I tried reverting and I did *not* get a warning. By design? Sebastian 04:38 Jan 25, 2003 (UTC)
ATTENTION!!! Content has been moved from this page. Please see the following pages for extremely important but long, drawn-out discusson, votes and polls which are still in progress.
As proposed in meta:Edit wars: Why not automatically page protect an article for 48 hours when it has been reverted to the same version three times?
Don't be hasty to revert. You can always revert later.
On the "vile mailing list", there's been some talk of making the so-called three revert rule an enforceable rule. I can see two obvious routes to enforcement, each with advantages:
I'm leaning towards the protection option, personally. What do folks think? Martin 23:17, 8 Mar 2004 (UTC)
I'm more inclined to ban anyone who accumulates six reverts of things which are not simple vandalism in one day for 24 hours. That indicates a pattern of reverting which is unlikely to be in a single article and suggests that protection is getting in the way of legitimate contributors. We know that there are comparatively few contributors who habitually choose revert wars rather than discussion and consensus seeking. Better to try to act against those not choosing to act within the community than to protect every page those people choose to fight over. If someone accumulates seven bans for this reason within some moderately short period, that suggests to me that it's arbitration committe time. Jamesday 23:28, 8 Mar 2004 (UTC)
I'm liking the protection option. -- Cyan 22:02, 9 Mar 2004 (UTC)
Temp ban is much better. Page protection hurts everyone. But temp bans are a bad idea too. If a logged in user continually adds various different non-obvious vandalism (but trolling nonetheless), many people would quickly run out of their reverts for the day. Anthony 06:22, 10 Mar 2004 (UTC)
Protection is, of course in my personal view, better for the wikipedia community. Protection encourages discussion, banning disencourages. It's good if the protecting sysop follow the principle of reverting to a page whose supporters are less aggressive, but maybe such a rule ought not be written in stone. There must be space for the sysop to make a sensitive choise - and also for doing mistakes without being heavily criticized.--Ruhrjung 11:46, 10 Mar 2004 (UTC)
Why not combine the two options? Martin's protection option on a per article basis and then have a three-strike rule for users: If a user gets more than three articles protected (the user violates the 3-revert rule) in some time period, then the user gets a 24 hour time out. --mav
What objective evidence do you have that people who revert are the problem? What if the problem is POV writers that need reverting? It makes just as much sense - that is to say little - to tempban people who have more than 3 edits reverted, than the other way around. Stirling Newberry 17:51, 28 Jun 2004 (UTC)
I'm liking mav's idea. Given Jimbo's support, I've edited wikipedia:protection policy. We can add 24hr temp-bans to that, if folks support it (I'm personally abstaining for now). Martin 17:16, 14 Mar 2004 (UTC)
For the record, Anthony DiPierro (as of now) believes that this is not policy. That's something of a meta-issue: the more relevant question for Anthony is - why does he oppose the policy (if indeed he does). Martin 01:08, 16 Mar 2004 (UTC)
Well, page protection does hurt everyone, but the change to protection policy is for cases where we would be protecting anyway. It's not a change to when to protect, only what to protect. So I don't think that argument (while valid) is relevant to the recent change in protection policy.
Yes, the fact that one person is reverting slightly more (or very much more) is not necessarilly relevant to the value of the reversion, but the protection policy is about discouraging revert wars, not evaluating reversions, so I'm not seeing the link here.
Page protection lasts until unprotection, and since any sysop can unprotect at any time, it relies on the judgement of ~200 sysops, not just one. So I think this is just incorrect, unless I misunderstand you.
I understand that you believe that reversions aren't necessarilly a bad thing. The many people in the community who occasionally revert pages, including myself, agree with you. I'm guessing what you mean to say is that repeated reverts (four or more to the same page, on the same day, by the same person) aren't necessarilly a bad thing. Do I understand you correctly on this point? Martin 01:24, 16 Mar 2004 (UTC)
So far it looks like people are reverting more, without discussion on the talk page, as they feel that as long as they outnumber the other side they have no need to compromise and attempt to reach consensus. It's too early to say for sure if that's what's going to happen, but that's my suspicion, and so far evidence is leaning that way. Anthony 04:52, 16 Mar 2004 (UTC)
One immediate practical effect is that it is going to cause a lot more escalation of user issues. Perhaps this is good, there are an awful lot of contributors who don't write NPOV. Before after a revert war, generally NPOV won out. Now with the restrictions and the rule favoring the numerically superior side, it is going to mean a lot more gets kicked upstairs for review. Good, bad ? I don' know. Stirling Newberry 17:48, 28 Jun 2004 (UTC)
Take a look a Jerusalem for an example of this policy being used inconsistently. Anthony 12:01, 17 Mar 2004 (UTC)
I'm currently involved in an edit war, so take my comments from the POV of someone who this potentially applies to. Reversion banning rules favor users with less invested - they "force a draw" by vandalizing or POVing. The sensible user reverts (Strike 1), the Povian reverts (Strike 1) and so on. Meaning the vandal has a first mover advantage. He reverts twice without cost, and the sensible user is faced with a 24 hour ban, or allowing the POV version to stand for some period, at which point the cycle repeats and the POV version stands, again. And so on. Any policy should look at the history of the page and the users to determine what should be done. Page protection for "cooling off" is much better, and allows time for the other mechanisms to run their course. Stirling Newberry 15:36, 27 Jun 2004 (UTC)
My advice would be when a page gets protected, a draft page would be created where the editorial wrangling can continue. Discussion could continue on the talk page and people could edit war till the cows come home on the draft page, but the main article wouldn't get unlocked until somebody reached an agreement. And the page would best be locked in a most recent stable condition, not the most recent condition represented after somebody or somebodies started making uncollaborative changes.
Failure to participate the editorial process on a draft page would result in the page going front without that parties interests included. This would put both parties in a dispute in the position where neither could go forward without winning consent from the other side, and would provide a more workable format in which to work out comprimise.
This site is all about making up rules it can never enforce, based on an assumption that disputes are immutable conflicts, often involving bad people, rather than a result of people with conflicting views being invited to create the same article. - User:172.198.46.33
Is anyone else finding it almost impossible to work out where the new comments are being added to this page? Does anyone care to volunteer to archive and refactor so that it's easier to follow the debate? I got lost about three days ago. And unsigned comments don't help. Tannin 10:24, 12 Mar 2004 (UTC)
Does the 3-revert limit apply based on a 24-hour period, a UTC day, or some other meaning of day? Current practice has been to use a 24-hour period (see User talk:Wik#Arbitration committee ruling). This also makes the most sense, since otherwise people could get in up to 6 reverts in an hour or so and claim not to have violated this rule/guideline. --Michael Snow 21:52, 31 Mar 2004 (UTC)
I only cited a single instance, because it's the most clearcut example, but the current practice extends to far more than just one person. For example, the quickpoll for Lord Kenneth is based on 4 reverts in a 24-hour period, but only 2 in any one UTC day, yet it currently has five voters in favor (while you commendably stick to your principles in your opposing vote).
To the extent that quickpolls make this a rule, we need to provide a clear definition of the time period involved (there is no "just that" meaning of day, as your own uncertainty demonstrates). I believe that is best accomplished by using a 24-hour window. We shouldn't have to rely on the user's time zone, because the information is difficult to prove and not readily available. And using a UTC day would provide an unfair advantage to users who, due to their location or personal schedule, frequently edit from, say, 23:00 - 01:00 UTC (gee, look what time it almost is now). --Michael Snow 22:28, 31 Mar 2004 (UTC)
Well, we probably disagree on how closely it approaches consensus. But obviously we need more opinions here, so maybe we should advertise the discussion a little more.
I don't understand what reverting first or last has to do with it, that's not the issue as I see it. People have to accept that the last revert before protection will probably be the wrong version. My point about the advantage is that if someone edits during the window I suggested (whether the day is UTC or any other time zone), they can revert at 23:25, 23:40, 23:50, 00:05, 00:10, and 00:20 without ever violating the rule. Six reverts in less than an hour - that's a classic instance of "gaming" the rule, which using a 24-hour window prevents. --Michael Snow 22:55, 31 Mar 2004 (UTC)
Is there any reason for it not to apply to a 24-hour period? It saves having to work out timezones, and as this is what the arbitration committee have already applied to Wik, it seems sensible to continue that. Angela. 21:34, Apr 1, 2004 (UTC)
It's not harder to figure out, because if we leave "day" up to people's individual definitions, then it will be utterly impossible to figure out what a day is. And I have no problem "penalizing" somebody who reverts at noon one day and 11 am the next. People who revert frequently in a short period should recognize that they're on thin ice, and edit accordingly. The rule is supposed encourage discussion instead of reverting. --Michael Snow 23:10, 1 Apr 2004 (UTC)
Being able to identify violations is important both for the person doing the reverting, and for other people, as long as any kind of sanction is attached to the guideline, such as quickpolls. Team-reverting is a separate issue, and anyway there is little benefit to it under the rule since once the "day" is over, you would just have to go back and fight the same war all over again. In my opinion, any limitation of reverting inherently encourages discussion. --Michael Snow 00:02, 2 Apr 2004 (UTC)
It is important for the person doing the reverting if that person wants to avoid having sanctions imposed. Martin's advice is an attempt to communicate the spirit of this rule to someone who seemed primarily interested in maximizing the "legal" use of reverts. Limiting reverting encourages discussion because when you exhaust the ability to revert, the only options left are to discuss the problem or walk away. --Michael Snow 00:38, 2 Apr 2004 (UTC)
Someone being banned is a kind of sanction. Protection is another tool to prevent reverting (by technical means for non-admins, and by community pressure for admins). Realizing that endless reversion is futile is a form of walking away. Without developing appropriate tools and policies to limit it, reverting can go on long enough to harm the community (this assumption, not "reverting forever", is what my argument is really about). And reverting is much easier than discussing, so there's an incentive to that as the path of least resistance. If we create disincentives to revert, they will necessarily encourage people to choose other options. --Michael Snow 03:32, 2 Apr 2004 (UTC)
Given that you have already stated your opposition to this guideline, Anthony, it seems obvious that you will also oppose anything that makes it a "tighter" rule. I think we can take that for granted. I'm more interested in knowing if anyone who supports the guideline has a problem with day = any 24 hour period. Martin 23:24, 1 Apr 2004 (UTC)
I'm not opposed to this guideline. In fact, I voted in favor of it. anthony 23:41, 1 Apr 2004 (UTC)
Oh, my apologies Anthony: I saw you'd voted for it being pointless, and didn't realise you'd voted in favour of it despite that. Martin 23:50, 1 Apr 2004 (UTC)
More specifically, I voted that I "think its [sic] pointless to declare an 'end to revert wars' without advocating another means of solving disputes". I don't think the guideline itself is pointless. I think presenting it as a strict rule without having some other way to determine what page to display is pointless. In fact, considering that a strict rule would essentially enforce plurality rule (most people=most reverts), it's perhaps even worse than pointless. anthony 00:00, 2 Apr 2004 (UTC)
Hmm, it only amounts to a plurality for a single day (and even there, it's weighted by motivation). Over the longer term, it reduces to multi-party iterated prisoner's dilemma - not fantastic, but not terrible either.
Anyway, thanks for clarifying for me. Martin 17:30, 2 Apr 2004 (UTC)
(I don't see how it's redundant. It could mean as determined by the reverter.)
Well, if there are less that 24 hours seperating four reverts, there are less than 24 hours seperating four reverts. That's not going to change depending on who is "determining" things. Hence, any 24hr period. Perhaps that's unclear? Martin 21:58, 2 Apr 2004 (UTC)
Having it as 24 hours makes logical sense. Change the policy to "no more than three reverts in any 24 hour period" if the term "day" is the sticking point. --Ben Brockert 23:59, May 3, 2004 (UTC)
Why does this (and so many other policy/civics articles) have a title that is uselessly long? Add to this target links... Wikipedia:How_to_revert_a_page_to_an_earlier_version#Revert_wars_considered_harmful_(the_three_revert_guideline) SeewhatImean? -SV(talk) 02:07, 5 Apr 2004 (UTC) PS. Quicklink - WP:3RR -Should be Wikipedia:Revert > Wikipedia:Reversion
Could somebody please explain to me why a rollback by a custodian is marked minor by the software? I find that they are seldom minor edits. Thanks. moink 00:01, 27 Apr 2004 (UTC)
When a banned user comes back with vengeance, on multiple IPs, it seems clear that people should ignore the 3-rv rule (and simultaneously work to block an aggressive vandal). There may be only ten interested parties online, and a vandal can easily blank a page thirty or fifty times in a day. Some people rv'ing vogelisms today apparently (iir[#irc]c) stopped after three per page, perhaps an unfortunate side effect of this guideline. I added two relevant sentences to section 4. +sj+ 19:45, 2004 May 10 (UTC)
A user asked me to roll back a page that had undergone several vandalism-reversion cycles, because the vandal had included rude remarks in the edit summaries. That is, he wanted the offensive edit summaries removed. I couldn't figure out how to do it. For the time being, I moved the page to an inconspicuous place, deleted the original page, then re-created it with the same contents. Advice welcome. Sorry if I missed something obvious. (The "rollback" function simply appears to be a convenient way to revert, and preserves all previous page history). Dpbsmith 22:28, 26 May 2004 (UTC)
The way I understand things now, sysops can quickly rollback a page by clicking a link on the page history, whereas anyone—even anons—can achieve the same effect by editing an old version of the page. Is there any reason for this difference? It seems to me that if anyone can revert a page, there's no point in making non-sysops go through an extra couple of clicks and pageloads (which can be a real pain on a slow server day). Does the rollback link do something sysop-only that I'm not aware of? If not, can the rest of us have the link? Thanks for reading. -Etaoin 05:49, 27 May 2004 (UTC)
I've been going through recent changes (live) a little lately, and frequently have to revert vandalism. I find that there are too many steps involved:
If there were an "edit this version (for reversion purposes only!)" button in the diff-viewing page, it'd help:
I propose that. (In fact, if there were a "revert to this version with summary ____" form on the diff page, it'd be even fewer steps:
— but I'm not proposaing that. I think it would be too easily abused and take up too much room on the page.) What think you all? —msh210 14:11, 7 Apr 2005 (UTC)
I see that many users reverting newbie tests use the edit summary "rv vandalism". Is this the proper summary or should something "kinder", such as "rv testing" or just "rv" be used? After all, newbie tests are specifically listed under what Vandalism is not. With that said, RC-patrollers are making very valuable contributions! Sjakkalle 08:43, 29 Apr 2005 (UTC)
The article says Reverts never cause an edit conflict. And then goes on to say that they can. Or do I get something wrong? DirkvdM July 8, 2005 17:46 (UTC)
And another thing. I suppose that a revert only works for the last edit unless you're ok with reverting all the edits that follow it as well. That sounds obvious, but maybe it should be mentioned. I just experienced something like this in the Cuba article; I wanted to revert several edits except one (or two), so I just made notes of the ones I wanted to keep and then did those edits again. DirkvdM July 9, 2005 07:50 (UTC)
Wikipedia:Administrators' how-to guide says that there's a rollback link on the article history page for admins. Well, I'm an admin, and I can't find one. There's one on the diff page, and on the user contributions page, but not on the history page. Having one at the history page would speed up revertions. — JIP | Talk 16:35, 13 September 2005 (UTC)
I've noticed one issue surrounding page reverts that I find annoying.
Several times, I've reverted obvious vandalism—but with an unexpected result. This has happened when I was reverting to a version that was credited to me. The article reverts to the last edit made by someone else other than me.
I can somewhat understand why this is, assuming it's Wikipedia policy... but it makes it VERY inconvenient to revert a page when I know I made the last helpful edit.
Any comments? – Dale Arnett 15:57, 26 September 2005 (UTC)
How do I do such thing? I revert a lot of vandalism, but sometimes, I see people who use that line in their summary. How can I do this? Is this even possible to do automatically, or is it manual? Sorry, I'm still quite new to Wikipedia. Could someone maybe put it in my talk in case I don't get back to this page? Thanks -Tcwd (talk) 21:41, 13 October 2005 (UTC)
Increasingly, I see reverts not for any problem in the improvements I contributed, but with the given reason that my little essay explaining them insufficient to placate the reverter. The essay appears only in the History section. Our mission with Wikipedia is a world-beating encyclopedia, not a world-beating set of edit descriptions.
Part of the burgeoning problem may be a snippet that I saw somewhere in Wikipedia's instructions incautiously encouraging reverting changes as a way to get discussions started. This section should be clarified to indicate that the reasons for the revert should be stated in full -- i.e., giving the reasons why the original article text was deemed superior. What I'm seeing instead is demands to make the little essay explaining the changes (which are in any event self-explanatory) a better essay, without saying what was lacking with it in the first place (my little essays often take up much or all of the space allowed in the little "Edit summary" field as it is).
The proliferation of these "Revert Trolls" diminishes my enthusiasm for contributing to Wikipedia. An effective mechanism for identifying them and shutting them down or educating them about the inadvisability of failing to follow the reverting guidelines is needed.