Wikipedia talk:Edit warring/Archives/2016/June
![]() | This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Did I do one revert or two?
At Norman Finkelstein I deleted[1] some material that had been in the article since at least 2012[2] with various wording. It was restored and I then reverted the editor restoring it. At Talk:Norman Finkelstein#Edit Wars on Palestinian Occupation I'm being accused of making two reverts and only getting away with it because I'm not pro-Palestinian (and that if I were the WMF (sic) would have blocked me), which is ridiculous. I do sympathise a bit with what I think is the mistaken belief that I reverted twice - I recall bringing similar issues here as 3RR violations to be told that the first edit wasn't a revert, so I've gotten that wrong at times. There doesn't seem to be any clear statement about this. Should there be? Or am I wrong and just working on bad advice? Doug Weller talk 18:07, 17 May 2016 (UTC)
- My opinion of long standing: two reverts. The first edit is a removal of material, which is an undoing of someone else's action and therefore fits the definition of a revert. The only time I don't count the first edit is when it is adding brand new material, although I sometimes don't count the first edit if it is insubstantial. A certain amount of administrator discretion is called for. BTW, this subject has come up many times before, and some administrators disagree with my position. What a shock.--Bbb23 (talk) 18:19, 17 May 2016 (UTC)
- But it's undoing several editors actions, not one. That's often the case if you remove a section. But my main concern is that this hasn't been clarified and thus editors aren't being treated consistently. Do we need an RfC? Doug Weller talk 18:30, 17 May 2016 (UTC)
- Thinking again, what you are saying is that removal of any text is considered a revert. Then policy should say that. Doug Weller talk 18:36, 17 May 2016 (UTC)
- It's not what I'm saying, and I've been through this so many times before it wears me out, so after this comment, I'll drop out. Editors are treated inconsistently by administrators all the time. As long as an administrator's decision isn't out of process, such inconsistencies are acceptable. Making it too iron-clad eliminates discretion to evaluate cases on a case-by-case basis, which benefits the project.--Bbb23 (talk) 18:39, 17 May 2016 (UTC)
- That makes sense. Context is always important. Doug Weller talk 18:51, 17 May 2016 (UTC)
- But I still think of reverts as actively reverting a specific editor. "To revert is to undo the action of another editor." - does that actually mean editors plural? If I remove a paragraph that has been worked on by several editors is that a revert? And with 1RR being more common on articles, maybe we do need to offer better guidance. Doug Weller talk 20:51, 17 May 2016 (UTC)
- Your problem is you keep thinking. Don't you know how dangerous that is? If you take the revert definition literally, undoing the action of multiple editors necessarily means that you're undoing the action of each one. Still, back to some common sense. If the issue is 1RR, which is most common on pages subject to arbitration sanctions, I personally would be less likely to block without two clear reverts and not necessarily count the first one. It also depends on how long the material that the first revert is removing has been there and whether it is related to the subject of the dispute. Which brings me back to discretion and your word: context.--Bbb23 (talk) 20:58, 17 May 2016 (UTC)
- But I still think of reverts as actively reverting a specific editor. "To revert is to undo the action of another editor." - does that actually mean editors plural? If I remove a paragraph that has been worked on by several editors is that a revert? And with 1RR being more common on articles, maybe we do need to offer better guidance. Doug Weller talk 20:51, 17 May 2016 (UTC)
- That makes sense. Context is always important. Doug Weller talk 18:51, 17 May 2016 (UTC)
- It's not what I'm saying, and I've been through this so many times before it wears me out, so after this comment, I'll drop out. Editors are treated inconsistently by administrators all the time. As long as an administrator's decision isn't out of process, such inconsistencies are acceptable. Making it too iron-clad eliminates discretion to evaluate cases on a case-by-case basis, which benefits the project.--Bbb23 (talk) 18:39, 17 May 2016 (UTC)
- Thinking again, what you are saying is that removal of any text is considered a revert. Then policy should say that. Doug Weller talk 18:36, 17 May 2016 (UTC)
- The 1RR rule is intended to catch early-stage edit wars and flag them for admin attention. I suppose we could tweak the definition of 1RR to make it more specific. Here is a suggestion:
- To break 1RR, you need to have made at least two edits of the article within 24 hours.
- Any edit that is marked as 'Revert' or 'Undo' in the edit summary is a revert
- Anything that undoes, wholly or in part, a recent contribution of another editor is a revert, where 'recent' is within seven days
- The edits don't have to be of the same material to count toward the total of two
- Two edits within 24 hours that *are* about the same material are a 1RR violation if the #2 edit restores the article to the same version as the #1 edit, in spite of one or more intervening edits by others.
- If this version of 1RR became accepted then the Norman Finkelstein case described by Doug above would in fact be a 1RR violation. EdJohnston (talk) 19:43, 19 May 2016 (UTC)
- Putting aside the precise particulars of the proposal, the initial problem I have is we have always said that 1RR and 3RR are the same except for the number. Are we changing only 1RR, or are we changing both?--Bbb23 (talk) 19:56, 19 May 2016 (UTC)
- It would get too complex if we tried to change 3RR to match this proposal. I suggest that (if this is needed at all) we only modify 1RR. Whatever is adopted should be easy to understand, so that people don't have so many questions about whether 1RR was truly violated. EdJohnston (talk) 21:33, 19 May 2016 (UTC)
- You'd have to justify changing one and not the other, and even if you did, it wouldn't stop there. There'd be a push to change 3RR to match. Finally, even if changing 3RR at the same time is complex, enforcement of two different rules would have its own set of problems.--Bbb23 (talk) 21:48, 19 May 2016 (UTC)
- Did I do one revert or two or three? I count one revert, yet I get this Warning: Violating the three-revert rule. How does this tally with the above?? Bosley John Bosley (talk) 21:57, 19 May 2016 (UTC)
- @Bosley John Bosley: That warning was created using WP:Twinkle, and is intended to make users aware of 3RR, not necessarily accuse them of "violating" it. I have previously suggested that the edit summary should be changed. I have linked the above comment there. Burninthruthesky (talk) 09:55, 1 June 2016 (UTC)
- Did I do one revert or two or three? I count one revert, yet I get this Warning: Violating the three-revert rule. How does this tally with the above?? Bosley John Bosley (talk) 21:57, 19 May 2016 (UTC)
- You'd have to justify changing one and not the other, and even if you did, it wouldn't stop there. There'd be a push to change 3RR to match. Finally, even if changing 3RR at the same time is complex, enforcement of two different rules would have its own set of problems.--Bbb23 (talk) 21:48, 19 May 2016 (UTC)
- It would get too complex if we tried to change 3RR to match this proposal. I suggest that (if this is needed at all) we only modify 1RR. Whatever is adopted should be easy to understand, so that people don't have so many questions about whether 1RR was truly violated. EdJohnston (talk) 21:33, 19 May 2016 (UTC)
- Putting aside the precise particulars of the proposal, the initial problem I have is we have always said that 1RR and 3RR are the same except for the number. Are we changing only 1RR, or are we changing both?--Bbb23 (talk) 19:56, 19 May 2016 (UTC)
- But it's undoing several editors actions, not one. That's often the case if you remove a section. But my main concern is that this hasn't been clarified and thus editors aren't being treated consistently. Do we need an RfC? Doug Weller talk 18:30, 17 May 2016 (UTC)
Ban aspect of "3RR exemptions" section
Part of the "3RR exemptions" section states, "Reverting actions performed by banned users in violation of their ban, and sockpuppets of banned or blocked users."
With this edit Od Mishehu stated "this applies to TBANs and IBANs, but only when the edit was a ban violation" and added "in violation of their ban." I don't see how that text makes it clear that the ban aspect applies to WP:Topic bans and WP:Interaction bans, so my issue was more so the rationale that Od Mishehu was using. As this link shows, I reverted, stating, "Huh? That material is about editors who are banned from Wikipedia. It is not narrowed to topic bans and interaction bans." And Od Mishehu replied, "Actually, the information is about both. Any edit by a site-banned user is a violation; as is an edit by a topic banned user on that topic, but not on other topics."
My issue is that when it comes to a one-way interaction ban, for example, can we really state that reverting the editor is an automatic exemption to edit warring? If the editor who does not have an interaction ban is the one doing harm, I do not see that this editor should be exempt when it comes to the WP:3RR policy. Same goes for a topic ban matter. WP:BANEX, which notes exemptions to a ban, is issue enough for editors. Generally, the 3RR exemption with regard to bans has applied to those who are site-banned. Flyer22 Reborn (talk) 04:27, 3 June 2016 (UTC)
Note for clarification here. Flyer22 Reborn (talk) 04:31, 3 June 2016 (UTC)
- Hi
- It seems quite clear to me that that the action a banned person, when within the purview of WP:BANEX, isn't a violation of his ban. So, you seem to have provided the solution for your issue. From where I am standing, Od Mishehu's contribution seems a good one that actually prevents automatic reversion of a banned person's action when that action is justifiable.
- Best regards,
- Codename Lisa (talk) 09:37, 3 June 2016 (UTC)
- The examption from reverting related to banned users applies to edits by a topic-banned user when they violate the ban, but not otherwise. If Strangesad edits an article on Christianity, this is a violation of his/her ban, and reverting the edit is exempt from 3RR; if (s)he edits an article about weather, it isn't a ban violation and reverting the edit wouldn't be exempt from 3RR. If a user is site-banned, then any edit they do anywhere on the site is a ban violation. עוד מישהו Od Mishehu 11:10, 3 June 2016 (UTC)
- Codename Lisa, what do you mean by "It seems quite clear to me that that the action a banned person, when within the purview of WP:BANEX, is a violation of his ban."?
- Od Mishehu, I was stating that the 3RR exemption with regard to bans has generally applied to those who are site-banned. Above, I was trying to clarify that your rationale puzzled me. I noted that with this edit, where I stated, "It seemed to me that you were focusing on the less serious bans when that bit is generally used for site bans." Your edit has attempted to extend the exemption portion of the policy to topic bans and interaction bans, but I do not see how your addition of "in violation of their ban" helps extend the policy to cover that. One could state that you are not extending the policy, but rather clarifying it. Either way, I was stating that I have not seen the exemption used for topic bans or interaction bans. In the case of a one-way interaction ban, for example, I have not seen edit warring and then the 3RR exemption card used as a pardon. I do think it can be a good thing for the exemption portion to cover topic bans or interaction bans, but I also think it can be a bad thing in the case of those who WP:Game the system. Flyer22 Reborn (talk) 05:26, 4 June 2016 (UTC)
- Silly mistake on my part! Fixed. —Codename Lisa (talk) 06:59, 4 June 2016 (UTC)
Time periods on WP:1RR
I've made a slight change to the text in this section, so it now reads "sometimes the phrase "24-hour period" is replaced by some other time period, such as "1 week"" (my additions in bold). The specific motivation for this is this unusual, but community-endorsed editing restriction which mandates no more than one revert per page per month, but I think not hard-qualifying "1 week" does accommodate these sort of edges cases more generally. Ritchie333 (talk) (cont) 11:23, 7 June 2016 (UTC)
- @Ritchie333: Although I left your change in place, I'm not certain we want to embody this kind of unusual restriction imposed on specific editors. My main concern is that it is misleading as to what 1RR has traditionally meant and continues to be meant in the vast majority of cases (don't know the statistics on the "exception" you added). The community's decision is not really a change to the policy but a home-grown sanction based on the conduct of a particular user.--Bbb23 (talk) 11:42, 7 June 2016 (UTC)
The latest example of the insanity of 1R of any kind per editor per article per day
Here and here. We cannot punish good faith, on either side of that coin. 🖖ATS / Talk 02:19, 13 June 2016 (UTC)
- Completely agree. Reverts tend to fall into three categories: Clearly righteous, clearly detrimental to the encyclopedia, and content-related differences of opinion. Current 1RR (and 3RR) policy treats them all the same, but they couldn't be more different. The main problem is the lack of workable alternatives, short of having an admin referee present at all times at every contentious article (which is not feasible and therefore not workable).
But the place for this discussion is WP:VPP, not here. ―Mandruss ☎ 02:39, 13 June 2016 (UTC)- Much obliged, Mandruss. 🖖ATS / Talk 02:51, 13 June 2016 (UTC)