Wikipedia talk:Edit warring/Archives/2020/July

Latest comment: 4 years ago by Paradise Chronicle in topic Revert rules


There's no mention of status quo here

I noticed that (unless I'm missing it) there seems to be nothing here or in another policy/guideline that says explicitly that, during a dispute where the prevailing consensus is not immediately clear, the status quo should be retained while discussion takes place. WP:STATUSQUO, while frequently cited, goes to a section of WP:Reverting, which is only an essay (although perhaps it should be upgraded). Should we codify this de facto rule? {{u|Sdkb}}talk 18:01, 2 July 2020 (UTC)

You need to take this to WP:Village pump (policy). As you know, WP:STATUSQUO is an essay. And it is not going to be given guideline or policy status in a back door way. Flyer22 Frozen (talk) 03:53, 3 July 2020 (UTC)
I don't think Sdkb is suggesting granting something guideline or policy status in a backdoor way. I think they're trying to ascertain, in a forum presumably with people who care about such a thing, whether it would be a good idea or not. A future step might involve going to VPP for further discussion or for a formal proposal. As to whether I think it's a good idea or not, I think rather than upgrading STATUSQUO, we could find a way to incorporate some essence of it here. Linking to it as well is something I'd be open to but don't feel strongly about one way or another. Best, Barkeep49 (talk) 04:21, 3 July 2020 (UTC)
I'm in favor of making a rule that status quo ante bellum should be maintained while consensus is being determined part of the policy against edit warring. It would be a much clearer rule to stop edit warring than the current complicated revert counting rules and head off the inevitable "I'm not edit warring; you are". But I would insist that such a policy make it clear that this is just while editors are actively trying to develop consensus and does not mean that status quo has priority as a matter of policy in the underlying dispute. Bryan Henderson (giraffedata) (talk) 17:28, 3 July 2020 (UTC)
If anyone ever creates RfC on WP:Village pump (policy) on the matters please {{ping}} me out. I'm in favor of creating more reliable functionality to achieve quick WP:DISPUTE resolutions, rather than falling back to something that is considered default. AXONOV (talk) 18:52, 4 July 2020 (UTC)

Clarifying the definition of 'revert'

I had posted the below proposal here in § Clarifying the definition of ‘revert’. Seeing no objections, I effected the change. MrX reverted with edit summary "This changes the meaning and makes it less clear. This would need to be discussed on the talk page first."

So here we are as I originally wrote:

To my reading, the text in the red box and in the para following the red box should be reworked for clarity (inserts indicated in italics below):

An editor must not perform more than three reverts on a single page—whether involving the same or different material—within a 24-hour period. An edit or a series of consecutive edits that undoes other editors' actions—whether in whole or in part—counts as a revert reverting edit. A series of consecutively saved reverting edits by one user, with no intervening edits by another user, counts as one revert. Violations of the rule often attract blocks of at least 24 hours. Fourth reverts just outside the 24-hour period may also be taken as evidence of edit-warring, especially if repeated or combined with other edit-warring behavior. See below for exemptions.

A "page" means any page on Wikipedia, including talk and project space. A "revert" means any edit (or administrative action) that reverses the actions of other editors, in whole or in part, whether involving the same or different material. An "edit" is here read broadly to include administrative actions; and "editors" to include administrators. A series of consecutively saved reverting edits by one user, with no intervening edits by another user, counts as one revert.

where 1) the 2nd sentence in the red box has been simplified to refer to a reverting edit so it can lead into the clear definition of revert in terms of reverting edit that I moved up from the last sentence in the subsequent para; and 2) the 2nd sentence in the subsequent para — which repeated text in the red box with minor elaboration — has been reworked to focus on that elaboration.

In final form, that would be:

An editor must not perform more than three reverts on a single page—whether involving the same or different material—within a 24-hour period. An edit that undoes other editors' actions—whether in whole or in part—counts as a reverting edit. A series of consecutively saved reverting edits by one user, with no intervening edits by another user, counts as one revert. Violations of the rule often attract blocks of at least 24 hours. Fourth reverts just outside the 24-hour period may also be taken as evidence of edit-warring, especially if repeated or combined with other edit-warring behavior. See below for exemptions.

A "page" means any page on Wikipedia, including talk and project space. An "edit" is here read broadly to include administrative actions; and "editors" to include administrators.

Further discussion welcome. Humanengr (talk) 06:11, 24 May 2020 (UTC)

Further re motivation: The existing text has 3 sentences that define 'revert':
  • An edit or a series of consecutive edits that undoes other editors' actions—whether in whole or in part—counts as a revert.
  • A "revert" means any edit (or administrative action) that reverses the actions of other editors, in whole or in part, whether involving the same or different material.
  • A series of consecutively saved reverting edits by one user, with no intervening edits by another user, counts as one revert.
where those 3 sentences have considerable duplication and the last one introduces the term 'reverting edit' without defining it. Humanengr (talk) 20:47, 26 May 2020 (UTC)
I have a couple of comments:
  1. The sentence "A series of consecutively saved reverting edits by one user, with no intervening edits by another user, counts as one revert." is not accurate. It should read "A series of consecutively saved edits by one user which include reverting edits, with no intervening edits by another user, counts as one revert." The reason is, a series of consecutive edits can consist of reverts, additions, rewording, typo corrections, reformatting, and so on.
  2. The sentence "An "edit" is here read broadly to include administrative actions; and "editors" to include administrators." is less clear that the previous wording. Specifically, "is here read broadly" is unnecessarily word to the point of obfuscation.
I'm open to any rewording that actually makes the policy clearer, especially for inexperience editors. - MrX 🖋 22:01, 26 May 2020 (UTC)
Intervening edits are sometimes irrelevant to whether it is one or more separate reverts, specifically if the subsequent edits do not undo the intervening edit. Can we try again? --Bsherr (talk) 22:42, 26 May 2020 (UTC)

[See revised proposal below] Adjusting only for MrX's comments, the proposal becomes:

An editor must not perform more than three reverts on a single page—whether involving the same or different material—within a 24-hour period. An edit that undoes other editors' actions—whether in whole or in part—counts as a reverting edit. A series of consecutively saved edits by one user which include reverting edits, with no intervening edits by another user, counts as one revert. Violations of the rule often attract blocks of at least 24 hours. Fourth reverts just outside the 24-hour period may also be taken as evidence of edit-warring, especially if repeated or combined with other edit-warring behavior. See below for exemptions.

A "page" means any page on Wikipedia, including talk and project space; "edits" include administrative actions; and "editors" include administrators.

@MrX: Does that suffice? A q: By 'by another user' do you intend 'by -any- other user' or 'by a user whose edits were reverted'? (I presume the former; just making sure.) Humanengr (talk) 23:25, 26 May 2020 (UTC)

@Bsherr: Would replacing "with no intervening edits by another user" with "with no intervening edits that (in turn) revert those edits" help? Humanengr (talk) 23:25, 26 May 2020 (UTC)

On an aside to much of the above (perhaps not too aside), I don't really understand why this page is trying to stretch into topics already covered by WP:WHEEL when it speaks to ... "edits" include administrative actions; and "editors" include administrators. It seems like needless clarification in context already covered elsewhere. --Izno (talk) 02:58, 27 May 2020 (UTC)

Makes sense to delete that; reduces complexities for newbies. Humanengr (talk) 03:13, 27 May 2020 (UTC)

@MrX, Bsherr, and Izno: Does the following updated proposal address your concerns:

An editor must not perform more than three reverts on a single page—whether involving the same or different material—within a 24-hour period. An edit that undoes other editors' actions—whether in whole or in part—counts as a reverting edit. A series of consecutively saved edits by one user which include reverting edits—with no intervening edits by another user that (in turn) revert those edits—counts as one revert. Violations of the rule often attract blocks of at least 24 hours. Fourth reverts just outside the 24-hour period may also be taken as evidence of edit-warring, especially if repeated or combined with other edit-warring behavior. See below for exemptions.

A "page" means any page on Wikipedia, including talk and project space.

Humanengr (talk) 08:54, 27 May 2020 (UTC)

Yes, I think that's an accurate definition. That's good work. --Bsherr (talk) 16:49, 27 May 2020 (UTC)
Thx … A minor shortening: "A series of consecutively saved edits by one user" --> "Consecutive edits by one user". With that, are we good to go? Humanengr (talk) 21:13, 28 May 2020 (UTC)
With a shout into the dark room that "revert" is not a noun, yes by me. --Bsherr (talk) 16:58, 31 May 2020 (UTC)
@MrX and Izno: I had found interpreting this passage very vexing, and so was hoping to relieve (seemingly many) others of that burden. Are there any further issues? Humanengr (talk) 01:01, 3 June 2020 (UTC)
@Bsherr On your grammar point, it looks like WP:EW is the policy page with the most 'revert(s) as noun' (along with several 'three/one/zero-revert rule' labels). The noun uses are easy to fix; there are only a few other instances on other policy pages. Humanengr (talk) 05:36, 10 June 2020 (UTC)
Shall we give it a try? --Bsherr (talk) 14:09, 10 June 2020 (UTC)
Really, given the various debates that have been had about defining a revert (including on this very talk page), I'd rather that the text stay as it is unless having wide support. How is the proposed change supposed to help things? Please don't ping me if you reply. Flyer22 Frozen (talk) 21:02, 10 June 2020 (UTC)

I put the proposed changes (the red box and subordinate text; revert(s) (n) --> reversion(s)) in a draft. @Bsherr, I thought it best to retain the 'three-revert rule' label as a WP coinage. But at least the noun-verb issue is resolved in the other text. Thoughts? Humanengr (talk) 18:51, 13 June 2020 (UTC)

That does make it easy to see what would be changed. I don't see a need to change "reverts" to "reversions." And I think I like "A series of consecutively saved reverting edits by one user, with no intervening edits by another user, counts as one revert." better than ""Consecutive edits by one user which include reverting edits—with no intervening edits by another user that (in turn) reverts those edits—counts as one reversion." Flyer22 Frozen (talk) 21:09, 13 June 2020 (UTC)
This new draft of that § reflects your comments. No revert->reversion changes proposed there or elsewhere in the article. (On the latter, FWIW, AFAICS, other than this page, which uses 'revert(s)' as noun 8 times and twice as a verb, only 5 other policy-level pages use it as noun: WP:VD, WP:ONLYREVERT, WP:TED, WP:CV, WP:DR; most of those also as verb. 15 other policy-level pages have it as verb only.) Humanengr (talk) 03:00, 20 June 2020 (UTC)

Bringing the proposal here for reference:

An editor must not perform more than three reverts on a single page—whether involving the same or different material—within a 24-hour period. An edit or a series of consecutive edits that undoes other editors' actions—whether in whole or in part—counts as a revert reverting edit. A series of consecutively saved reverting edits by one user, with no intervening edits by another user, counts as one revert. Violations of the rule often attract blocks of at least 24 hours. Fourth reverts just outside the 24-hour period may also be taken as evidence of edit-warring, especially if repeated or combined with other edit-warring behavior. See below for exemptions.

A "page" means any page on Wikipedia, including talk and project space. A "revert" means any edit (or administrative action) that reverses the actions of other editors, in whole or in part, whether involving the same or different material. A series of consecutively saved reverting edits by one user, with no intervening edits by another user, counts as one revert.

Humanengr (talk) 00:19, 23 June 2020 (UTC)

Flyer, I had incorporated your two suggested changes: "I don't see a need to change 'reverts' to 'reversions.' And I think I like 'A series of consecutively saved reverting edits by one user, with no intervening edits by another user, counts as one revert.' better than "Consecutive edits by one user which include reverting edits—with no intervening edits by another user that (in turn) reverts those edits—counts as one reversion.'" into the draft I created in my sandbox which I later posted above. In your reverting edit summary, you said "I was clear that I disagree." What's the issue? Did you miss that I had incorporated your suggestions? Humanengr (talk) 21:46, 23 June 2020 (UTC)

I don't like the "reverting edit" language. No one says that. You stated that you "incorporated [my] two suggested changes." I didn't suggest anything, though. I simply objected to changing "reverts" to "reversions" and noted that I prefer the "A series of consecutively saved reverting edits by one user" wording to the "Consecutive edits by one user which include reverting edits—with no intervening edits by another user" wording. But I didn't notice that both of those wordings are yours. What I prefer is the current "An edit or a series of consecutive edits that undoes other editors' actions" wording that is in the policy now.
I still don't see why it's necessary to change the current wording to any of your proposed wordings. Why are any of your proposals an improvement? I'm trying to understand why you are proposing the changes you've proposed. Flyer22 Frozen (talk) 00:15, 24 June 2020 (UTC)
I second Flyer22 Frozen's question. As to the proposal: (1) Don't use both "user" and "editor", it will just make newbies wonder if there is a difference. (2) If someone consecutively does revert, add-new, revert that counts as two reverts by this proposal, but it has always been regarded as one (this problem was noted above). Zerotalk 06:07, 24 June 2020 (UTC)

@Flyer, To explain more re motivation: As I said: I had found interpreting this passage very vexing, and so was hoping to relieve (seemingly many) others of that burden. I still find it vexing. Note also that MrX said above: I'm open to any rewording that actually makes the policy clearer, especially for inexperience editors.

I'm realizing now a contributing factor to that vexation stems from what Bsherr noted: 'revert' is not a noun. Bsherr is roughly correct: 'revert' is used only relatively rarely as a noun. Compare the definitions in Wiktionary for it as a noun with that for it as a verb. The verb use is comparatively familiar. And only the 4th noun definition — in computing — approaches anything like how it's used here. I, for one, had never encountered its use as a noun. Within Wikipedia policy-level pages, it appears as a noun on only 5 other pages — and in all those it also appears as a verb. In contrast, 'revert' appears as verb (and only as verb) on over a dozen others. (It does appear as part of three other WP terms of art, 'three-revert rule, one-revert rule', or 'zero-revert', on this and 4 other pages along with verb use on all but one of those. But those should be considered as compound constructions separately from the isolated 'revert' noun.) The use of 'revert' as a noun — and particularly as a noun with the unique definition given here — is indubitably off-putting as a strange construction. I hadn't realized this initially, but (thanks to Bsherr for setting me on this path) my introduction of 'reverting edit' softens the introduction of 'revert'. The new reader of Wikipedia:Edit warring will see the first sentence of the red box introducing 'revert' as noun and be puzzled by it. The former second sentence, An edit or a series of consecutive edits that undoes other editors' actions—whether in whole or in part—counts as a revert. doesn't resolve that — because it takes a complex construct that combines a singular quantity (An edit) in an logical disjunction (or) with a plural quantity (a series of consecutive edits) and relates it to another complex construct (that undoes other editors' actions—whether in whole or in part) to compose a definition (counts as a revert. My proposal removes the hurdle of that first complex construct and ties it to a term that is a normal english construction using the participle form of 'revert (v.)' as an adjective: (reverting edit). That normal construction is then used as a bridge to complete the definition. Humanengr (talk) 06:09, 24 June 2020 (UTC)


@Zero, right re your #2; I had accidentally dropped that point from MrX; so with your #1, and also correcting which -> that, we have

An editor must not perform more than three reverts on a single page—whether involving the same or different material—within a 24-hour period. An edit or a series of consecutive edits that undoes other editors' actions—whether in whole or in part—counts as a revert reverting edit. A series of consecutively saved edits by one editor that includes reverting edits, with no intervening edits by another editor, counts as one revert. Violations of the rule often attract blocks of at least 24 hours. Fourth reverts just outside the 24-hour period may also be taken as evidence of edit-warring, especially if repeated or combined with other edit-warring behavior. See below for exemptions.
A "page" means any page on Wikipedia, including talk and project space. A "revert" means any edit (or administrative action) that reverses the actions of other editors, in whole or in part, whether involving the same or different material. A series of consecutively saved reverting edits by one user, with no intervening edits by another user, counts as one revert.

Does that work? Humanengr (talk) 14:49, 24 June 2020 (UTC)

Thanks for explaining. I'm still stuck on the fact that we don't state "reverting edit." We simply state "revert." You stated, "Within Wikipedia policy-level pages, it appears as a noun on only 5 other pages — and in all those it also appears as a verb. In contrast, 'revert' appears as verb (and only as verb) on over a dozen others. (It does appear as part of three other WP terms of art, 'three-revert rule, one-revert rule', or 'zero-revert', on this and 4 other pages along with verb use on all but one of those. But those should be considered as compound constructions separately from the isolated 'revert' noun.)" Given the standard use on Wikipedia, I don't grasp your use of "only." I'm not seeing why this page should be the exception when it comes to use of "revert." It would make more sense to propose it more widely, and at WP:Village pump (policy), instead of just for this one page.
You argued, "The use of 'revert' as a noun — and particularly as a noun with the unique definition given here — is indubitably off-putting as a strange construction." You also argued, "The new reader of Wikipedia:Edit warring will see the first sentence of the red box introducing 'revert' as noun and be puzzled by it." I'm just not seeing that. During my long tenure with Wikipedia, this is the first complaint I've seen on the matter. Flyer22 Frozen (talk) 01:43, 25 June 2020 (UTC)
I agree with Flyer22 again. Although newcomers need to learn what a "revert" is, I have never met a case where its use as a noun is the source of difficulty. In addition, I think that defining it in a way that requires understanding of yet another concept (reverting edit) will make it harder and not easier to understand. Overall I do not think that this proposal is an improvement over the existing imperfect text. To editor Humanengr: I suggest you drop it as I see no prospect of consensus. Zerotalk 05:21, 25 June 2020 (UTC)

@Zero0000, I'm dropping that proposal. Flyer22 Frozen's feedback prompted a further rethink that does without 'reverting edit' and simply adds a clarifying footnote after 'consecutive edits':

An editor must not perform more than three reverts on a single page—whether involving the same or different material—within a 24-hour period. An edit or a series of consecutive edits* that undoes other editors' actions—whether in whole or in part—counts as a revert. Violations of the rule often attract blocks of at least 24 hours. Fourth reverts just outside the 24-hour period may also be taken as evidence of edit-warring, especially if repeated or combined with other edit-warring behavior. See below for exemptions.

* That is, with no intervening edits anywhere on that page by another editor.

Iiuc, this conservative interpretation of 'consecutive' is the dominant one. My reading was the more liberal one: 'consecutive' meant that as long as no-one undid -my- edits — rather than edit elsewhere on the page without changing my edits — I was in the clear. (That others have taken this view is seen here and here.) Humanengr (talk) 05:30, 3 July 2020 (UTC)

Your proposed definition of "consecutive" is the meaning that has always been assumed by administrators in my experience. But I don't like footnotes since they will either appear too far away (at the bottom of the page) or will break the page structure. How about another possibility: move the definitions of "revert", "consecutive" and "page" to a second paragraph inside the colored box? Nobody can miss them then. Zerotalk 08:14, 3 July 2020 (UTC)
Thx … I like the idea of putting the definitions inside the box and can mock something up with that. But first, a question: is the including administrative action phrase in the 2nd sentence below the box essential? AFAICS, it was inserted in 2009 and traced it to here, but could not locate rationale. Humanengr (talk) 21:54, 3 July 2020 (UTC)
How are you defining consecutive? And how are you sure that it's the definition that admins generally go by? No need to ping me. Flyer22 Frozen (talk) 05:18, 4 July 2020 (UTC)
What I found to go on is 1) the current text in the para below the box on the article page A series of consecutively saved reverting edits by one user, with no intervening edits by another user, counts as one revert, 2) the first link I gave a few cmts up which goes to a response by Bbb23 in a discussion titled Change to definition of "consecutive edits", and 3) the second link above which goes to a response by Monty845 in a discussion titled Definition of "revert" in which you were involved in 2015. It was from those that I got the sense I wrote as a footnote to the last box above 'no intervening edits anywhere on that page by another editor'. I'm not intending any change to community practice, just a clarification so all can readily understand it. Humanengr (talk) 06:36, 4 July 2020 (UTC)

I can't think of any reason for "administrative action" to be mentioned. An ordinary edit by an admin (like this edit I'm making now) is not an administrative action. Administrative actions like removal of copyvios, page protection, etc, would not be subject to 3RR anyway. Arguments between admins are handled according to WP:WHEEL, not according to this page. Zerotalk 09:23, 4 July 2020 (UTC)

Humanengr, the "Definition of 'revert'" thing makes me want to go back to what I stated above: "Really, given the various debates that have been had about defining a revert (including on this very talk page), I'd rather that the text stay as it is unless having wide support." Flyer22 Frozen (talk) 02:20, 5 July 2020 (UTC)

Notice of discussion

There's a discussion at Wikipedia:Village pump (policy)#WP:1RR: Who dunnit that I think might interest edit warring mavens watching this page. ~Awilley (talk) 15:33, 7 July 2020 (UTC)

Revert rules

I have seen that it is advised that a revert is explained in an edit summary or at the talk page. Some revert without edit summary, and some just put in the same repetitive argument that they have mentioned at the talk page and was refuted and replied with an other question. Couldn't there somehow be included a rule that a revert should be explained with an edit at the talk page? And that the explanation at the talk page is actually "on topic" of the dispute and/or the edit? If someone has the argument it is "sourced", repeats the argument as long as the discussion goes, but it is shown at the talk page and it is also well known that the "sourced" content is wrong, and this is also stated in the article, shouldn't the editor then bring up a secondary source to double up his claim. It would shorten the edit wars significantly. Also guidelines for the ones who revert unexplained during an edit war (it happens, more than some think) could also be introduced. Example of case is this one here an editor (Konli17/From now on Konli) explains his removeâl, an other editor (Ibn Amr al-Kulthoum/ from now on Ibn Amr) here refutes the explanation, here is a clarification by Konli17, and here now is an I don't care answer by Ibn Amr. Konli asks again for an constructive answer on the 30 May, then he asks again on the 17 June. Then here again the sourced argument comes from his first and second revert. Here is the edit historial from the dispute, and here from the discussion. In the end Ibn Amr reported Konli for edit warring and Konli was blocked, which I think was not fair. I have removed the content for now, also started a request at the DNR which failed, and now Ibn Amr has reported me and he still mainly has the sourced argument. And I think new revert guidelines should be inserted in order to deal with such a dispute.Paradise Chronicle (talk) 13:37, 10 July 2020 (UTC)

Maybe the guideline that edits should be explained "on topic" at the talk page should enter after the first reporting at a noticeboard?Paradise Chronicle (talk) 14:04, 10 July 2020 (UTC)
If you think the sources are "wrong" you can bring it to here [1] and mark them as deprecated sources. And after you failed with your arguments, you just deleted large stuff of sources under WP:quotations as new excuses. After i restored encyclopedic material which was not part of our talkpage discussion you just continue your edit war and amr reproted you. Shadow4dark (talk) 14:06, 10 July 2020 (UTC)

Hey look, Ibn Amr didn't answer at all before he managed to block Konli. From the 26 May until the 17 June his only argument was sourced. No admin has stepped into the dispute with arguments so far, not a single one. They only stepped in with blocks, weirdly in favor of the one ignoring the talk page in the first place and mainly arguing it is "sourced" even though it is stated in the article that several of the claims within the quotes in dispute are wrong, while it is of common sense that nothing should have be given back to ISIS, which the quote suggests.Paradise Chronicle (talk) 15:35, 11 July 2020 (UTC)

This page is not the page to argue about a specific incident of edit warring (that's here or here). The initial request makes it difficult to assess whether a change here would be valuable. Consider restating the objective accordingly, otherwise this thread will be ignored. --Izno (talk) 16:13, 11 July 2020 (UTC)
I have been involved in a dispute, where I have noticed some issues, that might could be enhanced in the process. I'd propose, that if someone edit-wars, the reverting editor should give an answer at the talk page and not only in the edit summary. Also, secondary reverts (revert for second time) with no edit summary should be regarded as contrary to the wikipedia guidelines if (it occurs within the time of a semi protection or an other protection of an article a notice could pop up). The notice could be included after an admin was adverted of the issue, or some other mechanism could be found. Of course, reverts of vandalism should be excluded from such rules.Paradise Chronicle (talk) 16:29, 11 July 2020 (UTC)
Also, the revert rules should apply only to the same repeated reason ( or no reason) for the revert. If at the first revert the editor reverts e.g. for content, but the next is because of Wikipedia guidelines, it is a different issue. Some admin "feelings" might be needed for this. Just an idea. Thinking about the issue.Paradise Chronicle (talk) 23:17, 11 July 2020 (UTC)