to retain it. If you're going to argue that ONUS requires affirmative consensus, is it all right for anyone to go around restoring disputed text without a clear consensus to support it? Or, I guess - my real point is, how do you see this working in practice? Let's say I remove some longstanding text in an article, and you feel that I'm whitewashing it by doing so. Is it all right for you to revert my removal without first establishing a consensus? If you do and I then start a discussion on talk that deadlocks and fails to produce a clear consensus for the disputed text, do you actually feel that we should go back and remove it again per your interpretation of ONUS? --Aquillion (talk) 18:46, 10 May 2022 (UTC)
Do you genuinely believe that inappropriate text is required to remain at Wikipedia indefinitely just because someone failed to notice it for a long enough period of time? Are people allowed to stonewall discussions in bad faith just to retain their pet text which is otherwise inappropriate for Wikipedia? Why is there a time limit when we can't remove inappropriate text from Wikipedia? --Jayron3218:55, 10 May 2022 (UTC)
Because text that has stood for a long time on an article enjoys implicit consensus due to the number of people who have read it without removing it. Your proposed version (ie. the version with parts of the article removed) enjoys no consensus, and it is completely inappropriate to insist that it remain in place during extend discussions - nor has policy or practice ever supported that. --Aquillion (talk) 19:17, 10 May 2022 (UTC)
Text that has stood for a long time on an article shows implicit ignorance that it has been a problem, as any number of people who may have (rightly) objected to the text when it was added didn't know it was there and if the text didn't belong, but should have been removed, it should not default that the person who added it gets "first mover" advantage in keeping text in an article so long as they pass some arbitrary time period with no one noticing it. --Jayron3219:40, 10 May 2022 (UTC)
I do think that (like I said below) we're not quite as in as much disagreement as might seem, since a lot of what you're saying focuses on "nobody noticing it" (ie. stuff that hasn't had many eyes on it, which would lack implicit consensus), whereas a lot of what I'm saying leans on implicit consensus (ie. the presumption that people have seen it and that the person who wants to remove it is just the first person to object to it.) I think any solution would do best to maybe hash out how WP:ONUS interacts with implicit consensus and what its contours are - though it's probably not something that can be nailed down with absolute precision, I think that "clearly nobody has seen this problem until now" is a valid argument for saying that something lacks implicit consensus and should therefore default to removal, whereas "clearly lots of people have read this text without objecting to it" is a valid argument that it enjoys implicit consensus and therefore requires affirmative consensus to remove. Part of the reason we disagree on what current practice is might also just be because we tend to edit different sorts of articles (ie. more established ones, controversial ones that have had a lot of attention, or even WP:GAs, where most of the text can be presumed to have implicit consensus; vs. newly-created ones or old but obscure ones that have very few edits and very few eyes on them, where it can't.) I do think that sometimes number-of-edits is a more useful measure of implicit consensus than just time-passed. --Aquillion (talk) 20:18, 10 May 2022 (UTC)
I have reverted the addition of the corresponding text at WP:NOCON, largely per my arguments above; including it at only one policy and not the other gives the impression that ONUS trumps NOCON, which is not the case, and risks undermining NOCON in general. Beyond that I consider NOCON to be the better-worded and more well-supported policy (and the one that largely reflects actual practice), whereas the sloppy wording of ONUS with regards to what qualifies as consensus and what disputes it is meant to cover are what lead to the sorts of disputes we see here. I would strenuously oppose having NOCON point to ONUS in its current form. --Aquillion (talk) 21:36, 11 May 2022 (UTC)
NOCON says (in hidden text) "This section summarizes existing policies and guidelines. It does not make any new rules. If this page and the more specific policy or guideline disagree, then this one should be changed to conform with the more specific page." It seems to me that in removing the sentence indicating that NOCON's summary might be incomplete ("CompareWP:ONUS (onus to achieve consensus is on those seeking to include disputed content)."), you kind of did the opposite of what NOCON says to do. WhatamIdoing (talk) 00:10, 12 May 2022 (UTC)
Hey, if I don't remember discussing this in the past then it didn't happen (;-)). I look forward to you finding the prior discussion so we don't have to reinvent the wheel here. - Butwhatdoiknow (talk) 15:18, 10 May 2022 (UTC)
There's that link there I just gave, I will see if I can find the other discussion, there might have been, probably was, more than one. It comes up fairly often, I think. Ultimately, I think what people want to avoid is situations being wikilawyered so a fair amount of flexibility is built in to the policies at the same time as indicating what might constitute a problem.Selfstudier (talk) 15:24, 10 May 2022 (UTC)
A flexible policy is good until, that is, it directly conflicts with another policy. Do these two policies conflict? That's my question. - Butwhatdoiknow (talk) 15:49, 10 May 2022 (UTC)
They are 100% not in conflict. WP:ONUS says that, when there is contentious material, leave it out of the article while discussion happens. WP:NOCON says that at the conclusion of the discussion, a "no consensus" result causes a return to status quo ante bellum. Which is to say that until the discussion concludes, we leave all contested material out. After discussion has concluded, then if there is no consensus, we return the article to whatever state it was in before the discussed changes were made. Notice two things 1) your initial edit changed "include" to "add". This is a BIG change, as it draws a distinction between newly added content, and content that had already been in the article, which WP:ONUS never had made such a distinction before you changed the text. 2) the text I put in italics notes key differences between ONUS and NOCON. Specifically, which you seem to miss, is that "during" is not a synonym for "after". --Jayron3211:09, 10 May 2022 (UTC)
Neither the word "during" nor the phrase "while discussion happens" appear in ONUS. Should we add text to make it clearer that ONUS only applies during discussion? - Butwhatdoiknow (talk) 15:31, 10 May 2022 (UTC)
That interpretation doesn't reflect current practice (excluding BLP-sensitive statements, which are a special case) - by that logic, I could remove any longstanding text from any existing article, argue about it on talk, refuse to accept any consensus, and start an RFC, at which point I could point to the RFC being in progress as an argument to continue excluding until it concludes (unless the RFC WP:SNOWed out or consensus was so glaringly against me as to be obvious.) The purpose of WP:NOCON is to ensure article stability; allowing people to remove longstanding text and then keep it out for as long as they can filibuster makes no sense and has never reflected any sort of practice or policy. --Aquillion (talk) 18:31, 10 May 2022 (UTC)
The phrase "The onus to achieve consensus for inclusion is on those seeking to include disputed content." Not add, and not remove, but "include", has been in the text of this page since at least 2017 (see this version), and in some form with slightly different wording since at least 2006 (see here [3], which states "Just because some information is verifiable, it doesn't mean that Wikipedia is the right place to publish it." which was later split out into WP:NOT, but was later returned to WP:V, as shown above. Also, we don't assume bad faith among fellow editors. If someone is abusing process, sanction the person. Good faith disagreements over content have always defaulted to "leave content out of articles", and has been codified here for at least 5 years, and at WP:NOT and other places for far longer. Bad faith behavior is dealt with through other processes, and not relevant to this discussion. There is no first-mover advantage. There's a consensus-having advantage, and while things are being negotiated, we talk it out. If a bad-faith actor is stonewalling consensus, deal with that elsewhere. But legitimately contentious material should be left out of articles unless there is consensus to include it. --Jayron3218:53, 10 May 2022 (UTC)
Good faith disagreements over content have always defaulted to "leave content out of articles" This is flatly untrue. Disagreements currently - and have always - defaulted to leaving the longstanding version in place until consensus is achieved. This is necessary because a lack of consensus is a failure state, and failure states need resolve in a way that preserves article stability; outside of situations like WP:BLP, the longstanding version, which has had many editors view it and has had no objections until now, is safer and enjoys more consensus than an article where removals potentially leave it lopsided or incomplete, with no degree of consensus backing it up. And my point is not that the situation I described is an abuse of process - if someone feels strongly that text is wrong and needs to be removed, they're naturally going to be disinclined to accept consensus to include it unless it is glaringly indisputable, especially given that that consensus partially depends on the strength of people's arguments; and will naturally argue at length to exclude it, including starting an RFC. This will result in instability in longstanding articles, and versions ending up stuck in place for as long as a month on end despite being unequivocally disputed and never enjoying any sort of consensus (not even the implicit consensus that the longstanding version enjoys.) --Aquillion (talk) 19:16, 10 May 2022 (UTC)
I'm not Aquillion, but here's my answer: Yes, there's a conflict. No, it's not so obvious that any passing editor will notice it at a glance. No, the conflict doesn't affect most edits, as it is (in practice) only seen when someone tries to remove long-standing text and editors can't come to an agreement about what to do. (NB: "remove", "long-standing", and "no consensus either way" are required for the conflict to matter. If you don't have all three qualities, then there is no conflict between the policies.) However: Yes, it's a big enough conflict that some editors seem to selectively name the policy that aligns with the outcome they prefer and "accidentally" forget to mention the existence of the other policy.
Fundamentally, if we are going to solve this, we have to decide whether the policy is – in the event that editors are unable to reach an agreement/consensus – either that we keep disputed long-standing text or that we remove disputed long-standing text. Right now, ONUS says that we remove disputed long-standing text unless editors agree to keep it, and NOCON says that we keep disputed long-standing text unless editors agree to remove it. WhatamIdoing (talk) 19:03, 10 May 2022 (UTC)
I'm okay with such a discussion. I have been operating for years that there was no such conflict, but it is clear that there are different interpretations of the policy, and clearer language would help ameliorate that. I think it's important that such policy changes be based not on "bad faith" actors, but instead should reflect best practices of good faith editors acting in the best interest of the encyclopedia. Aquillion's scenarios are not helpful, as literally they can be retooled to imply that ANY policy is open to the abuse of bad-faith actors, (as I demonstrate above) and we should instead focus on best practices rather. --Jayron3219:06, 10 May 2022 (UTC)
My point is not bad-faith actors (in fact, the fact that you characterize my description of the normal outcome of your proposed policy change as being "bad-faith" is a more severe indictment of your proposal than anything I could say about it.) My point is that, given that no-consensus outcomes are a failure state between people who feel strongly enough about their preferred version to result in an extended conflict, the default outcome of your proposed change would be disruptive and would result in a version with no consensus at all ending up on the article for an extended period of time over a version that enjoys implicit consensus - coupled with extended article instability, since any forceful objection to any longstanding text would require an immediate article change. --Aquillion (talk) 19:21, 10 May 2022 (UTC)
I sort-of mentioned this below, but I feel that the ideal solution is to incorporate the idea of implicit consensus more clearly into WP:ONUS. This basically reflects current practice and would answer most of the objections - it is presumed that when many editors have seen text without changing it, it enjoys a degree of consensus that requires affirmative consensus to remove, whereas low-traffic articles might not enjoy the same protection. (I also think that part of the reason these discussions go nowhere is because people are largely talking past each other in that respect - the people who side with WP:ONUS over WP:NOCON are picturing obscure overlooked bits of text saying something terrible, whereas the people who side with WP:NOCON are picturing longstanding text in a high-profile but controversial article, where someone removes a paragraph that has seen substantial editing and which has had a large number of eyes on it, then argues it should be kept out during discussions simply because it has never been discussed before.) At an absolute bare minimum, I would argue that the WP:GA version of any article automatically establishes at least a weak consensus for everything in it at the moment when it is promoted to GA - clearly removing and keeping something off of a GA requires affirmative consensus to do so. But this comes down to the more basic question that ONUS incorporates our concept of consensus, which is a bit nebulous by design. --Aquillion (talk) 19:10, 10 May 2022 (UTC)
My opinion is that they conflict to a degree, but that people have ignored the conflict by relying on the concept of implicit consensus to effectively render ONUS moot in most situations (which does answer some of Jayron's objections above, in the sense that text that has had few eyes on it lacks implicit consensus.) But clearly Jayron's suggestion does not reflect any sort of current practice - I've been editing for 18 years and the practice has always been to leave longstanding text in place during a dispute; people might sometimes WP:BOLDly remove it and sometimes it goes back and forth, but when it's clear the dispute isn't going to be resolved by editing the article is supposed to go back to the last stable version. Some people may disagree with that, or feel that that practice doesn't reflect ONUS, but the practice is 100% clear. In my 18 years here I've seen numerous disputes get stabilized by someone reverting to the last stable version while it has hashed out. I have never (not once) seen anyone follow Jayron's proposal to default to leaving disputed-but-longstanding text out outside of WP:BLP-sensitive statements, which are specifically called out as a special case in our policies. I have never even seen anyone suggest that it would be an appropriate way to approach a dispute outside of arguments on this talk page. When there's an intractable dispute on a stable, longstanding article (ie. one that has existed long enough to have a degree of implicit consensus), you go back to the last stable version until it is resolved - anyone who has edited controversial articles knows that this is current practice. --Aquillion (talk) 19:10, 10 May 2022 (UTC)
It looks like some discussed edits would be a major change, yet this this discussion is hard for anybody to understand or participate in. IMO any proposed change should be clearly specified here. Sincerely, North8000 (talk) 16:08, 10 May 2022 (UTC)
Nearly every editing decision also takes into account other variables. Including in defining / interpreting the terminology in the discussed provisions. How new is the proposed change/material? To what extent is it contested? Are other policy considerations being invoked in the conversation? (WP:Ver, WP:weight etc.) To what extent is the previous version "established"? This makes an explicit or precise "flow chart" guidance for those trillions of different possibilities impossible, so I'd advise against attempting that here. But these two policy provisions do provide influential guidance. IMO they do somewhat conflict/overlap, and persons here trying to "make it work" have been espousing seeing this through the lens of some some good views which eliminate/navigate the conflict but which are not policy. I think that the common meaning of "onus" is for new material, but it doesn't say that. Perhaps making that change would substantially reduce the overlap/conflict? North8000 (talk) 20:01, 10 May 2022 (UTC)
I think that this raises some good points; and has things that I have not considered. I think this may be a good idea to start developing (not starting, but pre-writing) an RFC for clarification of the policy. I've always understood Wikipedia policy, per WP:ONUS, to be "It is better to be silent than wrong" however Aquillion's position, which is that longstanding text has implicit consensus, is also a reasonable interpretation (though I don't agree) of best practice, and I can see where those positions come into conflict. Perhaps it's time to start a new thread where we spend some time developing a neutral and concise RFC to bring to the community so we can see where consensus is on clarifying and harmonizing policy around this matter. --Jayron3220:09, 10 May 2022 (UTC)
I should be clear (I mentioned this above, but I should mention it here too since it might be part of any resolution) that I don't think time-passed is the primary source of implicit consensus, just a usually-convenient shorthand. What really matters is how many editors have read the text without objecting to it. An article that has existed for ten years with only one or two editors probably doesn't enjoy implicit consensus for any of its text, whereas a high-traffic controversial article where everything is approached with a fine-toothed comb can probably be presumed to have it after just a few months. --Aquillion (talk) 20:22, 10 May 2022 (UTC)
@Aquillion, what I haven't yet understood from your comments is this:
Something gets added to an article.
Many, many page views/edits later, that something is still present, so we say that it has "implicit consensus".
Someone disputes the long-standing text.
Does disputed text have implicit consensus? Or would it be more accurate to say that it used to have implicit consensus but now it is disputed?
Imagine further that there are significant, substantive discussions about whether this content belongs in this article. The result is that editors' views are very evenly divided. There are good reasons to include it and good reasons to exclude it. What editors really agree on is that there is no consensus about what to do.
When there is no consensus, is there also still an implicit consensus? It seems to me that these two states are mutually exclusive. WhatamIdoing (talk) 22:49, 10 May 2022 (UTC)
BLP-sensitive statements aside, text that is longstanding and which many editors have read without objecting to has implicit consensus even if someone disputes it - you'd need at least some form of consensus to overturn it, which just one person objecting doesn't qualify as. After all, if any objection was enough to overcome implicit consensus, it wouldn't have any meaning. When discussions break down, the older consensus remains in place - this is how it works in every other context, and WP:NOCON is very clear about how it works. --Aquillion (talk) 05:03, 11 May 2022 (UTC)
Imagine this scenario:
Alice boldly adds something to an article.
Many, many page views/edits later, Bob boldly removes it.
Many, many page views/edits later, Carol objects to Bob's removal.
How many times did we achieve implicit consensus? Which version counts as the "older consensus" if the discussion between Bob and Carol breaks down? WhatamIdoing (talk) 00:14, 12 May 2022 (UTC)
Or if its a small change that 100% of those discussing agree on, try just semi-boldly doing it? BTW, like usual, I'd advocate doing the homework 100% before putting out an RFC, and come up with something that we can get behind and make the case for. If we just put something this complex out too quickly, then we'd end up with 300,000 words and 10 additional proposals and no decision. North8000 (talk) 21:22, 10 May 2022 (UTC)
If I understood it correctly, which I never did all the other times, if we don't have all three of "remove", "long-standing", and "no consensus either way", then there is no conflict between the policies and when do have all three, the problem is selectively name the policy that aligns with the outcome they prefer and "accidentally" forget to mention the existence of the other policy. That right? Selfstudier (talk) 22:09, 10 May 2022 (UTC)
Almost?
When we don't have all three of the relevant qualities ("remove", "long-standing", and "no consensus either way"), then we're going to quote you a different relevant page, e.g.:
"You're trying to re-word things, rather than remove them. I don't like your writing style, so STYLEVAR says to stop messing with my Wikipedia:Brilliant prose."
In the case of not having all three relevant qualities, it doesn't really matter which page we point at, because all of them end up in the same practical outcome (namely, revert everything unless and until there is a consensus otherwise).
When we do have all three qualities, then we get divergent results:
"I don't like that long-standing content, so I'm removing it per ONUS. You have to prove to me that there is a consensus to keep this in the article, because Policy Says you have 'to achieve consensus for inclusion' before it can be included. If there's no consensus, the content stays out of the article."
"I love this long-standing content, so you can't remove it per NOCON and QUO. You have to prove to me that there is a consensus to remove this from the article, because Policy Says to 'retain the version of the article as it was prior to the proposal or bold edit'. If there's no consensus, the content stays in the article."
Yeah I've definitely noticed this conflict before. I think I usually invoke ONUS rather than NOCON in those cases (and ONUS is definitely the way to go for BLPs). I personally think that if some content is controversial, we should not be saying it without a consensus, and would support changing NOCON to reflect this. Saying something in an article that is contentious (and thus possibly wrong/undue etc) is worse IMO than not saying anything. Galobtter (pingó mió) 23:30, 10 May 2022 (UTC)
I don't believe that ONUS and NOCON conflict, as I articulated previously in this comment. I tried to clarify the policies with an RFC two years ago, but I think that what I had proposed was misunderstood, so I did not get a clear understanding of community's feelings. Kolya Butternut (talk) 02:01, 11 May 2022 (UTC)
I was trying to recall what I had said about this in a previous topic. Kolya Butternut helped out by linking to their post which was just under one of mine. I agree with WhatamIdoing's comment. Based on what we have now it's not clear to me if we should view the fact that something is long term stable as justification for keeping it in in the case of no consensus or removing per ONUS in such a case. Along the same lines how strong is implicit consensus? If 2 editors favor a change and 1 opposes is that a consensus for change or not (assuming reasonable arguments all around) Note even in highly trafficked articles questionable content/changes can be unnoticed. This is particulary likely if there is a debate about one part of the article and a change that looks minor is made to a part that wasn't being debated. It can also happen if a big, debated edit is made after lots of talk page back and forth. We shouldn't assume that editors have really scrutinized an older change simply because many other edits have been made since. Also note that ONUS only applies to new content (edits subject to WP:V), not other edits where a consensus related debate may apply. Examples that aren't subject to WP:V include the order of content in an article or what content should be in the lead vs just in the body. However, content that was added and not challenged at the time should still be subject to ONUS in my view. Thus any content that has never had explicit consensus established (talk page discussion, etc) would be subject to an ONUS based challenge. This would address questions like how long does content have to be in the article before we flip from NOCON=keep vs NOCON=remove. It also protects from a case where editors would have objected had they noticed (watched the article) when the change was first made rather than noticing it say 6 months later. If the page is well trafficked and no one talk about a change but editors support it, it should be reasonable to assume a talk page discussion will establish consensus after the fact. If it can't, the content probably shouldn't be included. Finally, since it was mentioned previously, I would keep long term content in the article during the consensus discussion. If some addition is being challenged a year later it isn't critical to remove it now vs at the end of the discussion if DUE-ONUS is the only concern. Springee (talk) 03:35, 11 May 2022 (UTC)
To me it's pretty simple if we understand one thing as a preface. Most editing decisions are made by weighing multiple factors. Most wp policies prima facie go against that by seeking to specify what should happen based on just one or two considerations, but then make themselves useful and workable in the Wikipedia world by softening themselves by adding wording that makes them non-categorical. The result is that they usually only put "fingers on the scale" in such decisions rather than unilaterally prescribing what should happen.
Onus Depending on how one interprets its common meaning, wp:onus puts one or two fingers on the scale. One also exists in the world of common sense which is a finger towards exclusion of disputed material. The "probably" second finger is towards exclusion of new material. The context and common meaning of the relevant sentence in ONUS tilts towards applicability only on new material, but it does not explicitly say so in its relevant operative sentence.
NOCON wp:Nocon puts a finger on the scale towards the status quo.
Sometimes these "fingers on the scale" are in opposite directions, but as long as one understands how wikipedia works, such is not a "conflict". Of course there are other "fingers" on the scale besides these policies. One is that we're here to add material and build an encyclopedia. Sincerely, North8000 (talk) 10:27, 11 May 2022 (UTC)
WP:ONUS is part of WP: Verifiability, which is a content policy not a conduct policy. This discussed portion of WP:ONUS merely reiterates how WP: Consensus works. Kolya Butternut (talk) 13:12, 11 May 2022 (UTC)
I tend to agree with you that ONUS reiterates how consensus works in practice (albeit for a rather narrow set of circumstances). However, the (relatively new) NOCON section of Wikipedia:Consensus is too simple. It puts a finger on the scale towards the status quo (assuming editors can agree on what that was, which all experienced editors know is not always simple), and it doesn't do a good job of saying "Um, except for all the exceptions, which include adding disputed non-BLP material". WhatamIdoing (talk) 00:19, 12 May 2022 (UTC)
WP:Silence is weakest form of consensus, and to make matters worse for that argument, you have to consider that "silence" no longer enjoys the status of "implicit consensus" once a challenge has been issued, or an objection has been raised so the very fact it even goes to discussion at all will be enough to question it and take that status away unless fully supported for keeping. Huggums537 (talk) 03:47, 29 July 2022 (UTC)
List of possible meanings of the third sentence of ONUS
The discussion above makes it clear that there is no current consensus regarding the meaning of this sentence:
The onus to achieve consensus for inclusion is on those seeking to include disputed content.
Following discussion which results in no consensus, disputed material may not be added or, unless it has (or had) implicit consensus, retained. - Butwhatdoiknow (talk) 17:36, 11 May 2022 (UTC)
Following discussion relating to verifiability which results in no consensus, disputed material may not be added or retained. - Butwhatdoiknow (talk) 17:36, 11 May 2022 (UTC)
Following discussion relating to verifiability which results in no consensus, disputed material may not be added or, unless it has (or had) implicit consensus, retained. - Butwhatdoiknow (talk) 17:36, 11 May 2022 (UTC)
If you want to include something in an article, and anyone strongly objects, then it's your job to prove there is an active, positive agreement to include your material in that article. The people who want to remove your material do not have to prove that there is a consensus to remove the disputed material; they only have to establish (a) that they dispute the inclusion of the material (for any reason at all, including bad reasons) and (b) that you have not provided written evidence of an active, positive consensus to include your material. It is not enough for you to claim consensus without evidence; "presumed", "implicit", or "silent" consensus does not count for this provision of policy, and WP:STATUSQUO (which is only about what to do with an article during discussions about the dispute, and not about what to do after the discussion has ended) is irrelevant. (per WhatamIdoing, below at 02:21, 12 May 2022 (UTC)). - Butwhatdoiknow (talk) 05:52, 16 May 2022 (UTC)
Discussion of list of meanings of ONUS third sentence
IMO none of the above. Those are all categorical and onus isn't categorical and, by the nature of how Wikipedia works, can't be. North8000 (talk) 17:52, 11 May 2022 (UTC)
I think what North8000 is saying is that ONUS is better understood as a rough guideline rather than a hard and fast rule. I do think that part of the problem is that people are trying to treat it as too categorical and giving it too much force. Our policies are usually looser, since that encourages discussion and consensus-building; some of the wording and interpretations of ONUS seem closer to the way WP:BLP works (and even BLP only has that degree of force for actually BLP-sensitive statements.) I don't think that that would help dispute resolution - we give BLP-sensitive statements that "always default to removal" force because of the specific risk of harm, it's not and has never been our general policy. More generally, we ideally resolve things through discussion and consensus-building, not through wrangling over legalese; making policies like ONUS too rigid goes against that. --Aquillion (talk) 18:12, 12 May 2022 (UTC)
During the last few (five?) years, I feel like the community (or at least the dramaboards parts of the community) has moved away from the kind of consensus building that involves compromise, common sense, and finding things we can all support, and towards wrangling over legalese. We have always had wikilawyers, and it's good to write policy to minimize the damage that a wikilawyer can cause by quoting policies and guidelines, but I think we have more people believing that citing WP:SHORTCUTS and demanding that their view be enacted in the article is the best, or even the only correct way to write good articles. WhatamIdoing (talk) 22:45, 12 May 2022 (UTC)
Right, the arguably loose language might occasionally lead to a discussion but consensus will still rule at the end of it. I guess what I am saying, is it really broken? User:Selfstudier (talk) 17:57, 11 May 2022 (UTC)
User:Selfstudier, I'm just asking what the "loose language" means. We can talk about whether it is broken or not later. Does the "loose language" incorporate any of the meanings above? Which one(s)? Does it also have additional meanings not listed? What are they? - Butwhatdoiknow (talk) 20:15, 11 May 2022 (UTC)
The problem with ONUS vs NOCON is what should happen when consensus cannot rule at the end, because consensus cannot be achieved. WhatamIdoing (talk) 00:21, 12 May 2022 (UTC)
I get what you're saying, but sometimes consensus-building mechanisms break down - if people can reach a consensus then they don't need to rely on WP:ONUSorWP:NOCON. But we do need some sort of guideline for what state articles end up in when there's no clear consensus - and currently what happens is that we end up with situations where one person says "removing this, get consensus per WP:ONUS", then someone else restores it and says "no, you need consensus to remove it per WP:NOCON." That's not desirable. The guidelines do not have to be strict - in fact, they shouldn't be; honestly one problem with ONUS is that it is simultaneously vague about what it covers and incredibly stridently-worded. I feel that making the guidelines too strict in either direction discourages people from coming to the table and compromising, as opposed to just smacking each other with policy shortcuts. But at the very least the two shouldn't directly contradict each other. --Aquillion (talk) 21:56, 11 May 2022 (UTC)
I agree strongly with this. The popularity of WP:BRD adds fuel to this fire, as removal of longstanding content is frequently seen as the B. I also want to echo Aquillion's point higher above that some reference to WP:IMPLICIT is needed. Doing so will lead to more status quo bias in the policy, which I'd prefer over bias toward exclusion of content. Firefangledfeathers (talk / contribs) 00:44, 12 May 2022 (UTC)
I believe a common POV among experienced editors is, in Jimmy's words, "Zero information is preferred to misleading or false information". I assume that you mean that you'd rather err on the side of including, e.g., accurate trivia and well-sourced but overly detailed facts, but you still want all serious problems removed. WhatamIdoing (talk) 02:31, 12 May 2022 (UTC)
You're reading me mostly correctly. We're talking, after all, about material that's already cleared the bar of verifiability. Technically, adding or acknowledging a little status quo bias in ONUS wouldn't lead us to only err on the side of inclusion, as I'd equally support continued removal of content that was removed and kept out over many months/revisions. Firefangledfeathers (talk / contribs) 02:44, 12 May 2022 (UTC)
About B's dozen sentences: I don't think that we should prescribe the state of the article under which discussions happen. Sometimes you should discuss with m:The Wrong Version showing, and sometimes you should discuss with The Right Version, but it depends on circumstances, the nature of the problem, the real-world importance of the problem, whether editors are willing to discuss (sometimes, removing their contribution is the only way to motivate them to join the discussion), and (of course) the fact that you don't actually know, until the end of the discussion(s), which version is The Right Version. ONUS does not specify that the state of the article during discussions is critical, and caring too much about the state of the article during discussions is a bad practice in general; therefore, I conclude that it should not be interpreted to apply strictly in such cases. Therefore, I reject 1 through 6 as being focused on the wrong point in the process.
I think that sentence 8 is closest to my understanding of ONUS. The reason that I reject the others are:
7 is a subset of 8; it is true but is too narrow. Focusing only on additions puts us back in the realm of bad practices of edit warring to have The Right Version in the article while the discussion happens. If I get my version, then ONUS would have no effect, because I wouldn't be "adding" it after failing to achieve consensus. I'd only be "retaining" it. This would promote edit warring. Version 10 has the same problem.
Material that is currently disputed no longer has implicit consensus. IMPLICITCONSENSUS says An edit has presumed consensus unless it is disputed or reverted. As soon as the dispute arises, the material no longer has any sort presumed/implicit consensus. Therefore 9 and 12 are wrong.
ONUS does not require that the dispute be related to sourcing. Therefore 10, 11, and 12 are wrong.
The current question is not what the third sentence of ONUS should say, but what it currently says. Are you saying that (a) 8 is what it currently says? If not, (b) is its current meaning in the list? If not, (3) what additional meaning should we add to the list to make it complete? - Butwhatdoiknow (talk) 00:54, 12 May 2022 (UTC)
What it currently says is "The onus to achieve consensus for inclusion is on those seeking to include disputed content", which is so obvious that I don't believe that's what you meant to ask. I assume that you meant to ask what that means.
If you feel that my response is too focused on what the policy should mean, then I could perhaps translate the current sentence's meaning in this more practical way: "If you want to include something in an article, and anyone strongly objects, then it's your job to prove there is an active, positive agreement to include your material in that article. The people who want to remove your material do not have to prove that there is a consensus to remove the disputed material; they only have to establish (a) that they dispute the inclusion of the material (for any reason at all, including bad reasons) and (b) that you have not provided written evidence of an active, positive consensus to include your material. It is not enough for you to claim consensus without evidence; "presumed", "implicit", or "silent" consensus does not count for this provision of policy, and WP:STATUSQUO (which is only about what to do with an article during discussions about the dispute, and not about what to do after the discussion has ended) is irrelevant."
I agree, WAID, that the article state during discussion is overemphasized. At least, I agree academically, as I do sometimes feel very invested in which version is up. Since that overemphasis is a frequent cause of edit warring, I think the policy needs to address the reality. Clearer policy here can lead to less edit wars. In practice, ONUS only comes up before and during talk page discussion, not after a discussion's consensus is evaluated. Mostly NOCON seems to come up at the end. Firefangledfeathers (talk / contribs) 00:56, 12 May 2022 (UTC)
About the academic agreement: Sometimes the Wrong Version is very, very wrong indeed, and I expect the discussions to last for weeks, or even months. It can be hard to focus in those situations. WhatamIdoing (talk) 02:25, 12 May 2022 (UTC)
I also think 8 is the correct meaning, and the ideal meaning.
It was noted above that the dispute here over ONUS vs. NOCON probably arises in large part from different editors picturing different scenarios. So I will explain why I support ONUS, and why I believe the opposing scenario meant to favor NOCON is not actually a problem under ONUS. This is a scenario I've seen evidence of firsthand multiple times.
So, imagine an inexperienced editor inserts text on an article. This text contains sources, but the content is actually FRINGE or OR. The article it is inserted on has moderately-low or lower traffic and is on a complex topic. The text is just convincing enough that while some other editors do see it in passing, they are not familiar enough with the topic to notice the problems with it. It sticks around for years, with other edits occurring to other parts of the article and it seemingly thus gains implicit consensus. However, eventually, another editor figures out it is really bad text and deletes it with a mean red -4,000 or whatever in the edit history. Reversion and discussion ensues, but there's only one other editor in it, and they support the text because it looks superficially convincing or for POV reasons (maybe it was the original adder). Perhaps the remover knows the topic but not Wikipedia bureaucracy and detailed policy and thinks there is nothing left to do but give up. Which side should policy support?
The editors who favor NOCON and the status quo seem to often have in mind scenarios where at a high-traffic but socially controversial topic, someone removes text and then games the system to keep it out. However, ONUS as written does not contribute to this problem - at such an article it is very easy to establish a quick consensus that supports the material; indeed one possibly already existed on the talk page.
Right now ONUS basically says 'if content doesn't have consensus, it doesn't stay', which is better than 'it has to stay unless we get a consensus to remove it'. In cases where there is not a consensus for material, it should stay out, lest the encyclopedia accumulate garbage. We should not privilege material just for happening to lack scrutiny and sticking around for a while. Think about how huge the encyclopedia is and how little-scrutinized most of it is. Most passing editors are not too familiar with a topic and are biased toward letting through (or not bothering with) something that looks superficially okay.
All this was a central concern at this aforementioned RfC, which was closed with this statement: There is an overwhelming consensus against this proposed change, primarily due to concerns that it would undermine verifiability and efforts to remove poorly sourced information. I supported this clarification (or something along those lines) of NOCON. Crossroads-talk-05:11, 12 May 2022 (UTC)
Read the footnote in BURDEN: "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."
In other words, BURDEN generally does not apply to (plausibly) sourced content. Generally, under BURDEN, editors are only required to provide one source. Requiring editors to provide an unlimited number of sources until an POV pusher agrees would create its own set of problems ("Sure, you gave me sources from the Pope, the Queen of England, and Albert Einstein to support this statement, but I reject them all as being completely unreliable. Bring me another rock – or don't, because I will never agree that any source is reliable unless the source supports my POV.") WhatamIdoing (talk) 06:08, 12 May 2022 (UTC)
Crossroads, apologies for bringing up previous disputes; my point here isn't to go into them in-depth here but to try and get an understanding of what you feel WP:ONUS means and how, under your reading, it would actually affect the sorts of disputes we run into every day. We've previously had disputes on, among other articles, Mermaids (charity), Gina Carano, and Heterodox Academy where I removed longstanding text from articles that I don't think had previously obtained a consensus; you restored it, and discussion on the talk page broke down without producing a clear consensus to include. In each case, my understanding is that the text remained merely because it is longstanding - in the last case I eventually had to go through an RFC to remove it. Is it your assertion that in each of those cases the text should be removed, or at least that the onus is on you to demonstrate consensus if you want it to stay (and that an inconclusive discussion that breaks down roughly evenly means they get removed absent a formal RFC to keep them?) eg. do you feel that the Heterodox Academy text should have been removed, and that the onus was on you, and not me, if you wanted to restore it, once it was clear it was disputed and there was no prior consensus? Again, my point isn't to rehash those issues here; I want to understand how you think this would work in practice, what you feel qualifies as a consensus necessary to retain text, and so on, and to make it clear what I think is the implications of pushing for option 8 with no concession towards implicit consensus. Because I want to make sure we're on the same page in terms of what the implications here are - the next time I remove something because I feel it's undue, and you object because you feel it's due, how do you feel WP:ONUS will apply to that? (If you feel implicit consensus applies in all of those situations, that's fine, but that's 3 and not 8, so I want to be certain where the dispute here is.) --Aquillion (talk) 17:38, 12 May 2022 (UTC)
I think option 8 doesn't account for situations such as when material has been in an article for years and it is just being disputed by one person. Kolya Butternut (talk) 17:01, 12 May 2022 (UTC)
If the material is truly disputed by just one person, then it should be possible to find one or two other editors who support its inclusion. In that case, we'd be able to form something like a consensus to retain the material, and ONUS wouldn't really apply. WhatamIdoing (talk) 22:38, 12 May 2022 (UTC)
Only one person did the original bold insertion. The policy doesn't say it will self-destruct after a deadline and adding a deadline would give a first mover advantage based on a vague ("years" or "longstanding") assertion which could trump multiple opponents afterward. Peter Gulutzan (talk) 13:19, 13 May 2022 (UTC)
Since all are categorical, none of them account for other factors relevant to the outcome. North8000 (talk) 17:56, 12 May 2022 (UTC)
My issue is that I think that the applicability of implicit consensus is the main factor in the dispute. Removing the concept of implicit consensus (or making it so ONUS overrides it, which amounts to the same thing) would have fairly drastic implications to how we resolve disputes. --Aquillion (talk) 18:07, 12 May 2022 (UTC)
I feel like WP:Consensus simply means, in part: The onus to achieve consensus for changes to consensus content is on those seeking the change. The third sentence of WP:ONUS is just part of this. Kolya Butternut (talk) 19:13, 12 May 2022 (UTC)
My note is mostly structural and two steps. For example, if Mary says "I generally like big dogs", first what she really means is that when deciding whether or not she likes a particular dog or breed, that bigness is a plus. Maybe outweighed by that particular dog being vicious and having bit her kid on the face yesterday.
So if we come up with very specific "meaning" ideas like "Mary likes all big dogs" or "Mary likes all dogs over 90 lbs" or "Mary likes all dogs over 80 lbs", through their specificity and explicitness we are inventing things that Mary never said. North8000 (talk) 19:33, 12 May 2022 (UTC)
In short, such efforts are in effect proposing a new revised version / meaning, not discerning a meaning of the current one. North8000 (talk) 23:14, 12 May 2022 (UTC)
A sincere answer is that the meaning is "The onus to achieve consensus for inclusion is on those seeking to include disputed content" applied and interpreted in the context of the preceding sentence, it's section header and how the Wikipedia system operates. If you would like me to attempt en explanation on that, the specific-simple-special-case version would be: If all other factors are neutral, inclusion of new disputed material requires a consensus. A more general case attempted explanation would be that in decisions on inclusion/exclusion of disputed content, apply an extra finger on the scale towards exclusion, and doubly so if it for new material. And in that context that a consensus puts a heavier finger on the scale than any of those. Sincerely, North8000 (talk) 00:21, 14 May 2022 (UTC)
Trying to put your general case in the simplified format I get: "If all other factors are neutral, the addition or retention of disputed material requires a consensus." Is that how you read the third sentence of ONUS? - Butwhatdoiknow (talk) 02:22, 14 May 2022 (UTC)
I understand your desire to get to the bottom of it, I still say the same, it's all worded so as to permit flexibility in edge cases and where it will never be entirely clear what the outcome might be. It is I suppose possible that nocon is an outcome (after endless discussions), most of the time consensus will sort it out. Like here, if this is really a concern, save some time and just go for the RFC. Selfstudier (talk) 10:39, 14 May 2022 (UTC)
@Butwhatdoiknow: You accurately restated the core principle in another post whcih you pinged me on. Regarding your second question, that can't be answered here....that would be determined by the Wikipedia system, and this merely specifies an input to it. North8000 (talk) 22:02, 14 May 2022 (UTC)
Butwhatdoiknow, that would not be a statement of my "general case" explanation, but would be special case version of it as applied to that particular situation. But such an example is useful for illustrating the underlying concept. North8000 (talk) 13:51, 14 May 2022 (UTC)
User:North8000: And the core concept is "in decisions on inclusion/exclusion of disputed content, apply an extra finger on the scale towards exclusion, and doubly so if it for new material. And in that context that a consensus puts a heavier finger on the scale than any of those," do I have that right? - Butwhatdoiknow (talk) 21:42, 14 May 2022 (UTC)
I would agree with that in principle, I wouldn't like to try and codify that, tho. My inclination is still that it ain't really broke and it's more like we are trying to find ways to break it. Selfstudier (talk) 21:52, 14 May 2022 (UTC)
Selfstudier, at this point I'm not trying to codify anything, I'm just trying to understand what the community thinks is the meaning of the current, "unbroken" third sentence of ONUS. Do you think the text I quoted from North8000 is the meaning of that sentence? - Butwhatdoiknow (talk) 03:28, 15 May 2022 (UTC)
@Selfstudier, it is broken, because we get divergent results, as described above. Aligned policies lead everyone in the same direction, including people who don't know all the policies; broken policies let me cite the SHORTCUT that supports my POV, and if you didn't know that another policy says the opposite – well, too bad for you, because I'll win our dispute. WhatamIdoing (talk) 05:40, 16 May 2022 (UTC)
Until another editor shows up who knows better? It's not supposed to be a contest anyway tho I agree it can be sometimes. Selfstudier (talk) 10:46, 16 May 2022 (UTC)
@Butwhatdoiknow: Exactly. Actually the first sentence is from wp:onus, the second sentence comes from how Wikipedia operates, but is good and needed explanation/clarification/information. North8000 (talk) 21:56, 14 May 2022 (UTC)
North8000, focusing just on the third sentence in ONUS, the core concept is "in decisions on inclusion/exclusion of disputed content, apply an extra finger on the scale towards exclusion, and doubly so if it for new material," right? Butwhatdoiknow (talk) 03:28, 15 May 2022 (UTC)
I think 7 or 9 is the closest to my interpretation. Basically, if you add something and someone reverts you relatively soon afterwards, then the text stays out unless you can get consensus. If you remove longstanding text and are reverted, then the text stays in unless 1) you can get consensus or 2) there is no consensus and retaining the text carries unusually high risk (e.g. negative BLP material). -- King of ♥ ♦ ♣ ♠05:03, 13 May 2022 (UTC)
@King of Hearts, when you read "The onus to achieve consensus for inclusion is on those seeking to include disputed content", do you really hear "If the disputed text has been there a long time, then the onus to achieve consensus is on those seeking to exclude the disputed content"? Or are you saying that's not what the text says, but that's what it ought to say/what the current practice is? WhatamIdoing (talk) 05:43, 16 May 2022 (UTC)
I would like to caution against giving implicit consensus too much weight as policy, as the conversation seems to be heading in that direction. While WP:NOCON acknowledges that we often maintain the previous version in the absence of consensus, it does not actually recommend that course of action as the desired result; and I'd be wary of encoding it as the preferred outcome by policy. There is the risk that it will encourage editors to revert articles back to the status quo ante version, using "there's implicit consensus" as the only reason for the revert; which is far from a good guideline of desired behavior. As WhatamIdoing recalled above, implicit consensus disappears as soon as someone challenges it.
At the very least, any clarification to ONUS and NOCON should warn editors to provide, each time, a summary of the arguments on which consensus was achieved. Even if the outcome is the same of retaining the original version, at least the editor challenging the content will have a basis on which to work to improve the article on top of that consensus, maybe looking for ways to get their desired outcome without contradicting those arguments; instead of the alternative of being stonewalled by a flood of camped editors undoing all attempts at improvement on the only basis of "revert to previous consensus per WP:NOCON" with the force of policy. Wikipedia already has a bad reputation of veteran editors behaving that way; let's not encode in policy that undesired behavior, and instead guide editors to avoid it by explaining the most constructive way to protect a challenged article. Diego (talk) 07:46, 13 May 2022 (UTC)
The types of "clarifications" described above would have huge impacts and would require consensus in a widely advertised RFC. Maybe not for a low impact tweak. North8000 (talk) 13:27, 13 May 2022 (UTC)
I, also, find the concept of "implicit consensus" to be problematic. Either we've had the discussion, and the consensus is taken from that discussion, or we haven't, and the status of consensus on any matter is merely "unknown". It doesn't become more known as time passes. This isn't always a problem; if something isn't contentious, the matter never comes up. But as soon as something DOES become contentious, an actual taking of consensus needs to occur. It isn't sufficient to say "no one objected until now..." as though that's somehow consensus. Once a good-faith reasonable challenge has been made, we need to assess the situation more closely. "No one objected until now" is only sufficient until someone objects. --Jayron3214:42, 13 May 2022 (UTC)
I agree with Jayron32. Most of our articles are thinly edited. We shouldn't codify any language that suggests the standing text has consensus merely because it's aged. For text that's been discussed or in articles that are frequently edited by many editors, implicit consensus has a valid basis, but consensus can change, new sources can become available, due weight can change, and we should not be publishing guidelines that may be misinterpreted in such a way that they hinder article improvement. SPECIFICOtalk15:42, 13 May 2022 (UTC)
Not necessarily. The operative part of that essay is already in the first sentence. People seem to think that there's some kind of hidden time limit at which time Wikipedia articles become written in stone, but that essay already makes clear that consensus exists "until disagreement becomes evident". As soon as someone removes some text from an article (excepting obvious vandalism, etc.), it is evident they disagree. Unless there has already been a discussion about that text, the text is now no longer in consensus. --Jayron3211:06, 16 May 2022 (UTC)
@North8000 Well yeah that's the thing, I don't believe you can solve this issue with a "low impact tweak" to the guidelines, that's why I'm raising the warning in this section. I agree if we are to include such clarifications it should be widely publicized, gathering feedback from the community at large. Diego (talk) 16:56, 13 May 2022 (UTC)
Maybe, maybe not. It might be possible to describe a best practice without insisting that it be followed every time. For example, "If someone cites this section, it may be helpful if you briefly outline your thoughts on the talk page. For example, you might post links to recent discussions about this matter on the talk page, which could help others understand why you believe there is support for including (or excluding) the material." WhatamIdoing (talk) 05:49, 16 May 2022 (UTC)
Yes, that would be a good course of action, in line with what I was suggesting. Though I'm afraid it also has the potential to require wide, well publicized consensus as North8000 said. Diego (talk) 20:59, 18 May 2022 (UTC)
Fixing this
One thing abundantly clear from the above discussion is that our policies ONUS and NOCON are not clear, open to multiple reasonable interpretations. What say we focus on clarifying them: let's talk about what they should say, and then draft an RFC to change them? Levivich20:49, 23 May 2022 (UTC)
I fear that jumping to "what should we do about ONUS and NOCON" is likely to result in "no consensus." End result: the existing text stays and there is no improvement.
It seems to me that the first issue is that ONUS is ambiguous. My plan: (a) collect all the possible meanings of the current ONUS text (we've pretty much done that), (2) do a survey regarding which meaning is correct for the current text and see whether we can, at least, have a majority view regarding what the current ONUS text says. (3) Revise ONUS to more clearly say what the majority of folks says it means. And then, finally, (4) work on resolving any conflict between ONUS and NOCON.
@Butwhatdoiknow: I'm certainly not opposed to your plan--hey, give it a shot--but I think you'll find that the answer to #2 is no, w don't have a majority view regarding the current ONUS text, and if we did have one, we wouldn't be having the problems with ONUS as we have now. So I don't think we're ever going to get to #3 (to "say what the majority of folks says it means"). I think the discussion above between Jayron and Aquillion was a rather perfect "microcosm" of the two leading views I've seen "in the wild" in editing discussions, and that's what we'd see on a larger scale if we had an RFC about the current meaning of ONUS. But of course I could be wrong... give it a shot. Levivich03:27, 26 May 2022 (UTC)
...and I personally wouldn't focus on the conflict between ONUS and NOCON. I think that's not the best place to focus. I think the current "system" as written is not particularly helpful to us editors by and large, and so trying to "tweak" the current system won't lead to meaningful improvement (or even to consensus for any tweaks). I think, instead, it's better to focus on the goals or issues that ONUS and NOCON seek to address, which, namely, are about CONSENSUS, SILENCE, BOLD, BRD, and things like that. In other words, I think it'd more productive, instead of talking about the meaning of the current language, or conflicts between the current language, to talk about things like: when should/shouldn't an editor revert? When should/shouldn't "longstanding content" be considered (and what does "longstanding" mean)? What sort of consensus do we need to add something to an article, remove it, or change it? These are the sorts of questions that I think, if the community could answer them, would lead us to the best language for ONUS, NOCON, and other policy pages. Levivich03:31, 26 May 2022 (UTC)
The onus to achieve consensus for inclusion is on those seeking to include disputed content.
For the survey below, please select from the meanings of that sentence identified in the discussion above:
LIST OF POSSIBLE MEANINGS OF ONUS THIRD SENTENCE
During discussion, disputed material may not be added.
During discussion, disputed material may not be added or retained.
During discussion, disputed material may not be added or, unless it has (or had) implicit consensus, retained.
During discussion relating to verifiability, disputed material may not be added.
During discussion relating to verifiability, disputed material may not be added or retained.
During discussion relating to verifiability, disputed material may not be added or, unless it has (or had) implicit consensus, retained.
Following discussion which results in no consensus, disputed material may not be added.
Following discussion which results in no consensus, disputed material may not be added or retained.
Following discussion which results in no consensus, disputed material may not be added or, unless it has (or had) implicit consensus, retained.
Following discussion relating to verifiability which results in no consensus, disputed material may not be added.
Following discussion relating to verifiability which results in no consensus, disputed material may not be added or retained.
Following discussion relating to verifiability which results in no consensus, disputed material may not be added or, unless it has (or had) implicit consensus, retained.
If you want to include something in an article, and anyone strongly objects, then it's your job to prove there is an active, positive agreement to include your material in that article. The people who want to remove your material do not have to prove that there is a consensus to remove the disputed material; they only have to establish (a) that they dispute the inclusion of the material (for any reason at all, including bad reasons) and (b) that you have not provided written evidence of an active, positive consensus to include your material. It is not enough for you to claim consensus without evidence; "presumed", "implicit", or "silent" consensus does not count for this provision of policy, and WP:STATUSQUO (which is only about what to do with an article during discussions about the dispute, and not about what to do after the discussion has ended) is irrelevant.
A combination of the above (please identify).
None of the above (please provide the alternative current meaning of the sentence).
From the list above, please select one or more meanings of the current third sentence OF wp:ONUSas written: "The onus to achieve consensus for inclusion is on those seeking to include disputed content". - Butwhatdoiknow (talk) 22:36, 3 June 2022 (UTC)
This means you! 13 (which is not very different in substance from 8, although I think it is much clearer about the "this means you" aspect). WhatamIdoing (talk) 03:15, 4 June 2022 (UTC)
14 by way of and 8 and 2; more specifically, the most important thing is what the result of a discussion is, so 8 is the primary meaning, with 2 as a secondary meaning. Crossroads-talk-04:15, 4 June 2022 (UTC)
15. Per my comments below, these are all far too categorical to accurately summarize the existing text; and the argument that a discussion ending in no-consensus results in omission reflects current practice is just obviously false, as anyone who has participated in that process is well aware. WP:IMPLICITCONSENSUS, meanwhile, is policy and is therefore incorporated into this policy's meaning. I also want to register my specific objection to the non-neutral wording of 13, which I feel is clearly prejudicial - it is totally inappropriate to devote several times the text to one option, invoking other policies to justify its position and so on. This "simple explanation" of ONUS is several times the length of the sentence it supposedly summarizes. --Aquillion (talk) 21:23, 16 June 2022 (UTC)
15. I think the text speaks adequately for itself. I don't think that attempting to further clarify the current wording is likely to be constructive. -- Visviva (talk) 21:28, 16 June 2022 (UTC)
13 though I get the concern that the others are summary statements while 13 is a more detailed statement. Regardless, 13 is how I would view ONUS. Springee (talk) 12:00, 21 June 2022 (UTC)
13 is how I've always interpreted it and how I believe it should be interpreted. valereee (talk) 17:22, 30 June 2022 (UTC) For those who are concerned about 13 being a non-neutral statement, 8+2 says the same thing IMO.
13. I think this is the correct interpretation. WP:SILENCE and "implict consensus" for articles that are infrequently edited or the topic has not been discussed on the relevant talkpage is has no real meaning. Hemiauchenia (talk) 22:22, 9 July 2022 (UTC)
The meaning of the CURRENT text at ONUS (discussion)
User:Blueboar thank you for adding to the survey. However, the question is not the original intent. It is the meaning of the current text as written. Do you think 13 best matches the current text? - Butwhatdoiknow (talk) 00:10, 4 June 2022 (UTC)
13 is just a long-winded way of saying 8 while preempting all the excuses people give for pretending that ONUS doesn't apply when the disputed content is something they want in the article. It is IMO "the meaning of the current text as written", just in a far more verbose and tetchy style. WhatamIdoing (talk) 03:12, 4 June 2022 (UTC)
In the archives… what eventually grew into ONUS was discussed in January and early February of 2013… and the initial language was added on 11 Feb, 2013 (by me). Blueboar (talk) 11:40, 4 June 2022 (UTC)
The original language does not appear to say anything about onus (what happens when there is no consensus regarding "whether specific information is irrelevant or trivial to the topic") or the process elaborated at 13. - Butwhatdoiknow (talk) 17:03, 5 June 2022 (UTC) @Blueboar: please reply. - Butwhatdoiknow (talk) 15:49, 6 June 2022 (UTC)
I don't believe the just-added text, "if a discussion results in no consensus then the disputed material may not be added or retained," is appropriate in a policy that many Wikipedians treat as having the force of law. (Nor can the tiny number of responses above be taken seriously as representing any sort of project-wide consensus.) Taken literally (which people will), this new language goes far beyond a mere burden of persuasion and effectively gives a formalheckler's veto to any motivated group of bad-faith actors. Of which we have many. (I am thinking particularly of the nationalists that occasionally swarm an article they don't like, but there are plenty of other examples.) They wouldn't even have to put in the work of edit-warring! More generally, WP:V should not be in the business of specifying page-level dispute resolution practices. I haven't responded to the survey above because I see no overwhelming need for exegesis here. Sometimes ambiguous policies are ambiguous for a reason. -- Visviva (talk) 06:07, 16 June 2022 (UTC)
I agree with Visviva. It certainly should not have done until the discussion was closed. From my experience, Visviva accurately describes what would happen if the new wording is included. I'm reverting it. Doug Wellertalk07:40, 16 June 2022 (UTC)
I agree with Doug Weller said and go far beyond that. Adding such a disastrously prescriptive and categorical statement to this high impact core area of a core policy would be just that and require much more that just interpreting it out of this general local discussion.North8000 (talk) 13:48, 16 June 2022 (UTC)
More generally, WP:V should not be in the business of specifying page-level dispute resolution practices. I haven't responded to the survey above because I see no overwhelming need for exegesis here. Sometimes ambiguous policies are ambiguous for a reason. This is an interesting statement (and one that seems to be excluded by even the exhaustive list of options above, which presumes that WP:ONUS should be categorical in some form.) Do you think it should be added to the RFC - something that makes it clear that ONUS is advisory or intentionally ambiguous and does not impose hard requirements? The fact that large numbers of people have contributed to the discussion without weighing in there suggests that there is a flaw or an omitted middle somewhere in it. --Aquillion (talk) 21:07, 16 June 2022 (UTC)
I think at least a "no additions are needed"/"the current text speaks for itself" option would probably be a good thing. Beyond that I'm not sure adding more options would be helpful give how many are already there. -- Visviva (talk) 21:21, 16 June 2022 (UTC)
@Visviva, we've had multiple long discussions about this. How many thousands of words do we need to spend discussing how confused certain editors are before you would agree that the current form isn't working for the editors who claim that they're confused by it? WhatamIdoing (talk) 07:07, 20 June 2022 (UTC)
FWIW, I am 100% in favor of removing the text entirely, as it already does more harm than good. I hadn't realized that option was on the table. I am skeptical that a consensus could be achieved for that either, however; many editors are likely very attached to their personal understandings of the policy. -- Visviva (talk) 18:59, 20 June 2022 (UTC)
I want to raise a particular objection to how this list provides a detailed rationale and explanation for 13 while all the other options are brief, terse sentences - that is obviously not a neutral way to word an RFC. Finally, every point where WP:IMPLICITCONSENSUS is mentioned raises WP:CONLOCAL issues in that declaring that implicit consensus has no applicability here would effectively moot it as a concept, which isn't something that can be done by a relatively small number of people on an unrelated talk page. The argument made in 13 in particular makes it clear that people are interpreting this as not being related to or confined to WP:V-related manners. In that case, why is it on this page, and why are we discussing it here? Policy set on WP:V cannot override WP:CONSENSUS, which establishes implicit consensus, without a truly broad and overwhelming consensus itself, but the fact that by interpretation 13 this has nothing to do with verifiability makes it even more confusing. --Aquillion (talk) 21:24, 16 June 2022 (UTC)
So we need the policy to explain that it must not be interpreted as 14(2<->8) or 13. I knew there was a reason I didn't understand it all the other times:) Selfstudier (talk) 21:29, 16 June 2022 (UTC)
I don't know what to say to that because, as I've said, I've almost never encountered anyone actually attempting to use anything that I would describe as remotely compatible with 13 or the 2 + 8 in actual discussions where text has implicit consensus. I would like to drill down into what the people who support those actually mean, ie. what they think discussions should look like when governed by this, because I feel like there's a fundamental disconnect. I posted a big question to Crossroads in that regard above (since he's the one who I've had the most past disputes with where I feel that this change in practice would drastically alter the outcome) but didn't get a response... which is fair enough because I can understand not wanting to rehash past disputes. But if the result of what people are pushing for here is "no-consensus RFCs default to removal; if someone challenges any inclusion, it must be immediately removed and stay out until / unless you can point to an unambiguous consensus to include, either via a near-unanimous / overwhelmingly lopsided discussion on talk or a closed RFC", then that would break a lot of our existing consensus-building mechanisms and would result in any controversial article either being much, much shorter and more empty, or an absolutely massive wave of RFCs for anything that is seriously disputed. It would be one thing for people to propose this as a change - I'd think it's a bad suggestion, but I'd at least understand what you're suggesting. For people to say they believe that this is how things work currently, though, is, to me, extremely worrisome because it suggests that there's still no real understanding of what such a seismic change would mean in practice (and from the comments above regarding the specific proposed change I'm not the only one who shares that worry). --Aquillion (talk) 21:41, 16 June 2022 (UTC)
Also, the other possibility, I guess, is that if the people who push for interpretations 13 or 2 + 8 actually believe that that reflects both the current practice and is what the current text means, then I guess to a certain extent the disagreement is illusionary because don't have to do anything; we can continue to default to the status quo or using implicit consensus to the extent that we currently do and nothing will actually change beyond occasional argument. But that doesn't seem to me to be possible because 13 or 2 + 8 so directly contradict that current practice - I can go over more examples of past disputes I've been involved in, maybe, and the people who push for those interpretations can say how they think those disputes should have gone, or ought to be going? --Aquillion (talk) 21:47, 16 June 2022 (UTC)
I don't know why you keep bringing me up, and I don't think disputes with me are of any particular relevance. In my experience, text found not to have a consensus ends up deleted, even without an RfC. As has been said previously, IMPLICITCONSENSUS goes away once a good-faith objection appears. Often it will be easy to re-establish a stronger consensus, and "near-unanimous" discussions are not needed to establish consensus. If they were, I'd turn the question around - why would we allow content to stay even if a majority - though not "overwhelmingly lopsided" - said it should go? Also as said earlier, we need to beware of making it too difficult to remove old, bad material that went unnoticed. As other editors stated: If the material is truly disputed by just one person, then it should be possible to find one or two other editors who support its inclusion. In that case, we'd be able to form something like a consensus to retain the material, and ONUS wouldn't really apply. And: Only one person did the original bold insertion. The policy doesn't say it will self-destruct after a deadline and adding a deadline would give a first mover advantage based on a vague ("years" or "longstanding") assertion which could trump multiple opponents afterward.Crossroads-talk-21:58, 16 June 2022 (UTC)
I spelled it out above, but in our past disputes on eg. Mermaids (charity) - I definitely think there has never been a clear consensus for the Training section. Do you believe it actually had explicit consensus? And, in your view, how would that be resolved, if I insisted otherwise? Would it require an RFC, and would it default to removal if that RFC reached no consensus? If you agree it lacked explicit consensus, what do you think the people involved should have done differently? My point isn't simply to dredge it back up but to try and get some understanding of how you feel WP:ONUS should work and apply to dispute resolution in controversial areas in practice. Rehashing specific discussions in depth can wait until / unless policy is actually changed or clarified here, but - I am using an example to try and understand what you think this interpretation actually means, in the sense of how it would apply to actual disputes. Because to me this is a drastic change that would overturn or throw into question numerous past status-quo decisions similar to that one, and would in practice lead to either article instability, controversial articles getting trimmed down to the bone, or an absolute avalanche of RFCs on every seriously disputed point. --Aquillion (talk) 22:10, 16 June 2022 (UTC)
If I recall correctly, as seen on the talk page, that section as it stands is the result of compromises, removals, and adjustments on the part of many different editors that are part of the process of consensus - which often is more complex than a simple keep/remove. I haven't reviewed that discussion for this, but removing it all after that agreement was reached would be throwing out that more explicit consensus on the part of one editor and hence inappropriate. As for RfCs, while I do believe that RfC closure as "no consensus" is best avoided since it sidesteps the whole point of an RfC which is to get a definite answer to an intractible dispute, if even after one there is such significant disagreement with merely including a matter, then yes, the encyclopedia is better off without it. I think this interpretation of ONUS is in practice what is usually done anyway, and would throw hardly anything into dispute again, as my experience in controversial topics is that usually most controversial text is the result of explicit consensuses on the talk page, or could have one easily developed in case of disruptive deletion. Crossroads-talk-03:52, 17 June 2022 (UTC)
That first part is interesting. I don't totally disagree because as I've said above I feel that the crux of WP:IMPLICITCONSENSUS is that the more people who have seen or, better yet, edited something without removing it, the stronger the implied consensus backing it; we use time only as a quick shorthand, obviously text doesn't gain implicit consensus just by lingering on an article with a single editor that nobody else has ever seen. But I still think that that's just implicit consensus and therefore would not count under either 13 or 2 + 8 proposals above. Part of this might be that we need to refine the concept of implicit consensus a bit better, or at least add more guidelines as to the difference between the hypothetical "clearly only one person has seen this ever, but it has been here for 12 years" text and "it's been here for three months but has been extensively edited by numerous editors on a high-traffic article without anyone objecting to its basic presence" text. By my reading, though, barring the use of implicit consensus to satisfy WP:ONUS would mean that the only thing that would satisfy it is a clear discussion or RFC about that specific text that definitely led to a consensus supporting it. Simply being discussed and failing to reach a consensus or being heavily-edited aren't enough - the first is WP:NOCON (leaving only whatever implicit consensus it had before in place) and the latter is just a strong implicit consensus. I would add a caveat that edits that clearly treat a disputed addition as controversial (eg. adding tags, but also edits that are solely attempts to tone it down or pare it back) do not really contribute to its implicit consensus - especially if, eg. someone removes an entire addition, gets reverted, then tries for a compromise that removes the most objectionable part, that doesn't mean they support it and per WP:NOTSILENCE can actually weaken or remove implicit consensus rather than strengthening it. It's important that any objection to a disputed new or no-consensus addition "breaks the seal" of implicit consensus and blocks it from achieving it afterwards, because otherwise you're discouraging people from compromising at all, ie. if an editor effectively says "fine, I don't want to get into an extended dispute on this, I think the whole thing should go but I'm going to at least remove the very worst part", that's not even implicit consensus, let alone explicit consensus. --Aquillion (talk) 05:00, 17 June 2022 (UTC)
@Aquillion, about "this list provides a detailed rationale and explanation for 13": No, it doesn't. There is no rationale given for 13 at all. The whole point of 13 is that it explains the existing wording in extreme, over-the-top detail. Some editors claim to have found it enlightening. Perhaps it offers a sort of "I really mean it – This means you!" moment for editors who had previously thought there was an implied exemption for their circumstances. But there is no rationale given for why 13 is the meaning. WhatamIdoing (talk) 07:13, 20 June 2022 (UTC)
I mean it goes on for several times as long as the others, with passionate wording that invokes arguments based on additional interpretations of other policies - like STATUSQUO and WP:CONSENSUS. Neither of those policies, obviously, can be changed or curtailed by discussions here; stuff on this page is only applicable to WP:V and to challenges related to V, per WP:CONLOCAL, and will not change how we assess consensus in any way. As it is, the wording reads to me as blatantly prejudicial - you don't ask people to choose between 14 sets of dry simple wording and a massive paragraph of passionate invocations. --Aquillion (talk) 07:52, 20 June 2022 (UTC)
Can we talk about this when you have a little more time to think it over? You've just called STATUSQUO a policy, when it's an essay; and if I could get you to click here to open the specific section of CONSENSUS, I think you'll find a very interesting and relevant note at the top of the section that provides direct and explicit information about what to do if that section diverges from other policies. As a result, your worries about CONLOCAL are probably misplaced, if not exactly backwards.
I'd have thought that my highly informal, over-passtionate wording, which was turned into 13, would have discouraged other editors from choosing it. Editors generally want a certain amount of formality in their policies. Perhaps the important point, though, is that there's a gap between "This sentence correctly explains the meaning" and "I think this sentence belongs in the policy". I'm not seeing much of the latter; are you? WhatamIdoing (talk) 22:38, 20 June 2022 (UTC)
There's yet another severe structural problem with trying to interpret a policy change out of this. Trying to think that one can interpret the results out of a local "what do you think this wording means?" into a policy change on a high-impact core policy is not correct. The policy is what what is written right now, with all of it's deliberate fuzziness. Any change would nee to get proposed as a specific proposed change. North8000 (talk) 00:14, 17 June 2022 (UTC)
I think that there is at least a lot of confusion over what implicit consensus is and how it works, which means that having a policy discussion that references whether to respect it but doesn't define it is mostly useless. At least by my reading, implicit consensus is any consensus at all that is not backed by a discussion that clearly reaches a consensus for the disputed text in question, or (in situations where there is any degree of reasonable doubt) an RFC; and, therefore, if ONUS does not respect implicit consensus, anyone who wants to restore any removed content must point to one of those two things first. I obviously don't think that's workable. The vast majority of our text has had either of those things. --Aquillion (talk) 05:06, 17 June 2022 (UTC)
I agree that WP:Consensus needs to better define what constitutes an implicit (silent) consensus, and how much weight to give it (to my mind, we should focus less on the amount of time between addition and challenge, and more on the number of intervening edits. That is a better indication that other editors have reviewed the original addition and are OK with it). However, that discussion really needs to happen at WP:Consensus, not here at WP:V.
All WP:V needs to include is a warning that Verifiability is only one of many factors to consider when determining whether to include or omit information… and those other factors can result in verifiable information being omitted. Blueboar (talk) 13:05, 17 June 2022 (UTC)
I agree with your last bit. Some people above are saying that ONUS has no relation to, or is not limited to, disputes over WP:V; at that point... why is it here? It's not part of our WP:CONSENSUS-building policy; it's in the wrong place for that. If ONUS just said that verifiability does not guarantee inclusion, and referenced other policies like WP:CONSENSUS to define how consensus is built (plus stuff like WP:NOT, that also have things to say about it), that eliminate the problem. Well, it would eliminate it here, at least, we might still have to work on CONSENSUS if people think that it is missing some vital part that is currently only in ONUS. Aquillion (talk) 18:03, 17 June 2022 (UTC)
To put my last post in short form, a "what do we think that the policy means?" discussion is not a basis for changing a policy. An RFC on a specific proposed change would be needed to do that. North8000 (talk) 13:17, 17 June 2022 (UTC)
What say we focus on clarifying them: let's talk about what they should say, and then draft an RFC to change them per Levivich above. Is that where we are at? Selfstudier (talk) 13:29, 17 June 2022 (UTC)
If we do have an RFC, I'd like to have an option along the lines of what Blueboar said above, eg. All WP:V needs to include is a warning that Verifiability is only one of many factors to consider when determining whether to include or omit information… and those other factors can result in verifiable information being omitted - that is to say, WP:ONUS should not make any sweeping statements about the consensus-building process, it should just indicate that it is necessary and then point people to WP:CONSENSUS. --Aquillion (talk) 18:03, 17 June 2022 (UTC)
+1. The ONUS sentence, "The onus to achieve consensus for inclusion is on those seeking to include disputed content", regardless of what it means or may be re-worded to say, really should be moved to WP:CONSENSUS (and clarified). The rest of WP:V#Verifiability does not guarantee inclusion should do what Blueboar said in Aquillion's quote above (and what the section heading says). That remaining part could probably be wordsmithed too. Levivich21:34, 17 June 2022 (UTC)
There already was a Village Pump RfC for a very similar proposal, and it was closed thusly: There is an overwhelming consensus against this proposed change, primarily due to concerns that it would undermine verifiability and efforts to remove poorly sourced information.
Tying implicit consensus to number of intervening edits instead of elapsed time per se doesn't really mitigate the issue. A lot of intervening edits can occur that are to, say, the lead of the article or some other portion, while those same editors don't notice the problem material. A lot of edits are copyedits, categories, AWB, and bots, and hence also aren't really much in the way of scrutiny. Crossroads-talk-01:27, 18 June 2022 (UTC)
That proposal failed because it proposed bad wording for a new onus section of wp:consensus; it's not really applicable to this idea. Levivich01:59, 18 June 2022 (UTC)
To put an exclamation point on the risks of overly categorical language in this section, there is at least one prolific AFDer who contends that ONUS means that deletion is the default outcome of AFDs, and thus a guideline that states "there is consensus that articles of type X are OK and should generally not be deleted" means that all articles not of type X should be deleted. I have, of course, pointed out that of all the things ONUS is not, it is definitely not a deletion or notability policy. But I would have to think that this sort of misunderstanding is probably not original with this user on this day. Imagine the damage such a misreading could do if it the language of this section were even more legalistic. -- Visviva (talk) 00:48, 20 June 2022 (UTC)
In that discussion, the only person voting "endorse" is probably the only person misreading anything. The "prolific AFDer" whose edit you link to has participated in a whopping 19 AFDs. The editor, in that discussion, does not contend that ONUS means that deletion is the default outcome of AFDs (their argument is more complicated than that). And you should probably ping editors when you use them as examples. Welcome back, BTW, glad the new admin inactivity requirements have spurred a return to editing :-) Levivich[block]02:40, 20 June 2022 (UTC)
It would have just spurred a desysopping if I'd had to spend any longer in the hospital, but I suppose we can leave that particular example of how policy change goes wrong alone. For my part, I would not wish to personalize a policy discussion in the manner you suggest or have it done to me, but ymmv. (I'll consider omitting diffs entirely in the future, although that hardly seems like an improvement.) I cheerfully withdraw the characterization, although I think anyone who has participated in more than a couple of AFD discussions (myself included) can fairly be described as "prolific" relative to the community as a whole. I do think I have accurately summarized the position in that comment; I am unclear what sort of "nuance" would lead to such an explicit inversion of NLIST if my summary is incorrect. In any event, I think my broader point here as to the risk of misinterpretation of this passage stands. As to me being "the only person voting 'endorse'" in that discussion, since I have been reliably informed (elsewhere on that very page) that consensus is now determined solely by strength of arguments rather than numbers, the fact that some of my fellow Wikipedians disagree with my correct interpretation of policy is, apparently, neither here nor there. What a time to be alive! -- Visviva (talk) 04:45, 20 June 2022 (UTC)
Okay, what about the other solution: removing the third sentence altogether? Or, at least, proposing its removal. By its own terms the onus would then be on those who want to preserve it to explain why we should "include" a policy statement when the community cannot come to a consensus regarding when it applies or what it means. - Butwhatdoiknow (talk) 03:41, 20 June 2022 (UTC)
Of course not. Policy text is not article text; there is no risk of spreading misinformation to the public, and they are the rules that underlie how editing works, so ONUS does not apply just like WP:V in general does not apply. In no way would this discussion be sufficient to remove any sentence (I assume you meant the fourth sentence). And changing so it applied to changes in general was rejected at the Village Pump RfC - obviously removing it has the same effect of changing the policy and the same reasons for rejection apply. Crossroads-talk-06:56, 20 June 2022 (UTC)
To borrow phrasing from North8000 below, removal would put a finger on the scale away from exclusion of controversial and dubious material, and that's bad. Crossroads-talk-01:29, 21 June 2022 (UTC)
@Kolya Butternut, in your vote, you wrote "IMPLICITCONSENSUS must be considered." Do you think that the sentence in question ("The onus to achieve consensus for inclusion is on those seeking to include disputed content") already/currently considers IMPLICITCONSENSUS, or do you think that its omission of IMPLICITCONSENSUS is something we should change? WhatamIdoing (talk) 04:08, 3 July 2022 (UTC)
What do folks think of my interpretation of ONUS: The onus to achieve consensus for changes to consensus content is on those seeking the change.? My point is really that ONUS is just reiterating part of CONSENSUS, so as Levivich said, WP:ONUS should be moved to WP:CONSENSUS (and clarified). Kolya Butternut (talk) 01:17, 9 July 2022 (UTC)
I think that's a substantial expansion. WP:V says you can't add information unless it's sourced; ONUS clarifies that having a bit of information sourced is necessary but not sufficient, and if you can't get both sourcing and consensus for the material you want, then out it goes.
You are proposing that even copyediting or which color to make a sports team's navbox ("change") is controlled by WP:V. It's not appropriate for WP:V to reach into questions unrelated to sourcing.
Your wording will also increase the number of disagreements about who is "seeking the change". We have altogether too many disputes already that involve all editors declaring that "No, you have to get consensus because yours is 'the change'. My edit is the 'non-change' that gets to stay in unless you can prove to me in writing that everyone says my change is bad." That my-version-is-the-true-status-quo is one of the undesirable problems that ONUS is meant to eliminate within the narrow context of including/excluding information. WhatamIdoing (talk) 03:31, 9 July 2022 (UTC)
This is not a proposal for new wording. You wrote: You are proposing that even copyediting or which color to make a sports team's navbox ("change") is controlled by WP:V. No, the whole concept of "onus" is not controlled by V, that's the problem with the sentence's location there, I'm just explaining what the concept is. Kolya Butternut (talk) 08:59, 9 July 2022 (UTC)
Changing it from requiring consensus to include, to requiring consensus to change, is massive difference in meaning and major tilt of the scale. This seems to be the same proposal that was overwhelmingly rejected at the Village Pump RfC. Crossroads-talk-02:48, 10 July 2022 (UTC)
IMHO the meaning of "The onus to achieve consensus for inclusion is on those seeking to include disputed content"
IMHO: The meaning of "The onus to achieve consensus for inclusion is on those seeking to include disputed content" must be understood in the context of the Wikipedia system. The "Wikipedia system" includes making multi-variable decisions (those fancy words describe how normal human decision making and wp:consensus works) based on considering and weighting numerous factors, policies and guidelines, this being (merely) one of them. (Examples of the many others include: how long the material has been in the article, how much tacit review/consent it has accumulated, and "last stable version" considerations) The meaning of this sentence is to put a finger on that decision scale towards exclusion when the the material is disputed. A few special-case derivatives/ examples of this are:
If the material is disputed and all other considerations are equal, the material stays out
If the material is weakly disputed, but other factors add up very strongly towards inclusion, it will be included
Like most Wikipedia policies, by necessity, the sentence is not written in a very specific, categorical way. Any attempt to do so would be something that preclude also taking into consideration other policies, guidelines and factors which is not how Wikipedia operates and would also create conflicts with other policies, guidelines and considerations.
+1. It looks as if it's broken but it ain't really broke in practice and it's designed to be a bit woolly so as to cover all sorts of possibilities. Selfstudier (talk) 17:09, 20 June 2022 (UTC)
Right. But if our policy documents did their job, they would read like what N8k just wrote here, which is a whole lot clearer than anything written in any of our policies. Let's actually document this, which the current sentence does a poor job of doing. Levivich[block]17:36, 20 June 2022 (UTC)
No objection here, might not have the patience for the discussion that tries to settle the precise words that will be equivalent to what N8k just wrote:) Selfstudier (talk) 17:46, 20 June 2022 (UTC)
Describing how the described aspect of the Wikipedia system actually works (including how it interacts policies, guidelines and considerations) could be called "out of place" here but besides solving the issue it could be the seed for something that would have astoundingly huge benefits. A tiny place for "make no small plans" :-) North8000 (talk) 19:22, 20 June 2022 (UTC)
What North wrote was the easy part. It can be shortened to "Consensus is not unanimity. If you have a consensus, then do whatever you all agreed to do".
The question is what to do when the objections to the material (however they happened to be measured by you/in that specific dispute) are exactly equal to the support for that same material. Imagine 100 editors with 100 medium-strength arguments in favor, against an exactly equal 100 editors with 100 exactly equal medium-strength arguments against. There is no consensus; we have not achieved anything like an agreement. Imagine for simplicity that it is not possible to compromise on some sort of halfway-in-and-halfway-out position: the disputed material (e.g., an image, a name, a link, a source) is either included or excluded.
In those rare instances in which you cannot form a consensus to include or exclude, what do you do? The non-random options are:
Default to an outcome (inclusion or exclusion)
Default to a precedent (the oldest or newest edit)
Default to a person/role (e.g., if one of the principal disputants is an admin, or whoever has made the most edits)
<preach> The real problem with the WP:ONUS sentence ("The onus to achieve consensus for inclusion is on those seeking to include disputed content.") is that it's written entirely wrong, and until we completely rewrite it--and completely change how we approach the problem--we will always have this underlying fundamental disagreement.
What's wrong with the ONUS sentence is that it frames the issue in a battleground way, by dividing editors into two camps: those "seeking to include disputed content", and everyone else (presumably, those opposing inclusion of the disputed content). To further encourage a battle between the two sides, the sentence gives one side the "onus" of winning inclusion--or in other words, it divides us into two sides and tells us which side is playing offense and which side is playing defense. Talk about encouraging WP:BATTLEGROUND and WP:GAME behavior!
In content disputes, nobody really "supports" or "opposes" inclusion. Everyone supports inclusion of what's WP:DUE, and opposes inclusion of what's undue. And we all read and interpret sources to determine what we think is due for inclusion. Of course, many people disagree about what's due, but to frame that as a disagreement between those supporting inclusion and others is to completely oversimplify what actually happens in real content disputes: people have many different and overlapping interpretations of sources that lead them to different and overlapping conclusions about what's due for inclusion. They discuss their interpretations until they come up with something that overlaps enough people's interpretations that everyone agrees it's more or less OK, and if successful, we call that "consensus".
The consensus-building process is not a battle between two sides, it's a collaborative effort with everyone working together towards the same goal. The fact that sometimes we have votes on things where people "support" and "oppose" is a granular detail, just a part of a larger process, and to focus just on that little detail by having rules about which side has the "onus" is to completely miss the forest for the trees.
Whenever I read the ONUS sentence, my initial reaction is always: "Who the hell cares where the onus lies?" How does it help us resolve a dispute to identify who has the onus of resolving the dispute, but not identify how the dispute gets resolved? Wikipedia is not a legal system and is not trying to be one, so why do we use these legalistic analogies like burden of proof? I suggest we don't care where the onus lies, we care about what to do with disputed content while we come to consensus about what's WP:DUE. That is what WP:ONUS should tell us.
And we already know half the answer: if it's newly-added content, it should stay out until there's consensus about whether/how to include it. The hard part is for content that's already been in the article for a while (however one defines "a while"), and when such "longstanding" content is disputed, what do we do with it while we figure out what's due for inclusion? I don't know the answer to that question, and I think it's a "multifactorial" analysis like what N8k wrote above, balancing multiple considerations on a case-by-case basis. We could clear up all this confusion about the third sentence of WP:ONUS by rewriting it so it answers the question of what to do with new content, and longstanding content, while inclusion of the content is being discussed. </preach> Levivich[block]18:18, 20 June 2022 (UTC)
Ideally, yes. However, sometimes it reallt does come down to a zero-sum 'include or exclude' outcome, with multiple experienced editors on each side. Crossroads-talk-01:36, 21 June 2022 (UTC)
Levivich, I don't agree that there are no battles between two sides or that it's always a collaborative effort with the same goal. Sometimes people just want to stick their thing in, and they don't really care about anything else. Consider the newbie who seems to be posting a WP:SELFCITE to a source we'd never use. Consider the anti-vaxxers or other real-world (or at least Twitterverse-based) POV pushers, who want Wikipedia to give more airtime to their views. Consider the paid promoters, whose goal is to get their client mentioned in Wikipedia no matter what. "Everyone" isn't working towards the same goal unless your "everyone" excludes a lot of people.
Also: We might care a little bit about what to do with content while we come to consensus, but what about those rare times when we don't ever come to consensus? Shouldn't ONUS care more about these years-long situations than about a brief situation? BRD tells you to let the other guy win during the discussion, and if you don't like that, QUO tells you to leave it at some old version during discussion, and EW tells you definitely to stop edit warring over it, but most discussions only take about a week to resolve. Quit worrying about the relatively brief "during the discussion" part. Worry about what happens when it is utterly and completely impossible to develop a consensus. What should policies tell you about that?
And what about those situations in which an editor refuses to engage in the discussion? About a dozen years ago, I encountered an editor who would add something to a hotly disputed article and refuse to discuss it – unless and until someone removed it. Every removal produced a revert and a polite comment. If we didn't remove it, the editor refused to engage in the discussion. If we were to set the rules as "No removal unless discussion supports removal", then it would be in the best interests of the person who wants the material included to have the consensus-oriented discussion fail. I crammed that in the article, you can't get it out with consensus, and I sabotaged the discussion, so it'll never come to consensus. Voilà, there is no consensus to include that particular disputed material, but you still can't remove it.
I would be interested in hearing what you think should happen if:
I add well-sourced but possibly inappropriate content (and it's been there a long time).
You want to remove it (for very good reasons).
We have a long discussion followed by an RFC that ends with the words "There is no consensus either way. Editors were not able to agree on whether to include this or to remove it. Both numbers and reasons are closely balanced. There is no obvious reason to believe that further discussion will produce a different result."
If you just added it, and someone objects, they should revert it, and nobody should reinstate it unless and until there is consensus to do so. If the subsequent RFC ends in "no consensus", it stays out.
If it's been there a long time, whether it should stay in during the discussion would be determined by a multifactorial analysis that looks at factors like: is this a BLP, is it sensitive/controversial content, how well sourced is it, how strong are the objections, how long has it been there, and how many edits since it was added, among other factors. As N8k put it above, the factors may weigh towards inclusion (e.g., if it's been there for 10 years over 1,000 edits, it's not a BLP, it's not controversial, and only one editor is objecting) or exclusion (e.g., it's been there for 1 week over 5 edits, it's controversial, it's a BLP, only one editor wants to include it), depending on the specific circumstances; that's a case-by-case. At the end of the RFC, if there is no consensus on whether to include or exclude, it should be excluded.
sometimes it reallt does come down to a zero-sum 'include or exclude' outcome, with multiple experienced editors on each side yes, and those times are examples of "trees" in the "forest" that is the consensus-building process. By focusing on those "trees", ONUS misses the "forest". Levivich[block]17:01, 21 June 2022 (UTC)
I see no real issue with the wording here, the meaning to me being obvious in that until there is an affirmative consensus to include material it stays out as disputed and it settles what "no consensus" should default to (exclusion). I think the better thing would be to adjust NOCON rather than attempt to adjust ONUS. NOCON defaulting to status quo ante should only be applied when there is an existing affirmative consensus, and no implicit consensus through silence or time is not that. A discussion where one can show a consensus, not necessarily an RFC, would be necessary for a no consensus to default to content being retained. nableezy - 18:41, 20 June 2022 (UTC)
I'm with you on the first part of the argument, but I am a little confused by your conclusion. The kinds of battleground-type disputes where ONUS would apply at all are, as you note, not at all representative of what we are doing, or trying to do, on the wiki. So it's hard to see what it's doing in WP:V, which does apply to pretty much everything we do in article space. Disputes that get to the point of ONUS wikilawyering are so weird and warped in different ways that they pretty much all give truth to the hard cases make bad law maxim. Core policies shouldn't be tailored to edge cases. ONUS, as I read it, is trying to give a rough rule of thumb for those edge cases -- but it can't really do the job, because adding another Rule to the pile isn't likely to help anyone in finding a solution when the situation is already rife with wikilawyering. -- Visviva (talk) 19:13, 20 June 2022 (UTC)
I think it ended up in WP:V for two reasons: The first is that we needed a place to stick it, and WP:V is "a place". Not every policy statement ends up in the most logical place, and this one could have gone in several places.
Two, there's a tendency to think that everything that can be well-sourced belongs in Wikipedia. I can find many very good sources about the color of Queen Elizabeth's hat at 12:00 UTC on the first Thursday of this month, but that doesn't mean that any Wikipedia articles need a sentence that says "At 12:00 UTC on the first Thursday of June 2022, Queen Elizabeth wore a powder blue hat", right? I know that; you know that; some editors are ...maybe a little shaky on that point.
The fact is that we're so used to removing badly sourced content that we kind of forget that there are other policies, so when we encounter a situation like that, we tend to think of WP:V before WP:BALASP or WP:NOT#Wikipedia is not an indiscriminate collection of information. So our shaky editor is saying "But I sourced it beautifully, and it fully complies with WP:V, so that means it can stay!", and we point to the section of WP:V where this sentence is: ===Verifiability does not guarantee inclusion===. It's not enough to perfectly verify that she wore a powder blue hat at that exact minute; also, if people are disputing your perfectly verified claims, then you have to convince them to leave your perfectly verified sentence in the article. They do not have to convince you to agree to removal.
We could have taken other approaches ("If you have a dispute about whether a bit of perfectly verified content should remain in an article, please see the following six pages:"), but editors ended up sticking the sentence here. WhatamIdoing (talk) 23:23, 20 June 2022 (UTC)
Are the editors who who stuck this sentence here the same ones who gave us wp:NOCON? I ask because I'm still having trouble seeing how that policy doesn't conflict with this one. - Butwhatdoiknow (talk) 05:15, 21 June 2022 (UTC)
No. I wrote NOCON, except for the sentence that conflicts with ONUS. The conflicting sentence is one that I (we) discussed repeatedly, at both WT:V and WT:CON, because it never felt quite right. We were thinking about it as a kind of jumping-off point, to be tried out and refined later, etc., except that we haven't been able to make progress on it. "What we usually do" in that situation is complicated. Also, what we usually do in some instances might not be best for articles.
The other, more 'structural', problem with NOCON is that it's meant to be a convenient pointer to all the other, actually-controlling policies, but instead some editors think that NOCON is in charge of everything, and all the other policies have to change to align with it, which is backwards. It'd probably be safer in a separate essay. WhatamIdoing (talk) 15:36, 21 June 2022 (UTC)
It might be helpful to have a little more detail about the history.
I started NOCON in November 2011. See this comment and the one by Kotniski that follows it for the most immediately relevant precipitating comment, but if you want to understand the motivation, you really have to read that entire section plus multiple discussions on that and other pages earlier that year.
Ring Cinema added the QUO-oriented sentence in January 2012. I refused to add it myself because I didn't think it was correct, but I didn't object to attempting to cover this subject.
Something that might be useful to know is that Ring pushed hard, for years, to have QUO be enshrined in that policy as the One True™ Way. See, e.g., March 2012 to remove the idea that QUO didn't always happen; March 2012 to remove the anti-QUO BLP rules; same thing in June 2012; claiming that changes to the original version of an article without consensus are anti-policy in October 2012; July 2013 to remove the idea that QUO was a tiebreaker instead of a dominant principle, with edit summaries like "sorry, no consensus means no change -- it's in this policy", and so forth.
For clarity, this isn't the only editor who holds this view. See, e.g., this edit in 2015, or this one in 2020, which says even COPYVIO isn't worth mentioning as an exception to the sacred principle of QUO. Some people are dedicated to the principle that QUO is the law of the wiki. (Tangent: I wonder if editors who hold that view are disproportionately likely to end up blocked, like Ring was, for edit warring.) WhatamIdoing (talk) 16:50, 21 June 2022 (UTC)
And to add: Of course we default to QUO, except:
BLP
CHALLENGE
COPYVIO
ONUS (at least theoretically)
No agreement on which version is m:The Wrong Version and which one is the QUO version
One side's more willing to edit-war than the other
Editors actually read QUO, and figured out that it says "During a dispute discussion, you should not revert", and doesn't say anything about what should happen after the discussion is finished.
Editors actually read QUO, and figured out that it says "During a dispute discussion, you should not revert" and does not actually say anything even remotely like "you have a duty to force the status quo version onto the page".
Editors develop a nuanced understanding of consensus ("We agree on this much, even if we don't agree on everything")
OK… more background history… back in the day (before we had ONUS), when there was a dispute over whether an article should mention factoid X or not, one side would say: “you need consensus to include this”, while the other side would say: “no, you need consensus to exclude it”. We needed something to break this stalemate. We figured that it was usually easier to achieve a consensus to include than to exclude (proving the positive rather than the negative), so we placed the “onus” to achieve consensus on those who wished to include. Blueboar (talk) 19:29, 20 June 2022 (UTC)
I'm going to guess that it ended up where it is because it relates to the sentences preceding it which I'll bet were an attempt to offset the oft-invoked false urban legend.....the verifiability is a reason for inclusion rather than just a requirement for inclusion.North8000 (talk) 19:35, 21 June 2022 (UTC)
Yes, but I mean that interpretation #13 seems to make BLPRESTORE irrelevant if ONUS works the same way for all articles as BLPRESTORE works for BLPs. Kolya Butternut (talk) 15:28, 28 July 2022 (UTC)
Of course, #13 is one of what appear to be an infinite number of meanings that ONUS may have. That said, yes, if THE meaning is #13 then BLPRESTORE is redundant. - Butwhatdoiknow (talk) 01:56, 29 July 2022 (UTC)
Just a note here to say that BLPRESTORE would be a "necessary redundancy"; sometimes, you have to say important things multiple times or in multiple places, so that more people will be aware of the rule. WhatamIdoing (talk) 20:23, 29 July 2022 (UTC)
The meaning of the CURRENT text at ONUS (conclusion)
1. Some editors believe its meaning is opaque. Some of these editors believe this is a flaw, others that it is a feature.
2. Other editors believe its meaning is clear. These editors do not agree on what that clear meaning is.
The meaning of the CURRENT text at ONUS (additional discussion)
My assertion is that it is clear in the context of a Wikipedia system that is unclear because it is complicated and not really described. :-) North8000 (talk) 16:45, 29 June 2022 (UTC)
But easier to fix than it sounds if we write the "grand unification statement" in big letters somewhere. North8000 (talk) 16:49, 29 June 2022 (UTC)
Great! Please propose a first draft of such a statement (we can worry about where to put it after a consensus version emerges). Butwhatdoiknow (talk) 19:57, 29 June 2022 (UTC)
There are folks who like the two apparently conflicting statements. This allows picking which serves their agenda of the moment. North8000 (talk) 22:40, 8 July 2022 (UTC)
I think we should resolve the disagreement between them, but it's also important to recognize that there's some value in policies that highlight conflicting requirements, since we do actually have conflicting requirements when editing. It's absolutely essential that verifiability alone not be sufficient for inclusion per WP:NOTINDISCRIMINATE. It's also policy that all changes, in principle, require consensus, which covers additions and sometimes removals. But it's also important to maintain article stability; and there are points at which removal itself is the more drastic change and therefore requires consensus. Figuring out where to draw the line is complicated. See some of the examples I referenced and the replies for what I think is an informative real-world cases; the fact is that most consensus is reached through editing - our articles are not built on explicit consensus and making it a hard requirement for every sentence on every controversial article is not feasible. So having policies that weigh different needs against each other is useful. I think the problem is that WP:NOCON and especiallyWP:ONUS try to be too strident. Policies that have some discretion to them and which highlight our ultimate goals rather than serving as bludgeons for people to hammer each other with during disputes are good because they push people to actually discuss and consider the specific context they're working with (as well as to seek compromises.) Policies that try to solve problems unilaterally from above by declaring an outcome by fiat don't usually work - even WP:BLP, probably the closest we have, is fairly cautiously worded at points, with stuff like strongly consider and the like. If you look at the arguments people make for sweeping versions of ONUS you can see some of this - there's an intense focus on theoretical bad actors and not enough focus on how this actually works for good-faith disputes in controversial articles. The fact is that any policy can be misused by bad actors; but even when editing in good faith, people aggressively removing things they disagree with is just as much of a risk as people aggressively adding stuff that slants articles. --Aquillion (talk) 12:59, 9 July 2022 (UTC)
Wikipedia has always had disputes between those who say “You need consensus to include that” and those who say “No, YOU need consensus to exclude it”. ONUS was an attempt to resolve these disputes. Blueboar (talk) 14:47, 9 July 2022 (UTC)
According to my post a few lines up, taking it within the context of the wikipedia system for making decisions, there is no conflict, there is only that these two provisions sometimes influence decisions in opposite directions. And if that reality isn't acknowledged, and one takes each as a stand-alone rule, there is no way to resolve what would otherwise be a "conflict" other than deletion of one of them. North8000 (talk) 19:13, 9 July 2022 (UTC)
I have think the phrase "and anyone strongly objects," in survey option 13 is a real problem. In Britain at the moment the War in Ukrain and the resulting energy crises have highlighted a political NIMBY problem. The governing Tory party have lots of constituencies in rural areas, to placate their voters the rules for new on shore wind turbines state that if just one person objects to building a turbine it will not be built (hence the emphasis on much more expensive off shore wind turbine farms). It seems to me that this phrase has similar implications as the "anyone" does not even have to base their objection on a rational argument based on Wikipedia policy just on "I don't like it". — PBS (talk) 17:44, 2 July 2022 (UTC)
I have not seen a discussion here on how WP:3RR fits in to what is being discussed would someone please explin how it does. For example one person objects to something which is sourced stating the source is not good enough. An editor with this on their watch list reverts and says take it to the talkp page WP:BRD, first editor reverts with no talk page discussion, a third editor reverts this second deletion. Knowing (s)he is loosing the revert war the initial editor takes it to the talk page saying (s)he "strongly objects" to the content, so must be removed while it is discussed? Should third editor self revert his or her revert? -- PBS (talk) 17:44, 2 July 2022 (UTC)
@PBS, thanks for these comments. I have the following thoughts, which might answer parts of your questions.
"Anyone strongly objects": Well, if nobody objects, then "Alice Adder" doesn't have to provide written proof that there's an active, positive consensus to include the information, right? It's only when "Rob Remover" not just objects to the addition, but objects strongly enough to really stick with the dispute, that we reach the point of invoking ONUS at all.
3RR: This is still the policy. Don't edit war, full stop. If nobody is edit warring to restore the disputed information, then nobody will ever be able to reach the point of 3RR anyway.
ONUS isn't really about what happens "while it is discussed". I suppose you could apply it during a discussion, but it's more like a tie-breaker rule about what to do with unclear discussion results. As a result, removers ultimately will need to "base their objection on a rational argument based on Wikipedia policy", because if they don't, the resulting discussion is likely to end up with written proof of consensus for inclusion.
ONUS-based scenarios usually look like this:
Someone adds sourced content.
I remove it (e.g., because I think it's off topic).
The adder says: "This has to be kept! Policy says I get to keep anything with a source! Besides, there's a Silent majority in favor of keeping it."
I say: "The policy says you have to prove, in writing, on wiki, that there's a consensus to keep it. If you get proof of consensus to include this, then of course we'll keep it."
This rule sidesteps arguments about:
the content's age ("That so-called hoax is 10 years old. It must be kept until there is written proof of consensus to have accurate articles containing only verifiable information!"),
spammers and UPEs making extra work for volunteers ("You have to prove to me that this sourced trivia about my client is too trivial to include! I expect an RFC with at least 45 participants, or I won't agree that there is consensus!"),
POV pushers wearing down editors over time ("You have to round up a bunch of editors for the fourth time this year and prove that everyone agrees that these 16 paragraphs about my special worldview are far too much for this article. All of my content must stay in until you prove there's a consensus against me!"), and
badly sourced content ("Just because you don't think this is well-sourced doesn't mean that you get to revert it. How about you spend the next three hours trying to find better sources all by yourself, or try to get other, extremely busy editors to come over to this talk page and very patiently explain to me why my content isn't perfect, while I go screw up edit another dozen articles?").
Of course, like Wikipedia:What editors mean when they say you have to follow BRD, if ONUS is applied early in the process, it can result in good content being temporarily removed from an article. One would expect in such situations that the good content would find a consensus in favor of restoring it. But the point of ONUS isn't really about what to do during the discussion; it's largely about what to do after the discussion, when you know that there is no consensus to include the disputed content, and especially when you know perfectly well that there is an active consensus against including it, but the would-be content adder is wrongly claiming that they won, or that there's no consensus, and that the alleged absence of consensus to include the content means that the content must be kept. WhatamIdoing (talk) 03:58, 3 July 2022 (UTC)
You did not really address my point User:WhatamIdoing. Instead you put togeter an alternative scenario. I put together a simple scenario and asked about that specifec instance. I deliberately kept it to 3 editors, because often dicussions over conteent involve very few active editors. I am interested to know how editors other than myself think that this specific senario ought to play out. For example who of the 3 participants should remove the content under discussion? If removing the content involves one editor breaching 3RR must one of the other editors regard less of their wish to keep the content remove it if they wish to discuss its retention on the talk page? To work around 3RR is the person who wants the content removed allowed to solicit removal in another forum or is that a form of canvassing? -- PBS (talk) 09:38, 3 July 2022 (UTC)
The phrase "work around 3RR" gives me pause. Just don't edit war (or canvass), there are other means of dispute resolution, typically the solution is to draw in more editors, the obvious method being an RFC. Selfstudier (talk) 09:43, 3 July 2022 (UTC)
User:Selfstudier In an ideal Wiki world one follows dispute resolution and it garners lots of intersted informed editors and they all agree what is best. My experiance is in lots of cases one is lucky to get more than a couple of others involved and their input may or may not be based on policy. However at the start of the process under survey option 13 who removes the content if the person proposing removal can not do so without breaching 3RR? -- PBS (talk) 10:06, 3 July 2022 (UTC)
Perhaps I didn't explain myself clearly, don't edit war. If one breaches 3RR one is edit warring, in fact, to have reached the point of breaching 3RR, one must have already reverted 3 times? Which is still edit warring, even if not sanctionable. Often topic areas have portals, neutrally inviting users at those portals is a step before an RFC, if the argument is that the material is not NPOV, there is the neutrality noticeboard, keeping to the principle of the more eyes the better. Selfstudier (talk) 10:17, 3 July 2022 (UTC)
Last year I was involved in a long running dispute over two different wordings of some the lead of an articlce. Under survey option 13 if all of the lead is contested is it acceptable to leave an article without a lead while the issue is resolved? -- PBS (talk) 10:06, 3 July 2022 (UTC)
"While the issue is resolved" should be wp:QUO, not wp:ONUS (or wp:NOT). Even if it is ONUS, as Selfstudier says, no edit warring. Tag the disputed text and discuss away on the talk page. - Butwhatdoiknow (talk) 16:26, 3 July 2022 (UTC)
Your scenario is:
A adds sourced content (possibly years ago).
B reverts/removes it because the source is weak.
C re-reverts/re-adds it and claims BRD, even though (a) it's not really a true BRD situation and (b) BRD says the person best adhering to its principles is the person who starts the discussion, not the one who says it's the other editor's job to start the discussion.
B re-removes it.
D re-re-reverts/re-adds it with the weak source.
and you want to know what ONUS says? ONUS says that B and D need to demonstrate that there is a positive consensus for including this information. ONUS does not establish a timeline for this; as far as ONUS is concerned, the material could be re-re-removed immediately or it could wait until the editors have finished six months of dispute resolution. There is no deadline in ONUS.
EW, on the other hand, says that if all of these edits happened during a 24-hour span, then both B and D are already at risk of being blocked for edit warring, and C might be, if A's content was added very recently and/or if A is the same editor as C. WhatamIdoing (talk) 21:02, 3 July 2022 (UTC)
This page is WP:V. It does not govern how we resolve disputes or what we do after disputes in any way; the only actual significant part of ONUS is that verifiability, alone, does not guarantee inclusion. The rest is merely explanatory in nature; no one should be consulting WP:V for details on how to resolve a dispute outside of the specific details of how verifiability is assessed and what it means. A single brief aside in the paragraph reminding editors that verifiability alone is not sufficient obviously does not have any great significance to our dispute-resolution process, let alone completely overriding large swaths of it, as you seem to be implying here. --Aquillion (talk) 12:31, 9 July 2022 (UTC)
Aquillion, I left you a note on your talk page. I do not agree that ONUS is disputed. The dispute involves various misunderstandings of it that result from editors extending its meaning beyond the four corners of its text. The disputed tag gives the impression that Verification may be sufficient to establish WEIGHT, for which I've seen no support as policy, but which we see in innumerable articles' text and the associated edit wars. SPECIFICOtalk12:54, 9 July 2022 (UTC)
Since someone added a dispute tag to NOCON, and since the dispute substantially relates to whether NOCON or ONUS takes priority, and since discussions on both pages have plainly failed to establish a clear consensus so far, I've tagged the current version of ONUS as disputed for parity reasons - we shouldn't give the impression that one of the two has more consensus than the other when that is clearly not the case (especially since I think it's clear that NOCON is the one that reflects actual current practice.) I'd be fine with removing both tags if people feel that it's inappropriate to leave a disputed tag on two longstanding policy pages, but I'm not fine with a situation where someone who consults or is directed to both is left with the impression that only NOCON is disputed and that ONUS is comparatively uncontroversial; so please do not remove one without the other unless you can point to an unambiguous consensus resolving the conflict in a one-directional way. --Aquillion (talk) 12:26, 9 July 2022 (UTC)
It's probably better to remove both tags. At this point, the WP:ONUS is on editors to get consensus that such a tag should be included this late in the whole above conversation (i.e., consensus that there is a true dispute between the two). It's one thing if a tag were placed early on, but a tag at this point definitely can come across as circumventing the lack of consensus in discussion. KoA (talk) 17:56, 9 July 2022 (UTC)
"tags in order to produce a discussion" I always thought they were in order to let readers know that the text was disputed. What is the basis for your statement? - Butwhatdoiknow (talk) 23:16, 9 July 2022 (UTC)
Thank you. However, "no tag without starting discussion" does not necessarily also mean "no tag as soon as a discussion is up and running." A "disputed" tag is a useful tool to reduce edit warring during a discussion (see, for example, wp:QUO). - Butwhatdoiknow (talk) 00:33, 10 July 2022 (UTC)
Selfstudier, as WhatamIdoing said elsewhere (discussing the same topic): "it is in everyone's best interest for editors to be aware of these discussions, and not just the handful of people who watch these pages closely." (I look forward to your response to this post and my July 10 post (above).) - Butwhatdoiknow (talk) 16:09, 26 July 2022 (UTC)
This page is on my watchlist, no need to keep pinging me. I already answered this, apologies if you don't like the answer. If you want to tag, I'm not stopping you? Selfstudier (talk) 16:37, 26 July 2022 (UTC)
I've gone ahead and removed the tag here. I haven't been involved here (though watching till now), but if someone feels strongly about NOCON and removing that tag or leaving it, that can be handled over at that policy since we shouldn't be dictating what happens at other policy pages here. KoA (talk) 20:26, 9 July 2022 (UTC)
I read BLUDGEON as discouraging pushing the same point over and over. If you read my pings you will find that they seek more information so I can understand others' positions. And it appears that my ping to Selfstudier has resulted in clearing up a misunderstanding (see above). Meanwhile, accusing me of BLUDGEON behavior is not a substantive response to my ping to you. Do you have one? - Butwhatdoiknow (talk) 17:05, 26 July 2022 (UTC)
Depends on the topic
The topic decides on which takes precedence, WP:NOCON, WP:ONUS, WP:BRD, WP:BOLD, etc. If the topic is emotionally charged & heavily watched 'or' barely given any attention at all? It will be handled differently. Any attempts to provide the same solution to articles of different topics? would be messy. GoodDay (talk) 19:30, 10 July 2022 (UTC)
Part of the problem is that this spills out into a huge range of other policies, including the ones you mentioned and, more generally, WP:CONSENSUS and our entire dispute-resolution process. Some of the arguments above essentially say that WP:ONUS somehow allows editors to completely ignore the well-established concepts of implicit consensus or consensus-via-editing. It's too much weight to put on a small aside paragraph in WP:V - ONUS is a useful reminder that verifiability alone is not sufficient to include something, nothing else. If we're going to change our core concept of consensus to only count explicit consensus, or revise our dispute-resolution process to default to removal, that needs to be done elsewhere and have more discussion there rather than fixating on a single sentence in ONUS. --Aquillion (talk) 02:31, 11 July 2022 (UTC)
I think the ambiguity that exists in ONUS allows for implicit consensus to be taken into consideration. It's possible that "those seeking to include disputed content" can make the case that implicit consensus exists based on the article's history, giving some leverage in a discussion where consensus is evenly split. Of course, this would still need to be shown, and the onus to do so is on those in support of inclusion. It's also possible for implicit consensus to be a weak argument, again dependent upon the article's history, level of activity/visibility, etc. ONUS, to me, leaves that argument on the table. --GoneIn60 (talk) 17:20, 11 July 2022 (UTC)
I don't believe that ONUS is ambiguous. I think that's part of the problem with the conflict: We're trying to pretend that there is ambiguity so that we don't have to deal with the sharp, clear words in ONUS. ONUS says:
If I add it +
You don't want it in the article, then
out it goes, unless and until we have proof of consensus for inclusion (which it would be my job to seek, not yours).
We need to stop pretending that this is unclear. There is no secret "well, if she added it months/years ago, then it wasn't ever 'added' and ONUS doesn't apply" clause in here.
BTW, this might be part of the problem. @GoneIn60 writes that "implicit consensus exists based on the article's history". Actually: No. That's SILENCE, not IMPLICITCONSENSUS. Implicit consensus is when I add something, and you improve it. You "implicitly" consent to the material being in the article when you choose to improve the material instead of removing it. If you remove it, you are (a) proving that there is no implicit consensus for the material I added and (b) speaking out against any perceived silent consensus. WhatamIdoing (talk) 23:26, 11 July 2022 (UTC)
If the content in question was improved upon, then that action would be a part of the "article history", would it not? But sure, that's a helpful clarification to make, and perhaps EDITCON should be more specific in that regard. While I think some ambiguity exists in ONUS (in a good way), I wouldn't necessarily call ONUS ambiguous. There are straight-forward examples like the ones you've provided that show when ONUS is most clear. It is less so in other situations. For example:
Alice boldly adds something.
Bob modifies Alice's addition.
Carol later removes it entirely.
Only Alice and Carol participate in the challenge discussion.
Does Carol automatically have the upper hand per ONUS? Does the implicit consensus count for anything if it pans out to a 1-1 draw? Maybe the default is still exclusion, but that seems less clear to me in a situation like that. --GoneIn60 (talk) 03:50, 12 July 2022 (UTC)
Was the content in question sourced? If so, it can't be a CHALLENGE. (It can only be a "dispute".)
Was the content in question both poorly sourced and about a BLP? If so, then BLP overrides every other consideration, and results in removal.
Assuming that neither of those a relevant, we proceed to ONUS. Bob's implicit support for including the material should be counted when determining whether consensus exists. You might see a discussion summary (let's pretend we have one for convenience) that says "Well, Alice and Bob are clearly support its inclusion in some form, but Alice gave no reasons except that she likes it, and Carol says it's a violation of NOT, and I'd say the fact that two editors like it is basically equal to Carol's one strong reason, so it's a 50–50 draw with no consensus". In that case:
ONUS says to remove it.
BRD says not to restore it.
EW says not to restore it and to seek more participants for further discussion.
EPTALK says to consider BRD or dispute resolution.
QUO says to remove it if it's new but keep it if it's old.
NOCON says that, in such situations, we "commonly" remove it if it's new but keep it if it's old.
The process shouldn't be consider the discussion in isolation, declare it a no-consensus draw, and then add on other considerations, like edits made by editors who didn't participate. When you are determining consensus, you should be considering the whole situation, (e.g., policies and guidelines that you are aware of, but nobody in the discussion happens to have mentioned). If our imaginary editor provided this summary: "Well, Alice says ILIKEIT and Carol says IDONTLIKEIT, so that's a 50-50 draw with no consensus", then our imaginary editor was thinking too narrowly, and upon discovering that Bob had previously supported this, should revise the summary statement to say "Alice says ILIKEIT, Bob previously supported it, and Carol says IDONTLIKEIT, so if we think it's safe to assume that Bob's views haven't changed per NOTSILENCE, that's probably a weak consensus for inclusion". WhatamIdoing (talk) 14:38, 12 July 2022 (UTC)
Thanks, yes it was meant as a simple example in a vacuum that assumes proper sourcing but doesn't take further dispute resolution into account.I think it demonstrates the point that in some situations, ONUS may seem less clear (or perhaps less decisive) in determining a discussion's outcome. Typically no consensus between editors actively discussing the conflict results in exclusion per ONUS, but other factors – not described by ONUS – can impact the outcome. And that's all perfectly fine, because I think that's the intent. My comments were more a response to Aquillion's concern about arguments saying "WP:ONUS somehow allows editors to completely ignore the well-established concepts of implicit consensus or consensus-via-editing". It does not, and any argument that pretends otherwise is missing the point. The purpose of ONUS is to kickstart discussion placing the ball in the court of supporter(s). It does not rule out potential talking points of that discussion.Perhaps we are more on the same page than first assumed, now that we've had a few paragraphs to flesh that out! ;-) -- GoneIn60 (talk) 16:16, 12 July 2022 (UTC)
Can you name a situation in which the result isn't exclusion? Looking at my comment above, for the simple example, it appears that the options are "exclusion per BLP" and "exclusion per ONUS". Either way, the result is exclusion. WhatamIdoing (talk) 16:24, 12 July 2022 (UTC)
Thought we just did? Let's say Alice lays out her case for inclusion, and Carol follows up with her counter. Assume both arguments are roughly equal in terms of applicable policies and guidelines. If discussion ends here, it is a no consensus draw, in which case the content is excluded per ONUS. Bob's implicit consensus can get completely overlooked when squarely focused on ONUS. However, the result could have been inclusion had Alice known to bring Bob's contributions into the mix, but you can't fault Alice for assuming it's a 1-1 draw. Bob is not actively "seeking to include disputed content", and the knowledge to take other factors into account, like implicit consensus, is prescribed outside of ONUS.SPECIFICO's comment below made me realize perhaps this tangent is straying too far off topic! --GoneIn60 (talk) 18:13, 12 July 2022 (UTC)
Your description here is basically "We initially did a bad job of assessing consensus, but when we assessed it correctly, we decided there was consensus for inclusion". ONUS supports including sourced information when you determine that there is consensus for including it. I do not understand how "Oops, we did a sloppy job of assessing consensus" could mean that ONUS is unclear or that factors other than ONUS determine what to do when the actual state of consensus does not support inclusion. The fact that editors sometimes do a bad job of assessing consensus does not amount to a Get Out of Jail Free card for any policy, including this one. WhatamIdoing (talk) 04:04, 13 July 2022 (UTC)
That's actually a good way of summarizing this rabbit hole. Essentially, the bad job occurs because of a misunderstanding of the order of operations. The revert or removal engages ONUS which states "to achieve" and "seeking", present tense descriptions that could be interpreted as active participation. So those in support of inclusion may think the responsibility is now on whoever participates in discussion moving forward. Actions like Bob's occurred in the past. But ultimately, I think you're right that it boils down to a proper understanding of consensus, and ONUS does link to WP:CON. --GoneIn60 (talk) 17:43, 13 July 2022 (UTC)
I should add that the position at WP:EDITCON about implicit consensus going away once challenged seems to me as a way to promote progress and change within an article. An editor's knee-jerk reaction shouldn't be, "This has existed for X months, and therefore, it has consensus. You must show it doesn't have consensus." My comments above are meant to address a situation in which discussion commences, setting implicit consensus aside, that ultimately reaches a stalemate. Implicit consensus could then be revisited and given some weight, if justified, tipping the scales in favor of retention. I'm not sure ONUS stands in the way of that. --GoneIn60 (talk) 17:35, 11 July 2022 (UTC)
It's perhaps also worth remembering that not all disputed changes are removals of materials. So while I think ONUS should apply to long term claims that haven't been challenged, I'm not as keen on including it for things like where content is located in the article, specific phrasing, content of the lead etc. So if fact A has been in the lead for a year without challenge or discussion and two editors disagree if that content is DUE for the lead, should ONUS come into play? Presumably the content is already in the body of the article. Same for the order of article sections. Basically I don't think putting more emphasis on ONUS means IMPLICIT effectively goes away. Springee (talk) 18:17, 11 July 2022 (UTC)
ONUS has always and only applied to whether a given bit of material is in the article. ONUS does not apply to moving the material to a different section. WhatamIdoing (talk) 23:28, 11 July 2022 (UTC)
Is the (lowercase) "onus" to achieve consensus for changes to consensus content not on those seeking the change, regardless of what that change is? Kolya Butternut (talk) 17:11, 12 July 2022 (UTC)
Depends on "consensus content". Is that consensus through editing or through discussion? If it's the former, EDITCON says that presumed consensus goes out the window as soon as it's challenged. But then there's different levels of editing consensus. If multiple editors have modified the disputed content over time, and the challenger is just one individual, does that then place the onus on the challenger? If so, what policy or guideline supports that? -- GoneIn60 (talk) 19:26, 12 July 2022 (UTC)
Kolya, I think the short answer to your question is "no". For most editing, the onus/obligation/duty to figure out what the consensus is for the article is on every editor who wants to see the article improved. There has been a fairy tale going around for some years about it always being the other editor's job to start the discussion, but that was never in any policy. In fact, policies like Wikipedia:Editing policy#Talking and editing focus on what "you" can do to edit collaboratively and resolve any disputes that come up – always "you", and never "the other guy". WhatamIdoing (talk) 04:12, 13 July 2022 (UTC)
Sure: An essay written by a couple of people presents that opinion. Also, making any edit to an article constitutes one form of "saying so". Have you ever seen anyone edit an article to communicate their belief that the version they're changing was better than the one they're posting? WhatamIdoing (talk) 16:19, 13 July 2022 (UTC)
Yes, onus is a simple concept. Like you said, For most editing, the onus/obligation/duty to figure out what the consensus is for the article is on every editor who wants to see the article improved.Kolya Butternut (talk) 16:54, 13 July 2022 (UTC)
As others have said above, "it" does not depend on the topic. I think you're just generalizing from having observed that some topic pages are widely and frequently edited and others are broadly neglected. To take two examples, many topic areas in current events are followed by hundreds of editors and talk page consensus is a reasonably good sample as to how the editing community of millions would resolve ONUS and WEIGHT and BLP. Then, on the other end of the scale, topic pages about social science and humanities, obscure works of fiction, and pop electronic media subjects (games, tik-tokkers, bloggers) are edited by a much smaller population -- and talk page discussion is likely to be dominated by a small number of enthusiastic POV advocates -- who are prone to good faith misapplication of NPOV. SPECIFICOtalk17:06, 12 July 2022 (UTC)
If you or any other editor(s) want to attempt a blanket treatment on all talk pages? Then by all means do so. There's the ideal world & the real world, which aren't identical. GoodDay (talk) 20:10, 12 July 2022 (UTC)
I am not saying there's no way to parse it. I am saying that you parsed it in the wrong dimension. If this interests you, I suggest you brainstorm some ways to parse it according to the edit environments that are typical across topics rather than the topics themselves, within which there are diverse editing environments and diverse factors that effectively make a topic-based "real world" an unnecessarily blunt instrument. SPECIFICOtalk21:11, 12 July 2022 (UTC)
I wonder if the word "topic" in that comment meant something closer than "a specific article, with its individual history and characteristics" than to "any and all articles on the general topic of ______, broadly construed". One could not really claim that all articles about (e.g.,) military history or films or the Middle East will be, or even should be, treated the same. WhatamIdoing (talk) 04:19, 13 July 2022 (UTC)
In general, I think it would depend on the specific content and sources. If the two sources are otherwise equivalent for the content being sourced, then yes, a free-to-access source would be better. If they're not equivalent, there's always the option of providing both. In this case, for citing a statement about Emmy nominations, the free source, Variety is probably equivalent (it's reliable for things like Emmy awards), and since it's free, it's probably better than WaPo. Now, if the content was about the war in Ukraine or something like that, it'd be a different story. Levivich[block]06:08, 13 July 2022 (UTC)
There is no official rule favoring the FUTON bias, although some editors personally prefer it and advocate for it.
"All else being equal" is an uncommon situation. "Not truly 'equal' but definitely 'adequate for this claim'" is much more common. People who don't like sourcing disputes often add their free-as-in-beer source next to the paywalled one, rather than removing the paywalled one and substituting their preferred one. WhatamIdoing (talk) 16:26, 13 July 2022 (UTC)
"IMO "all else being equal" is a simplified way of say "the decision involves taking both this consideration and other factors into consideration". IMO a source that is freely and electronically available in the language of the Wiki which cited it is going to be more useful to a typical reader and makes it easier to determine if WP:Ver has actually been satisfied compared to some combination of "paid only", off line and in a different language. . But we can't have wp:ver precluding any source based on those attributes. North8000 (talk) 16:48, 13 July 2022 (UTC)
I was more commenting on the fact that the more difficult it is to see and read the source, the more the editor has insulated the text from verifiability scrutiny. Not that that implies that we should be excluding such sources. North8000 (talk) 02:43, 15 July 2022 (UTC)
Sure, it's harder to check such sources. Sometimes we even cite sources that are truly difficult to check. Gartner reports on the tech industry can cost hundreds of dollars. Medical textbooks are notoriously pricey, and some reference works are worse. Springer's got a sweet Encyclopedia of Signalling Molecules that's on sale right now for more than US$5,000 (ISBN 978-3319671987; get it while you can!). Some rare books are extraordinarily useful, but only available if you can travel to where the book is. This is a very common problem with {{cite sign}}. The policy says all of that's okay, though. Ease of verification does not ever seem to have been a concern for this policy. WhatamIdoing (talk) 06:32, 15 July 2022 (UTC)
$5000? I don't think I'll even check whether their 75% off offer applies to that! ;-) I definitely agree that even if that is definitive many readers can be quite poor, and it is better to put in a second citation that lets them check the basics without needing to sell their kidneys. Being a free encyclopaedia is one of the major aims and the knowledge is not quite free if the verification requires money. I agree verifi8ability shouldn't mention this as a requirement but we should certainly keep it in mind! NadVolum (talk) 15:07, 21 July 2022 (UTC)
In my experience, small community libraries do not give patrons access to online technical literature, because of the expense, nor do they have physical volumes on the shelves. University libraries generally give access to physical volumes to whoever walks in the door, but you have to be a member of the faculty or a student to access the non-free online publications the university subscribes to. So I don't believe 90% of our readership has access to technical journals, even in first-world countries, unless they're prepared to travel to one of the few really large public libraries, like the New York Public Library. Jc3s5h (talk) 15:30, 21 July 2022 (UTC)
As I've said in the past on Jimbo's page - the WMF needs to fix the access issues if they expect volunteers to keep the quality of the project up to par. Jake & the TWL team are doing all they can to get us access, while the WMF capitalizes on our work with Enterprise...but I wonder how long that income stream will last if we keep moving toward lowering our standards relative to RS, and V. WP:NPP & AFC reviewers are already struggling with backlogs despite the WMF's ability to provide the funds & resources we need to develop the tools necessary to do our job - like access to articles behind paywalls. If they don't, then we can rest assured that we did our best to stop the downward spiral and loss of quality in what we publish, not to mention the loss of $$ contributors to the WMF, the loss of experienced, hard working editors who got burned out, and the inevitable paradigm shift that will occur as UPE/PE dominates the encyclopedia, making it appear more as a sales catalog than an encyclopedia. At the rate AI and decentralization is advancing globally, it probably doesn't matter anyway. Good luck with relaxing our PAGs. Atsme💬📧18:06, 21 July 2022 (UTC)
Jake left the WMF more than three years ago. Samwalton9 is the best contact for the Wikipedia:The Wikipedia Library now. If you want to talk about whether meeting Google's demand for a high-speed data funnel has ever been cost-free to the WMF (or to Bomis before the WMF existed), or whether providing Google with extra services to meet a legally enforceable Service-level agreement will be cost-free, then you're looking for Wittylama. WhatamIdoing (talk) 19:14, 22 July 2022 (UTC)
This is Wittylama - answering/commenting from my staff account, since this is definitely a 'work related' question. Interestingly, the topic is at the intersection of two things I'm directly involved in... Until last year I was coordinating the "WikiCite" grant which was about supporting things to do with references (access, reuse, datasets, outreach...), and as the outcome of that project I made this proposal for what our movement might be able to do in the future: I called it "Shared Citations". This is just a proposal, no funding/approvals yet, but I think that one of the biggest problems in how we handle references is that they are extremely labour intensive to manage and we can't do much analysis of the corpus of citations we actually use across wikimedia project. Hypothetically, with a citation management database we ought to be able to query what % of references in (for example) Medicine topics are behind paywalls, or whether there are any academic citations which are still 'in use' on WP articles but the publication has later retracted it. As you can see on the proposal's mockup of what I citation record might look like (in slide 8) it would be possible to link to various places where a source can be found (open, closed, preprint, pirate) at the discretion/editorial policies of the community.
With regards to the Enterprise API service that was mentioned... The status quo is that donor funds are subsidising the business model of large commercial organisations and because all of the API services currently offered are independent of each other, smaller external organisations are willing-but-unable to re-use Wikimedia knowledge in their products. The re-use of Wikimedia content for commercial purposes is and always has been allowed/encouraged in accordance with Free Licensing principles, but by providing no assistance for smaller commercial orgs, we've actually been limiting the number of ways our knowledge can reach different audiences around the world. So, by providing a service where the big companies pay for the operation of commercial-specific features they require (super high volume/speed, contractually binding SLAs, dedicated customer support) AND where smaller companies can access the exact same product/service at the volume they desire, that means that a) commercial usage subsidises the movement not the other way around, and b) there is a more level 'playing field' for access to free-knowledge. There's an Essay, FAQ and Operating Principles, if you're interested in more detail. The shorthand metaphor is that: "they're paying for some bigger pipes, the water is the same". Sincerely, LWyatt (WMF) (talk) 12:24, 27 July 2022 (UTC)
@WhatamIdoing - The en.wp WP:WikiProject Academic Journals project has not explicitly been involved in the design of that proposal, but several people involved in it are. in particular User:Evolution and evolvability who has been doing lots of thinking about the 'retracted' problem has been involved in the designing of the 'shared citations' proposal. And for the example of status change, this is emblematic of the problem for any kind of metadata maintenance: that we're doing it manually, to varying degrees of consistency, across each wiki independently. That's a burden on volunteer labor and introduces many vectors for data inconsistency and is especially burdensome on smaller language editions. LWyatt (WMF) (talk) 12:30, 1 August 2022 (UTC)
(1) Embolden the words "Please immediately remove" in the sentence "Please immediately remove contentious material about living people that is unsourced or poorly sourced." in the lead section.
(2) Add the following text to the aforementioned sentence, before the period at its end: and [[WP:LIBEL|libelous material]].
I've always known that it's a Wikipedia policy to remove libelous material from articles about living people, since it may seriously and negatively affect them. I've also read today about the lawsuit from April 2019(?), in which the Wikimedia Foundation was obligated by a German court to remove libelous material from an article's history. The author of the post in the Foundation's blog wrote that the Foundation was forced to oversight the article's history. So I was surprised when I read WP:LIBEL, since it isn't linked to from Wikipedia's content disclaimer, about page, the five pillars, or anywhere important really. To quote: "It is a Wikipedia policy to immediately delete libelous material when it has been identified. Page revisions containing libelous content should also be removed from the page history. Libelous material (otherwise known as defamation) is reasonably likely to damage a person or company's reputation and could expose Wikipedia to legal consequences." This is in contrast to the implication of the Foundation's blog post that oversight shouldn't be exercised in such cases.
So please, if you can make these changes to the article to make it clear that certain material must be immediately removed as of Wikipedia policy, especially if one such policy is the Foundation's policy due to potential legal consequences and surprisingly isn't widely mentioned. 85.64.76.29 (talk) 03:57, 31 July 2022 (UTC)
If you'd like to, you could restate your reasons in a new section, without needing to propose a "change X to Y"-style edit like you needed to for the edit request. Probably it's fine to leave this up, as this talk page has some fairly active participation. I don't strenuously oppose the change, but I'd prefer not to see more boldface added to the intro. WP:LIBEL is linked prominently from WP:BLP, which is the most likely place editors will be when working with contentious material about living people. For the scope of this policy, I think "contentious material about living people that is unsourced or poorly sourced" covers libelous material as well. Firefangledfeathers (talk / contribs) 04:22, 31 July 2022 (UTC)
Thank you. What about libelous material about organizations? And if this is such an important policy, it should be linked to standalone without relying on editors to read the BLP policy and somehow notice this too. At least that's what I think. 85.64.76.29 (talk) 04:39, 31 July 2022 (UTC)
I have no objection to adding this, but I am not certain that it's important to add the link, as the problem rarely arises, and when it does, people don't seem to have trouble doing the right thing. For example, the rumors in April 2019 indicated that that situation could have been handled under BLP rules anyway.
In my experience, the main problem with talking about Wikipedia:Libel on wiki is that we have a lot of armchair lawyers, and not very much appreciation for the variety of legal rules around the world. For example, are defamation and libel the same thing? Are accurate/true facts defamation? Is it legally possible for a corporation, political party, or other organization to sue over libel? Different jurisdictions have different answers to these questions, but editors tend to think first/only of the rules they're personally familiar with.
Then there is the impossibility of applying it to individual cases: Is "During a #metoo thread on Twitter, two people accused him of raping them" (a true fact, in that they really did say this) still defamation? How about meticulously documenting every time someone has been involved in lawsuits or other legal trouble? I knew two people who were in the habit of suing their employers and business partners. Having the "true facts" widely known would do significant harm to them, because nobody who knows those "true facts" would ever agree to work with them. But is it actually libel to say "He worked for X from 2000 to 2002, and then sued them for wrongful termination; he worked for Y from 2002 to 2003, and then sued them for wrongful termination; he worked for Z in 2004, and then sued them for wrongful termination"? The answers aren't simple. WhatamIdoing (talk) 19:10, 3 August 2022 (UTC)
But these are questions you raise regarding the meaning of the WP:LIBEL policy. Whether it means one thing or the other, it's still a Wikimedia-wide policy. Maybe the Foundation needs to narrow the policy or clarify the explanation in the WP:LIBEL page, but because of the official status of it and its importance, I still think it needs to be mentioned. I think the right place for these questions is in the talk page for WP:LIBEL (or perhaps emailed to the Foundation), and not affect whether it should be linked to from WP:V. 85.64.76.29 (talk) 15:35, 6 August 2022 (UTC)
What does the libel policy have to do with verifiability (i.e., that other people can verify that the information came from a reliable source, and not from an unreliable source/the editor's imagination)?
As I see it, there are two possibilities, since the text already as-it-is in the verifiability page mentions the BLP policy:
(1) Either the text could be amended to include my changes or completely removed if it doesn't belong in the page; or...
(2) The text could be left as-is if you don't care that much that it would be complete/precise/etc.
So... Do you suggest the text "Please immediately remove "''Please immediately remove [...]''" be erased from the article completely?
I think it was put in the page because verifiability helps prevent libelous information that is blatantly false, so it probably seemed relevant for the original editor of the page to request false info in BLPs to be removed. That's how I see it. 85.64.76.29 (talk) 17:14, 8 August 2022 (UTC)
Are prior publications the only way to establish someone as an expert for the SPS expert exemption?
The current wording of SPS offers definition of who can qualify as a expert, "Self-published expert sources may be considered reliable when produced by an established subject-matter expert, whose work in the relevant field has previously been published by reliable, independent publications." How would this apply in areas where expertise is better defined by ones resume vs ones CV? In a resent RSN discussion[4] I used light aircraft and automobile racing as examples. Would we treat as expert opinion the self published comments of Burt Rutan on the design of a light airplane, Rick Mears on performance driving, [Adrian Reynard] on the design of formula racecars or Ralph Firman on design and sales of jr level formula racecars? I think all of these people would be acknowledged as experts in their respective areas based on their easily documented accomplishments. However, our current SPS exception seems to only consider academic publication as proof of expertise while ignoring industry for the same. I've tried to find some prior discussions on this topic [5][6]. I think much of the concern in those was related to SPS where the person comes across as an expert (and perhaps is quite knowledgeable) but it's just editor's opinions of the person's work vs any sort of outside measure. I can see this as a legitimate concern since a slick website or videos can make a person of just average expertise present as more knowledgeable than they may actually be. However, I feel that if we can point to a solid list of accomplishments this shouldn't be a concern (with all other self published expert limitations applied). Springee (talk) 12:55, 8 August 2022 (UTC)
What if we aren't aware of any publications by the person? Using Adrian Reynard as an example, I'm not aware of any work published by Reynard (though I'm sure he has been interviewed, for argument sake lets assume he has never published), but if he were to offer an opinion on the cause of a NOTABLE crash of the Mercedes-Benz CLR would we treat it as SPS expert opinion? Springee (talk) 14:28, 8 August 2022 (UTC)
If an individual is widely cited/quoted by independent reliable sources on a topic (is treated as subject-matter expert by RSs), that seems like it would be a factor worth considering. Schazjmd(talk)14:47, 8 August 2022 (UTC)
I don't think so. The way that you get widely quoted by journalists is to prioritize replying to them quickly, and to hand them short, easily quoted statements. Knowing what you're talking about is not really a requirement. In some industries, some consultants use getting quoted in independent news and magazines as their primary marketing technique.
WhatamIdoing, I didn't search Rutan but that doesn't surprise me. I'm asking this as a hypothetical so to some extent I have to make up the facts as I go :D Looking at the other three, particularly Firman would would be little known outside of his industry, I would still be curious if we would allow him to be an expert in context of a SPS opinion. As an example, he might offer an opinion related to the impact Swift Engineering had on the Formula Ford market in the 1980s. I will note this is a slightly loaded example because I know Firman was asked exactly that in a book cited in the Formula Ford article. Springee (talk) 02:36, 10 August 2022 (UTC)
I would say these are probably the exception that prove the rule. Sure, these are people who know a lot about their respective fields, but are they any good at putting that knowledge into written words? That's what the "has authored independently published articles" requirement is so good at sussing out. There are probably dozens of people in every field who are very good at what they do, and they probably even have interesting opinions. But that would make them worth quoting when quoted by journalists who are reliable and publishing in RSes. It would not make them necessarily good (or reliable) at writing their own stuff. hence why they can fail SPS but be worth quoting as in RSOPINION. SPS is reserved for people who are subject-matter experts and good at writing about it. — Shibbolethink(♔♕)13:37, 10 August 2022 (UTC)
So continuing with my hypothetical, let's assume Firman has a personal blog or his company has a blog and on that blog he said, "The reason the Swift DB1 was so successful was XYZ". Would RSOPINION allow the use of that claim so long as it was presented as an attributed opinion? Springee (talk) 14:43, 10 August 2022 (UTC)
It'd be self-published and primary, so it might fail WP:DUE anyway. Generally speaking, any publication in which Alice says _____ is a reliable source for a statement that "Alice said _____". The fact that it's "reliable" is not really the main question. WhatamIdoing (talk) 15:18, 10 August 2022 (UTC)
I agree that once "expertness" has been established, next up is the relative importance or dueness of what is being said. So in the given case, if there was some back and forth about reasons for success, then that opinion might fit right in, maybe not if it is just sitting out there all by itself as a random opinion. Selfstudier (talk) 22:17, 10 August 2022 (UTC)
I presume the DUE part is similar to any other case of a SPS. For example, Alan Dershowitz would certainly count as an expert in constitutional law. If he opines about something the SCOUS does we would still have to establish that Dershowitz's opinion on the subject is DUE. I actually think that would be harder in a case where there are plenty of RSs talking about the SCOUS's decisions. Anyway, off the top of my head I can't think of a time I've run into a case where this question might apply but I wanted to seek clarification anyway. It's something I could certainly see coming up in articles about auto racing/specific race cars and likely a lot of other topics that have limited general interest. Springee (talk) 23:18, 10 August 2022 (UTC)
My inclination is to first do whichever test seems easiest, rather than going in a set order.
Reliable sources aren't limited to sources of general interest. I assume that there are still some magazines about auto racing out there. WhatamIdoing (talk) 01:32, 11 August 2022 (UTC)
RfC: A TikToker, ... , other accused constitute 'Public figure' or not?
Requesting inputs about WP policies regarding, WP:BLP protocols and naming of the accused in relation to mentions of allegations and counter allegations in the given article, against a female victim of sexual assault, her associates and also other accused.
WP:ECREE has become increasingly useful reference. I especially like the bullet points that identify what makes certain sources "extraordinary" in the right way and what does not.
To that end, I think a good link to include here is to WP:SENSATION as often sources that get touted as being extraordinary are found to run afoul of that. A brief bulletpoint that said something like:
Any source engaging in sensationalism by promoting extraordinary claims without considering ordinary or standard explanations.
@ජපස, I'm not sure this fits. We don't want someone to read a sensationalist "news" article and repeat that sensational story in Wikipedia's voice. But imagine a source that says something sensationalist plus also something very ordinary. We don't want to rule out the source entirely. Here's an example:
I saw a feel-good news story the other day. It contained a statement that many Wikipedians from Western cultures would consider a case of "promoting extraordinary claims without considering ordinary or standard explanations", insofar as it ended with a quotation in which the person to whom something nice happened said that it was the result of divine intervention. Many Wikipedians would scoff at that and believe that the real "cause" was random chance. (I have forgotten what the story was about, but it was something like having a medical emergency right in front of the right sort of healthcare provider.)
This would be a lousy source if you wanted to write "A bona fide miracle happened in the American South last week" but it would be a perfectly fine source if you wanted to write "Alice Expert is a specialist in scaryitis" or "Alice Expert was awarded the Minor Local Hero Award in 2022 for providing medical assistance during an emergency". WhatamIdoing (talk) 23:03, 23 August 2022 (UTC)
Hi! I missed your first comment, so apologies. But I think that it is always true that the sensationalist content is what is the problem. The issue is the extraordinary claim itself -- not the claims that are not extraordinary. As for examples, I think the one you are mentioning is great. I have seen accounts who have tried to use sources like that to argue for the veracity of extraordinary claims. jps (talk) 09:41, 24 August 2022 (UTC)
To keep the focus on the extraordinary claim, I don't think we could write "Any source engaging in sensationalism...". People would be rejecting sources for any and all claims, not just for extraordinary claims. I think we'd have to say "sensationalist claims, including extraordinary claims that could have ordinary or standard explanations".
It dawned on me that this is becoming a distributed enough problem that an essay is probably the right maneuver. WP:N should not be doing all this lifting. jps (talk) 20:32, 27 August 2022 (UTC)
I only just saw this thread. You should probably notify WP:N to remove it. It's CREEP that's redundant with V, NOTGOSSIP, and NOTNEWS. At the end they even just laundry list the policies that they say this "applies ... beyond". The "See also" makes zero sense. The wls to tabloid journalism, infotainment, etc. add zero useful information to the guideline ('sensationalism' is defined by being sensationalist, and here are other subtypes of sensationalism in other media, and that's supposed to help?). If it's replaced by an essay, the essay should be written from scratch, because this guideline is awful as is. SamuelRiv (talk) 03:34, 29 August 2022 (UTC)
Some of it is pretty awful. It looks to me like according to this anything supported by a primary source getting challenged becomes an exceptional claim, and the super hilarious part is that it will be especially true if the proponents say there is a conspiracy to silence them (because you know how important it is for us to focus on editors, and their mental health rather than content.) Haha! Huggums537 (talk) 03:51, 29 August 2022 (UTC)
If you think that the WP:SENSATIONAL should be removed from WP:NEVENTS, then I suggest that you make that proposal yourself on its talk page. If you haven't made this sort of proposal before, then I suggest that you figure out how long its been in this guideline and what discussions led to it. Then think about whether the information might be useful or informative to someone. (Not everything in policies and guidelines is a "rule". Sometimes, the goal is to explain why other editors might not share your initial opinion.) WhatamIdoing (talk) 15:13, 29 August 2022 (UTC)
Slap me with a WP:Trout! I just realized you were saying SENSATIONAL is awful, not ECREE. (I guess that is why I couldn't completely agree with you that all of it was awful, and I could only see those couple things wrong with ECREE that I mentioned above.) However, I do agree SENSATIONAL is a regurgitation, and would be better served as an explanatory essay. Huggums537 (talk) 20:12, 29 August 2022 (UTC)
The advantage of going the new essay route as WhatamIdoing (talk·contribs) suggests above is that we don't have to muck about in top-level WP:PAGs which should be let fairly stable in an ideal world. There are basically a few main points:
Some sensationalized stories are published by otherwise trustworthy media outlets. These can be tricky to navigate and often require careful corroboration through WP:Independent sources (I am reminded of a particular New York Times article on the most recent UFO craze which was authored by three authors that included one Leslie Kean (a famously credulous ufologist).
WP:RECENTISM and fog-of-war type reporting can end up causing real problems too. Wikipedia being "non-innovative" should lag behind most "breaking news" anyway, but in cases where sensationalism is running amok, there may never be a point where we reach a well-sourced situation. News cycles move on, people lose interest, and wild claims can end up unchallenged in various outlets without being noticed by others. It is not Wikipedia's place to correct that misinformation, but Wikipedia is also under no obligation to spread it.
.jpg)
Not done for now: please establish a consensus for this alteration before using the 