Wikipedia talk:Three-revert rule/Archive 6

Regarding people that 3RR in just over 24 hours
Is the 24 hours a hard exact limit? If someone reverts three times within 24 hours+5 minutes, then they don't violate the policy? —Zachary talk 17:42, 8 January 2008 (UTC)


 * I suppose one hasn't technically violated 3RR, but the application is not precise. If someone's edit warring they can get blocked short of 3RR anyway, and that's more likely to happen if there are other signs of trouble like contentiousness or wikigaming.  4 reverts in 24 hours + 5 minutes would appear on the surface to be gaming the system.  Wikidemo (talk) 22:24, 8 January 2008 (UTC)

That's why bright line rules suck. Of course, no one actually ever thought to do away with the forerunner to this rule. --Kim Bruning (talk) 02:11, 10 January 2008 (UTC)


 * Editor review??? MB83 (talk) 04:51, 13 January 2008 (UTC)


 * Kim means WP:EW, or Edit war. To answer Zachary's original question, it's worth considering the history of this page. Back in the days of yore, it was not uncommon to see edit wars filling whole screens worth of page histories. Edit warring was always considered to be a bad thing, but sometimes it was hard to say just whether someone had been edit warring or not. Sometimes even people who were quite obviously edit warring would take advantage of the general nature of the prohibitions on edit warring by wikilawyering. So in 2004 the three-revert rule was invented, to be a bright-line rule that would cover the really obvious cases.
 * What happened after that unfortunately is that people tended to forget that you could still be edit warring without actually breaching the three-revert rule, for example by making three or fewer reverts a day for several days, or reverting across a number of pages, and so on. Many types of edit warring will not be so immediately obvious as making more than three reverts in 24 hours, but they're still not acceptable. --bainer (talk) 05:07, 13 January 2008 (UTC)
 * Exactly as Thebainer says. In addition, people are often blocked for near-violations of the 3RR like that.  Mango juice talk 17:54, 17 January 2008 (UTC)

Bright line rules can be very helpful; if 25 hours are too few, why not just change the rule to 3 reverts in 3 days?TVC 15 (talk) 03:28, 2 October 2008 (UTC)


 * No one objected to my suggestion above, and it's been nearly a week, but I really hesitate to try to edit a rule without more support (or even objection) from others with more experience. The policy on edit warring says, "Deliberate engagement in edit warring instead of discussion is a breach of Wikiquette and may cause user blocks from editing."  The earlier comments above seem to assume that the person making the reverts is acting in "bad faith" and refusing to discuss, but in fact I recently had the opposite situation: one side (two people) reverted four times while refusing my requests to discuss .  I kept inviting discussion until I got blocked for complying with 3RR.  Once burned, twice shy, I will hesitate to go near 3RR without an exorcist to protect me from its "spirit."  Nevertheless, I don't appreciate being branded with the scarlet letter of "bad faith" when in fact I was trying to encourage discussion and follow the rule.  If you file your taxes on April 15, even though maybe you could have filed on March 20, the IRS doesn't accuse you of "bad faith" and send you to Guantanamo for 24 hours to protect America from your selfishness.  (Maybe that's next, on the rationale that your tax dollars are needed asap to support the patriotic spirit of another war.)  If the sign says "school zone, 25 mph" and you drive 25, they don't arrest you for violating the "spirit" of driving slowly.  (Maybe somewhere they do, "Hey look at the bad guy, driving as fast as he's allowed to go, better suspend his license for a day to protect the community!")  Blocking people who don't even violate the rule is like arresting someone for going 17 in a 25 zone: what's the point of a 25 limit if Officer Rhetorical's radar gun only goes to 19 and so he suspends the license of anyone anyone driving 20?  WP is a collaborative effort of volunteers and trying to follow the rules shouldn't result in being blocked and stigmatized.TVC 15 (talk) 07:38, 8 October 2008 (UTC)
 * Well, 3RR is really only one way to measure edit warring. That's the real rule: don't edit war. We can't make enough bright-line rules to cover all possible edit warring situations; they'll always be gamed. So we make one and use it fairly often, while keeping the not-a-bright-line rule in use as well. It's not really correct to compare this with tax deadlines and speed limits for this reason. (Comparing Wikipedia policies to real-world law is in general a shaky proposition.) Heimstern Läufer (talk) 08:04, 8 October 2008 (UTC)


 * Thanks for the additional context. Perhaps the better solution would be to make the 3RR definition track the definition of edit warring, i.e. reverting three times (in one day or three) "instead of discussion."  That would better match the proffered spirit, i.e. promoting discussion instead of brute force.  It might also better approach the situation described above, where one side (two people) reverted four times while expressly refusing to discuss.  To the extent reasonably possible, a policy should say what it means, and mean what it says; otherwise it's a "gotcha" trap, where well-intentioned people can fall victim to unseen spirits and goblins.TVC 15 (talk) 21:29, 8 October 2008 (UTC)

reverts of vandalism
What happens if discussion page content is removed by a registered member who is intent on vandalising my comments by interspersing his own into my text? If I revert, do I breach 3RRs? Should I report him to admin? At first I neatly separated his comments from mine and pasted them below, but he was intent on removing my comments entirely, and replaced them with fairly bad sarcasm.--mrg3105mrg3105 11:40, 12 January 2008 (UTC)


 * If someone persists in interfering with talk page discussion, WP:ANI is probably the place to go. MB83 (talk) 04:50, 13 January 2008 (UTC)


 * I'm not sure mrg is referring to talk page comments. It is reasonable to revert someone modifying your comments once, but it's a dumb thing to get into an edit war over.  It's usually possible to sidestep the issue: i.e. instead of undoing the changes you can leave a new message saying your previous message was altered over your objections, and link to the diff.  Mango juice talk 17:56, 17 January 2008 (UTC)
 * It seems to me that editing someone else's comments should be subjected to a 1RR rule. Unless it's done in response to a violation of Wiki policy, changing someone else's words is acceptable only when it's done it a good faith belief that the other user would be okay with it; once that other editor reverts you, that GFB is gone.Heqwm (talk) 08:39, 19 January 2008 (UTC)

Responding to specific concerns
Suppose that an edit summary cites a specific concern as justification for a revert. Is addressing that concern a separate edit? For instance, suppose I add a fact, another editor deletes it claiming that the reason for the revert is lack of citation, and I add the fact back in, this time with the requested cite. Does my second edit count as a separate revert? It seems to me that both the letter and the spirit of the rule say "no".Heqwm (talk) 08:48, 19 January 2008 (UTC)


 * I looked though the archives a few days ago, and I saw many variations of this question. None of them were answered. Per the letter of the rule, it would seem that replacing a  tag with the citation it requests would count as a revert.  3RR Administraors? MB83 (talk) 05:56, 21 January 2008 (UTC)


 * The question is, is there an edit war? I would count the following as a violation:  A adds fact without reference.  B reverts.  A re-adds fact with reference.  B reverts.  A reverts.  B reverts.  A reverts.  C reverts.  A reverts.  Here, A only "reverted" 3 times, not more than 3, but the re-adding of the fact the first time is still the undoing of another user's edit.  Note the presence of C here; if B had done it, B would also have broken the 3RR.  That kind of situation looks like an edit war.  I wouldn't normally hesitate to readd something with a reference if it was removed for not being referenced, but if I was in an edit war I would be cautious; probably better to discuss in that situation.  Mango juice talk 16:19, 21 January 2008 (UTC)


 * I think the question is more along the lines of: A adds fact without reference. B reverts.  A reverts.  B reverts.  A reverts. B reverts.  A reverts.  C adds a fact tag.  A replaces the tag with a source. A has now undone C's addition of the tag, but it's A's fourth revert, technically. MilesAgain (talk) 10:14, 22 January 2008 (UTC)


 * Nobody would block A is their last edit was to add a reference. This follows from common sense, which is more important than the letter of the policy. &mdash; Carl (CBM · talk) 13:03, 22 January 2008 (UTC)


 * Replacing a fact tag with a source isn't a revert so I certainly wouldn't block for that. If C had reverted, saying the item was unsourced (instead of tagging it), and A returned the item with a reference, I would have to look carefully at the situation.  If, for instance, C reported A for 3RR, I would probably block, because if C perceives a problem with A's behavior, there's a problem: C may have objected to more than sourcing... or A's source may have been inadequate.  In A's place, I would advise against that kind of action; after B's multiple reverts and A's multiple reverts, it's time to go to discussion.  Mango juice talk 15:38, 22 January 2008 (UTC)

template reinsertion
Is the reinsertion of obviously applicable tags like these after they are repeatedly removed without discussion included in Three-revert rule of "simple and obvious vandalism?"-- brew crewer  (yada, yada) 19:21, 25 January 2008 (UTC)
 * No; it's not vandalism you're reverting, it's edit warring. That said, the other editor there needs to explain themselves.  I'll leave them a note.  Mango juice talk 20:11, 25 January 2008 (UTC)
 * If you are correct, then I'm also guilty of a 3rr. Oh well. -- brew crewer  (yada, yada) 20:27, 25 January 2008 (UTC)


 * If I insert a template disputing the title, and it is removed three times, no title can ever be disputed without breaching the 3RR. Is my only option to initiate a dispute resolution?--mrg3105 (comms) If you're not taking any flak, you're not over the target. 02:56, 19 February 2008 (UTC)

Why is it 3RR instead of, say, 4RR?
Let's say an article is in version A (original) and version B (controversial new version). A typical sequence of edits might be:
 * Article changed from version A => B (no reversions)
 * Article reverted from version B => A (revert #1 editor for A)
 * Article reverted from version A => B (revert #1 editor for B)
 * Article reverted from version B => A (revert #2 editor for A)
 * Article reverted from version A => B (revert #2 editor for B)
 * Article reverted from version B => A (revert #3 editor for A)
 * Article reverted from version A => B (revert #3 editor for B)

As we can see from this sequence, the editor supporting the original version is always the one who ends up getting punished by the WP:3RR rule. In fact, for any odd number, the one restoring an article to a balanced and stable version will always end up violating the WP:3RR rule before the editor who pushes a new and highly controversial version; whereas, for any even number, the editor who makes a change will hit the number of reversions before the person who tries to maintain the original version. Why was an odd number chosen for the rule? ← Michael Safyan (talk) 05:10, 28 January 2008 (UTC)
 * Umm...it'll be the same with an even number. Basically, whoever makes the first revert will hit the number of allowed revisions first.Bless sins (talk) 05:13, 28 January 2008 (UTC)
 * Actually, I am not sure why 3 was the magic number; I am sure it comes from early talks with Jimbo, though. However I remember an attempt to make it 4 instead of 3. We have a WP:1RR, too. In any case, the reason of 3RR is not to use all your reverts, but instead give you time enough to realize you are edit warring. If I edit and someone reverts me, it means someone does not agree with me. If I revert him, I revert knowing that someone is not agreeing with me and yet I want to make my position the definitive regardless of his opinion. The second revert means someone greatly objects my change, or that there are at least two users not agreeing with me. If by that time you haven't yet realized you should stop and discuss, you are not likely to stop. Blocking is not punitive, but preventive, and if you haven't realized you have people who is not agreeing with you, the rule allows admins to block someone since he is not likely to stop after he has done a third revert. -- ReyBrujo (talk) 05:27, 28 January 2008 (UTC)
 * And by the way, no matter which page you protect or is reverted at, it is always The Wrong Version. -- ReyBrujo (talk) 05:28, 28 January 2008 (UTC)
 * (ec) This is why I don't believe in mechanical enforcement of the 3RR. In cases like this, where both users are edit warring, I feel admins should block both or neither, regardless of whether one has exceeded the arbitrary limit of three in one day (possible exceptions if other factors are in play, such as if one has a long history of edit warring or is being uncivil). Heimstern Läufer (talk) 05:30, 28 January 2008 (UTC)

Suggested exception to 3RR: grammar and spelling
Does it seem reasonable to suggest an exception to the 3RR rule where the edit in question relates solely to the correction of poor grammar or incorrect spelling?

For example, assume that a User had already made his or her three reverts or changes and had been intentionally amending the content of an article (ie. changes which would indeed count towards 3RR). The User then makes a fourth change during the 24 hour period but solely to fix spelling or grammatical errors introduced by other editors and which does not seek to perpetuate the direction of the earlier three edits. Should this fall within an exception?

If this were acceptable, I would not intend it to apply in circumstances where a discussion of grammar or spelling itself was the focus of the article in question, such as in English Grammar or English spelling reform; but rather in all other cases where the correction of the spelling or grammar improves the readability of the article without changing its direction. Kind regards-- Cala braxthis (talk) 07:29, 1 February 2008 (UTC)
 * But then wouldn't we need to add poor grammar and incorrect spelling as forms of vandalism? If someone comes and adds nonsense, and you revert three times, and then someone comes and adds a new paragraph that is not understandable, really bad written or does not add anything to the article and you revert again, I don't see anything wrong with that. If it is not salvageable, it should not stay in the article. However, sometimes they are useful and should be rewritten instead. -- ReyBrujo (talk) 12:21, 1 February 2008 (UTC)
 * Thanks for your thoughts. Obviously vandalism connotes an element of bad faith which is lacking in cases of poor spelling and grammar.  But essentially the idea would be if an editor corrected the latter (ie. rewrote the text rather than deleted it), there should be a carve-out from the 3RR rules.  Perhaps this would only arise in very unusual circumstances, but it is cases which are at the margin where issues of fairness are felt most keenly.  Kind regards-- Cala braxthis (talk) 16:53, 1 February 2008 (UTC)
 * It says more than three reverts, not more than three edits. The fourth edit in the situation you describe would only be a revert if someone had intentionally changed correct spelling to incorrect spelling, which would be vandalism -- Gurch (talk) 16:41, 27 August 2008 (UTC)

Sandbox
Does the 3RR apply to the sandbox as well? —Preceding unsigned comment added by Red45aqua (talk • contribs) 20:30, 1 February 2008 (UTC)
 * Do people edit war at the Sandbox? If so, Lord help Wikipedia. Well, I don't know that the Sandbox has any exemptions beyond the usual ones. Is there a specific situation you're wondering about? Heimstern Läufer (talk) 21:13, 1 February 2008 (UTC)

No I was just wondering If I could pratice many new things with editing In there as much as I want.Red45aqua (talk) 21:28, 1 February 2008 (UTC)
 * Yes, because practicing editing is not reverting, and therefore not covered by this rule. Heimstern Läufer (talk) 21:29, 1 February 2008 (UTC)

And people laughed when I proclaimed my intent to engage in Sandbox Patrol (see #2) at my RfA. They laughed, I tell you. Well who's laughing now? BWA HA HA HA HA HA HA Ronnotel (talk) 13:42, 20 February 2008 (UTC)

Reverting edits by anonymous users applies to this rule?
I was wondering if reverting more than three edits made by the same IP address and it's not a clear vandalism will broke this rule. Tasc0 It's a zero! 01:58, 14 February 2008 (UTC)
 * I would guess in most cases, yes, unless the material added is unsourced and violates WP:BLP. Both you and the IP are bound by 3RR, so your best bet would probably be to wait a day and revisit it later.  Often the anon user won't return, or another editor watching the article will revert the information before that.  &mdash;To rc.  ( Ta lk.  ) 02:10, 14 February 2008 (UTC)
 * Thanks for replying. But that information it is not specified in the rule page. Or atleast I haven't see it. Tasc0 It's a zero! 02:13, 14 February 2008 (UTC)
 * What information is not specified? The exceptions?  Or that IPs are not exempted from 3RR?  &mdash;To rc.  ( Ta lk.  ) 02:16, 14 February 2008 (UTC)
 * That reverting more than three times edits by a single IP address will not break the rule. <b style="color:#002BB8;">Tasc0</b> <sup style="color:#002BB8;">It's a zero! 02:28, 14 February 2008 (UTC)
 * No, it will break the rule. If you revert an anon IP three times, that's a 3RR violation.  If an anon IP reverts you three times, that's a 3RR violation on them.  Anons and registered users have the same set of rules; it says as much in the very first line of the article.  &mdash;To rc.  ( Ta lk.  ) 02:41, 14 February 2008 (UTC)
 * But it doesn't specify that about IP users. <b style="color:#002BB8;">Tasc0</b> <sup style="color:#002BB8;">It's a zero! 02:58, 14 February 2008 (UTC)
 * "The three-revert rule (often referred to as 3RR) is a policy that applies to all Wikipedians." There's no reason to call out IPs more than registered accounts - they're already covered.  &mdash;To rc.  ( Ta lk.  ) 03:40, 14 February 2008 (UTC)
 * Don't worry about what anyone else is doing, worry about what you are doing. More than three reverts in 24 hours is breaking the rule. Maybe someone else broke it too, but that's irrelevant in answering the question of whether you broke it. --bainer (talk) 03:13, 14 February 2008 (UTC)

Reverting edits made by anonymous users does not make those reverts an "exception" - IP editors are not second-class citizens. If you find an article under assault of inappropriate changes made by IP editors, request semi-protection at WP:RFP. Mango juice talk 18:00, 14 February 2008 (UTC)


 * There might be a little confusion about the meaning of "edit". If an IP did eight different edits in a row, reverting to the version before that chain of edits counts as 1 reversion and not 8.  You'd have a tally of 1 reversion and the IP 0.  If the IP reverts your change then the tally is 1 yours and 1 IP reversion.  -- SEWilco (talk) 19:14, 22 February 2008 (UTC)

24-hour rule no longer applies?
I recently placed a on the Chişinău article. Although "An editor must not perform more than three reverts, in whole or in part, on a single page within a 24-hour period. A revert means undoing the actions of another editor, whether involving the same or different material each time.", and I did not perform template restores more then 3 times in 24 hours, I was never the less told that I am being disruptive by an administrator in Wikiquette alerts. I thought this was strange, so was wondering if I can have a clarification on this.
 * PS. Is reverting of the same edit by three different people countering the original editor also a subject of the 3RR?--mrg3105 (comms) If you're not taking any flak, you're not over the target. 09:48, 20 February 2008 (UTC)


 * Read the policy. The three-revert rule doesn't establish that users have the right to revert up to 3 times in a day: rather, it makes it clear that more than that is unacceptable.  That doesn't mean the fewer is acceptable, and in fact it is very often not acceptable, especially if the pattern continues over time.  If you are reverting frequently you are nonetheless edit warring, and a block would be appropriate if you seem to have no intention of stopping.  And no, 3RR does not at all apply to a group of separate editors.  Part of the point of that is so that a lone editor who dissents can't hold the article hostage.  Mango juice talk 19:21, 20 February 2008 (UTC)


 * The policy is weasel worded. Either "three" means "three", in which case admins are not allowed to play mind readers as this one did and ascribed intent which mrg3105 denied, or it doesn't mean "three" (which actually seems to be the reality) and any admin who doesn't like you for any reason from racism based on your name to how well you spell can block you for making any change to any article just by hanging the label "disruptive" on you.


 * If "three" doesn't really mean "three", then give the rule another name instead of being Orwellian about it. -- Davidkevin (talk) 06:58, 16 October 2008 (UTC)

Reverting pure vandalism
Lets say a user vandalises a page. I revert it and warn them (1). They vandalise it again. I revert it and give them warning level (2). They vandalise it a third time. I revert it a third time and give them a level (3) warning. They vandalise it for a fourth time. What next!? Should i not revert it and still give them a level (4) warning? - Thanks, help on this matter would be appreciated. TheProf07 (talk) 20:42, 20 February 2008 (UTC)


 * The policy page says that 3RR does not apply to reversions of "simple and obvious" vandalism. However, what is simple and obvious to one person may not be to another.  At the minimum I would only do it if it's so obvious that no person could in good faith disagree.  Even there, edit warring is a touchy business and you are risking being warned or blocked yourself, rightly or wrongly, if it looks like you're one of the people perpetuating an edit war.  I would probably sit back and let another editor jump in, perhaps give the vandal a 3RR warning in addition to the vandalism warning, and report it to an administrative notice board, rather than taking the whole thing on myself even if policy permits it.  Wikidemo (talk) 23:12, 20 February 2008 (UTC)

Encountering a revert war
A 3RR issue recently came up on AN/I that isn't addressed by this policy, but probably should be. It got heated there, so I won't go over the particulars and just frame it generically.

User A and User B have rung up three reverts each on a page; it's a content disagreement, not an obvious case of vandalism. User C comes along, decides she likes User A's version, and reverts to it. Bad idea, generally. User A has treated three reverts like an entitlement and to revert to him, validates the behaviour. User B is suddenly going to feel that, though both may have sinned, he is now more sinned against and become angry. There would obviously have to be caveats—particularly, if User B was disrupting long-standing wording without discussion it probably is best to revert. But I think it would be useful to address the general problem. Marskell (talk) 15:57, 21 February 2008 (UTC)


 * In the perfect world User:C would take it too the talk page and use sound reasoning to persuade User:B that User:A's is a preferable version. Certainly appearing and taking sides on the article will just inflame the situation and make the article less stable. David D. (Talk) 17:13, 21 February 2008 (UTC)

There's another possible scenario - of editors deliberately acting in concert to get their way by combining their reverts to overwhelm a small group of opponents. User A disagrees with users B and C over a content issue. In order to get his own way, user A calls in ideological allies in the shape of users D and E, perhaps coordinating with them off-wiki or on a talk page somewhere. A, D and E have, between them, nine reverts. B and C have six reverts between them. A, D and E can overwhelm B and C simply by reverting more times than their opponents can. This isn't a theoretical possibility; I've seen this sort of thing actually happening on highly contested articles. In at least three cases that I can think of, I've seen hard evidence of ideologically allied groups of editors engaging in prior coordination of contentious editing activities (using e-mail, instant messaging or other web forums to coordinate their tactics). Unfortunately we don't seem to be very effective at dealing with gaming tactics of this nature, particularly if they involve established editors rather than an influx of POV-pushing newbies. -- ChrisO (talk) 20:46, 21 February 2008 (UTC)


 * The 3-revert rule can't solve everything by itself. The question is whether the scenario you describe constitutes  Edit warring or not.  From the way it sounds, I would say that's edit warring, and I would consider warning all those users that they need to engage more productively, and blocking them if they continue to revert-war, and probably protecting the page.  But it could also be that B and C have a minority position that they know is a minority position and are just trying to hold the article hostage against a clear consensus.  The 3-revert rule does give an edge to the larger group -- that doesn't mean the larger group is always right, but it does mean that they have an advantage by default.  I'm okay with that - more often than not, the larger group has consensus on their side.  Mango juice talk 15:33, 22 February 2008 (UTC)


 * The issue of course is that everybody knows that editors A/D/E in ChrisO's example (or editors S/C/J if you prefer) are a longstanding editing/adminning tag team. It doesn't violate the letter of any policies, so we all shrug and accept it. I don't see how we could write a policy that would prohibit this since in practice there's no bright line between people who just happen to have similar interests and those who have formed a bloc. Raymond Arritt (talk) 18:38, 22 February 2008 (UTC)
 * Except you're missing a step: A's second revert in this case was already edit warring, the third was certainly, the other editors aside.  Regular editors are blocked for edit warring even if they don't reach three reverts.  Sandy Georgia  (Talk) 19:00, 22 February 2008 (UTC)


 * I was shrugging as well on AN/I, Raymond, because there is no bright line and it's hard to police this behaviour. But that doesn't mean the policy should say nothing—it's not uncommon, after all, to come across other people's edit wars. ChrisO is very right about editors beyond C, but people aren't going to conceptualize that far, so I think any wording would have to stick to two parties + third party. Marskell (talk) 20:06, 22 February 2008 (UTC)


 * Tandem edit warring is clearly gaming; doing it in a pattern as a reverse form of wikistalking is enabling and manipulative of good faith and etiquette. Raymond's initials of the involved edits are the elephant in the room example. What's going on is cabaling, not collaborating. What we need are admins willing to use judgment to make bold decisions that obviously need to be made and not pussyfoot around about the necessity of "bright lines."
 * I've been conceptualizing some language in my thoughts as proposed wording. Words in policies will help make this explicitly actionable to reticent admins. I'll write it down later for discussion. SchmuckyTheCat (talk)

Edit war vs. 3 reverts
I was thinking about posting a suggestion to modify this policy on the village pump and thought I'd see what any regular editors here thought first.

This policy appears to have a few ways for the system to be gamed so that it can actually be used as a tool in an edit war rather than preventing them. If the idea is to limit edit warring wouldn't it be better to address that instead of concentrating on the number of reversions alone? Don't get me wrong, 3RR is a symptom of edit warring, but by no means the only one. For example sock puppets are often used in an edit war just as much as three or more reverts. Gaps are created in enforcement which helps both trolls and good faith editors who refuse to engage in discussion by essentially giving them the ability to continue, without having to discuss their differences. As long as they don't break one rule too much since they know people on the 3RR board aren't interested in sock puppet problems, so long as they aren't egregious.

In short, why not create an enforcement/noticeboard designed to curb edit warring, with behavior like 3RR, sock puppetry, coi, or whatever is discussed with an eye toward how they relate to possible edit wars. An example post would be something like this though not necessarily in table form of course:


 * I was thinking about adding a section titled "What the 3-revert rule is not." Specifically, it could include an entry like "The 3-revert rule is not a weapon" that explicitly condemns attempts to get others blocked by continuing to revert.  Also, it's not an entitlement, and it's not the only rule against edit warring.  What do people think?  Mango juice talk 06:22, 23 February 2008 (UTC)


 * I think what you are looking for is our policy on Edit warring. Dmcdevit·t 07:42, 23 February 2008 (UTC)

I totally agree that your ideas could do something to help Mangojuice :) I was thinking of consolidating enforcement of the 3RR and others related to edit warring into its own noticeboard so that people conducting edit wars can't use the separate nature of the 3RR noticeboard/Suspected sock noticeboard/etc. to game the overall system.

Dmcdevit I'm actually aware of the policy, and my point is that we have so many different areas devoted to addressing specific symptoms of edit warring but nothing (noticeboard wise) for it. (Granted there are routes such as RfCs or 3O but these solutions rarely seem to work or are simply ignored in favor of the incident noticeboard. Anynobody 02:24, 24 February 2008 (UTC)

Introduction
Except in special circumstances? What special circumstances? Trying to make the article better? Imperial Star Destroyer (talk) 16:57, 26 February 2008 (UTC)


 * I personally would like this question answered with a simple yes or no. If you have reverted an article three times and it gets vandalised (clear vandalism that is) again, can you reverted it again? Thanks TheProf | 2007 17:07, 26 February 2008 (UTC)
 * Yes, simple and obvious vandalism (not subject to 3RR). see exceptions to the three revert rule--Hu12 (talk) 19:57, 26 February 2008 (UTC)
 * I'm glad you said that. Because an Article ive reverted 3 times on today has just been vandalised again! TheProf | 2007 20:01, 26 February 2008 (UTC)
 * Just had a look at the article you mentioned, your last revert is perfectly fine. No worries, thanks for reverting it.--Hu12 (talk) 21:52, 26 February 2008 (UTC)

Removing AFD notices
An editor has twice removed an AFD notice, despite the notice itself stating that "this notice must not be removed". If he continues to remove the notice and I continue to undo his removal of the notice, am I potentially guilty of violating 3RR? I am prepared to report him for a 3RR violation if he continues, but I want to make sure that I myself am not violating 3RR. For now, I am treating his removal as vandalism and my undoing of that vandalism is excepted from the rule. If I am wrong, then I will self-revert my restoring of the notice. Sbowers3 (talk) 04:09, 5 March 2008 (UTC)
 * I've always considered removal of AFD notices to be tantamount to vandalism, assuming the user has been told not to remove them. At any rate, though, I've just checked your contribs and found the article in question and watchlisted it so I can help if the user keeps it up. Heimstern Läufer (talk) 04:13, 5 March 2008 (UTC)

Replace this rule?
Does this rule have to be so mathematical? There are already so many exceptions and additional stipulations that the numerical criterion of "3 reverts" effectively doesn't (or at least shouldn't) apply anymore. So what if someone's made three reverts to protect the encyclopedia from unreasonable edits? They shouldn't be blocked for that; they shouldn't even have to worry about the possibility of being blocked. The only criterion to be applied should be reasonableness of behaviour. Since the first shot in an edit war isn't a revert, the rule currently implies that in a simple one-on-one edit war, the party defending the status quo is assumed to be in the wrong (and of course the rule can easily be manipulated/evaded anyway by the use of puppets).

I would judge reasonableness in edit warring largely by willingness to engage in dialog. If one side initiates a talk page discussion, and the other side keeps reverting without engaging sensibly in that discussion, then we have a situation where a block could be considered. Let's replace this arbitrary 3RR with an UEWR (Unreasonable Edit Warring Rule) or something along those lines.--Kotniski (talk) 16:26, 28 March 2008 (UTC)
 * I believe the policy that you are looking for is WP:EW. That policy pretty much covers edit warring regardless of the number of reverts. I've always viewed 3RR as a way to prevent a single user from forcing their opinion on an article. If the status quo truly has consensus, then it shouldn't only be one editor defending it. If it is one editor defending the status quo, then that is probably evidence that the status quo doesn't have consensus and that discussion should be taking place on what is the consensus. It also tends to prevent WP:OWN as a single user can't stop multiple editors from making changes to an article and claim that they are defending the status quo. --Bobblehead (rants) 17:54, 28 March 2008 (UTC)
 * That makes sense for major articles which are being watched by multiple online editors, but not in cases where only one person happens to be currently online and interested. For example, I don't see why someone can make the same misinformed edit to an article four times (only the last three counting as reverts), declining all offers to discuss on the talk page, and I then have to leave his version in place for 24h because if I revert it again I'll be in danger of being summarily blocked.--Kotniski (talk) 18:42, 28 March 2008 (UTC)
 * Unless it is a BLP violation or vandalism(which are exempt from 3rr), then it can stay up on Wikipedia for a few hours until someone else comes along. I know it can be aggravating, but until someone else chimes in then it's just a he said, she said. If it's a protracted content dispute and the person continues to not respond, you can always ask for assistance on AN/I, and/or solicit assistance from one of the wikiprojects on the article. Basically just ask for an extra set of eyes on the article. --Bobblehead (rants) 20:45, 28 March 2008 (UTC)
 * OK, I'll try that next time it happens.--Kotniski (talk) 20:57, 28 March 2008 (UTC)

Proposed amendment to rule
The 3RR has come up again at the Village Pump - a policy change is being proposed there, whereby a 3RR block would only be possible if preceded by a warning and if the offender persists. This seems only reasonable, since a similar courtesy is extended to vandals under the vandalism policy, and over-zealous edit-warriors can hardly be thought to merit harsher treatment than vandals. (Can they?)

The relevant thread is at WP:VPP - the proposal in question appears towards the end of that section.--Kotniski (talk) 12:54, 8 April 2008 (UTC)


 * This part of the blocking policy already covers this, and that should not be duplicated here (or worse, inconsistent material should not be added here). --bainer (talk) 13:57, 8 April 2008 (UTC)


 * Well, I'm not sure that the present wording of "here" is consistent with "that". It certainly seems skewed that the wording of the vandalism policy places so much emphasis on the need for a warning first, while this 3RR policy hardly mentions it. --Kotniski (talk) 14:06, 8 April 2008 (UTC)

Proposed rewording of the libel exception
"Reverts to remove clear violations of the biographies of living persons policy, including libellious material and unsourced or poorly sourced contentious material."

Verfiability is only one half of the BLP policy - we must also take into account NPOV - given dedication, a user can write a brilliantly sourced but absolutely negative article about a person, and efforts to fix a violation of a foundation issue aren't given the same protection as removing violations of a (non-negotiable all the same) wiki-by-wiki issue. Sceptre (talk) 22:10, 1 May 2008 (UTC)

"Not an entitlement" section added
I've very frequently seen this rule used as a defense in unblock requests from people blocked for edit warring: "I didn't make three reverts!" and the like. I've added that section to make it a little more clear that that is, quite frankly, irrelevant in the larger scheme of things. Gaming the system to avoid breaking rules is, if anything, more disruptive than simply edit warring, and WP:EW notes that blocks can be made just for edit warring as well. If anyone disagrees with it, I'm open to discussion, but it seems this is more-or-less just a copy of what's posted elsewhere. <em style="font-family:Bradley Hand ITC;color:blue">Hers <em style="font-family:Bradley Hand ITC;color:gold">fold  (t/a/c) 15:48, 9 May 2008 (UTC)
 * I think there should also be something against another form of system gaming - trying to trap an "opponent" into breaking 3RR in order to get them blocked. In fact, as I've said somewhere above, I think the making of a fourth revert is pretty irrelevant in trying to identify guilty parties in edit wars, and the numerical aspect of this rule should be abandoned or at least toned down.--Kotniski (talk) 16:37, 9 May 2008 (UTC)
 * I think that's kinda of covered by the admonition to "follow the spirit of the law, not the letter," and users who do so would likely be blocked for trolling as well as edit warring. You're welcome to add it in if you like, however. I don't see much problem with it, as again, it's basically restating what's already elsewhere. <em style="font-family:Bradley Hand ITC;color:blue">Hers <em style="font-family:Bradley Hand ITC;color:gold">fold  (t/a/c) 15:30, 11 May 2008 (UTC)

"Fairly" instead of "equally"
In similar situation the Arbitration Committee has started using the word "fairly" rather than "equally" so it is clear that individual circumstances will be taken into consideration for each editor. Both side of a dispute need to be examined and treated fairly based on their individual level of disruptive editing in the current and previous situations. I changed the wording here as well since that is what happens and editors need to understand it. FloNight&#9829;&#9829;&#9829; 12:21, 24 May 2008 (UTC)

Bold change to exception
The rule for the exception to vandalism reverts and BLP reverts were not parallel. I made them parallel with this edit:. I encourage others to improve it, if they can, but the basic idea should be that it should be obvious BLP violations. This clause could be used as a get-out-of-jail-free card otherwise, which is not its intention.

ScienceApologist (talk) 00:59, 11 July 2008 (UTC)
 * I don't agree (and have reverted). WP:BLP violations, even if not immediately obvious to the casual user, need to be robustly removed (for legal and moral reasons). It oughtn't to lead to get-out-of-jail-free situations, since the "accused" would still have to provide convincing arguments to show that they were acting under this clause.--Kotniski (talk) 06:11, 11 July 2008 (UTC)

Clarification - editors or numbers?
Hey. So.. on this page, an admin wrote that "The WP:3RR does not require the same editor to be reverted, only the same content." An editor (who has been reverting fairly frequently) has been repeatedly recalling this rule, but I can't find any evidence on this page backing that up. Is it actual policy, or is just the opinion of one editor? Thanks. &mdash;  Hello Annyong  (say whaaat?!) 14:39, 19 July 2008 (UTC)


 * The wording "A revert means undoing the actions of another editor, whether involving the same or different material each time" says to me that it does not matter which users you are reverting. It does not have to be the same person or the same content to be a violation, simply reverting too many times is enough. Chillum  14:47, 19 July 2008 (UTC)


 * Er, alright. I was thinking more of like, one editor making edits and multiple editors reverting the other editor's edits. In other words, the following actions:
 * Editor A makes an edit
 * Editor B reverts that edit
 * Editor A reverts the last edit
 * Editor B reverts the last edit
 * Editor C reverts the last edit
 * Editor A reverts the last edit
 * In this case, can editor B and C be counted as 3RR, since collectively they reverted three times? &mdash;  Hello Annyong  (say whaaat?!) 15:08, 19 July 2008 (UTC)


 * Ah. Well no, B and C are not in technical violation of 3RR(unless they are the same person socking of course). B and C could be considered to be edit warring, and depending on the content of the reversion it could be considered disruptive. I think A needs to take B and C to the talk page. Chillum  15:27, 19 July 2008 (UTC)


 * It's edit warring without a doubt, but I was just curious as to the interpretations of 3RR in this context. &mdash;  Hello Annyong  (say whaaat?!) 00:41, 20 July 2008 (UTC)

'Unwanted Edit' addition
I reverted as as written the section about other exceptions is listing what you can't revert to on your user page, and I think that if people want to revert to vandalism on their own page they should be able to. Certainly having sections refernce one annother could lead to unintended consequences if only section is edited but not the other. Narson (talk) 14:49, 7 August 2008 (UTC)

Ambiguity in revert definition
The definition of a revert is currently 'undoing the actions of another editor'. Suppose the page history is like this: How many reverts did Carol make so far? The wording on the 3RR reporting page suggests that #3 is not a revert since there is no prior version that is being reverted to (i.e., Carol made one revert). However, in edit #3, Carol did undo the work of Alice and Bob, which falls under the definition of a revert (i.e. Carol made two reverts). I would appreciate a bit less ambiguity. Han-Kwang (t) 15:34, 9 August 2008 (UTC)
 * 1) Alice writes A
 * 2) Bob expands it to AA
 * 3) Carol replaces AA by BB
 * 4) Dave reverts BB to AA
 * 5) Carol reverts AA to BB


 * "Replace" is not the same as "revert". Help:Reverting, which is linked to in the second sentence in the body of this policy, says, explicitly, Reverting involves returning a page to a previous version..  -- John Broughton  (♫♫) 21:27, 28 August 2008 (UTC)
 * Presumably we need a more robust definition here, though. Otherwise people might return to a slightly different version (by rewording their own preferred version slightly) and claim that they are not reverting.--Kotniski (talk) 21:58, 28 August 2008 (UTC)
 * I notice just under the link you mention, there is actually a specific definition of revert for the purposes of this rule (with a disclaimer that it is such). Perhaps the word "revert" should be de-linked to avoid confusion. (Doing this.)--Kotniski (talk) 22:03, 28 August 2008 (UTC)


 * Number 3 in your example is not a revert because it involves the introduction of substantial new content. A user who repeatedly replaced other users' content with substantial new content would not be violating the three-revert rule – but could still be blocked for edit warring if it was disruptive. Don't forget that Edit war is policy too; this policy does not and is not intended to cover all edit warring, just a particular scenario -- Gurch (talk) 10:46, 29 August 2008 (UTC)
 * Although I agree with you, my point is that this kind of distinction was and is not clear from the definition of a revert in the context of 3RR ("undoing an action"). I see that you changed the wording into "reverse actions", but that does not resolve the ambiguity IMO. If I change the above edit history example into this (Example 2):
 * 1. Alice writes A
 * 2. Bob expands it to AA
 * 3a. Carol reverts to A
 * 3b. Carol replaces A by BB
 * 4. Dave reverts BB to AA
 * 5. Carol reverts AA to BB
 * then the end result is the same, while it would be counted as a revert here. A 3RR that treats #3 in the first example differently from #3a/b in the second example is bad rule IMO. Han-Kwang (t) 20:12, 29 August 2008 (UTC)
 * 3RR is a bad rule in my opinion, for this reason and several others... However, given that the community seems to be wedded to it, we have to do the best we can. If you can think of a tighter definition of revert, then do tell. But as Gurch says, there are other policies besides 3RR to deal with edit warring, so we don't have to cover every possible case on this page.--Kotniski (talk) 15:49, 30 August 2008 (UTC)

User page 3RR
Should a user who removes something like a sock puppet warning repeatedly from his user page be considered in violation of 3RR? This is the issue with a suspected sock puppet (and all but definite one-shot account) currently pushing an agenda. I'll not name names, but if you check my contribs you'll probably figure out who I mean. --coldacid (talk|contrib) 01:34, 15 August 2008 (UTC)
 * I would rather see the 3RR criterion apply to all pages; there's no reason an editor should be edit warring on their user pages. &mdash; Carl (CBM · talk) 01:42, 15 August 2008 (UTC)
 * FYI, coldacid is talking about this incident Adam233 (talk) 02:17, 15 August 2008 (UTC)
 * Note that you started the reverting, though, which puts you in violation before Gene. --coldacid (talk|contrib) 02:27, 15 August 2008 (UTC)
 * Note that already Gene's first edit violated Wikipedia policies, which puts Gene in violation before me. Adam233 (talk) 02:30, 15 August 2008 (UTC)
 * And which one was that? This enquiry of mine is strictly from the POV of the 3RR, without other policies interfering. Rather than continually reverting like you did, you could have actually offered up some evidence in your favour. That you didn't, however, only goes on to increase suspicions.
 * If you want to keep debating that, your or my user page would be a better place. Note that you're probably not the first, nor the last, to have potential 3RR issues in your user space, and the discussion here should allow for other scenarios as well. --coldacid (talk|contrib) 02:41, 15 August 2008 (UTC)
 * coldacid, it is clear that Gene Poole asked you here to vandalize my user page, because Gene Poole and you tried to provoke a 3RR-violation from my side. However, this plan didn't succeed because I posted this note to the administrators' noticeboard instead of starting an edit war. Adam233 (talk) 02:58, 15 August 2008 (UTC)
 * You and him already got into the 3RR war before I jumped in on your user page. Also you meant to say "as well as" rather than "instead of". If you want to keep arguing about this, do so there. As far as I'm concerned, my question for here is answered, and the issue in relation to this talk page is closed. --coldacid (talk|contrib) 03:08, 15 August 2008 (UTC)
 * (EC)Adam didn't breach as he had the right to remove a false claim which wasn't backed-up. I think it would be best to start proving at WP:SSP then continue to make claims which you can't backup and are breaching WP:BITE. Bidgee (talk) 03:16, 15 August 2008 (UTC)
 * On the other side, you took part at Gene's vandalism of my user page . Adam233 (talk) 03:14, 15 August 2008 (UTC)
 * Please see WP:3RR: Reverts performed by a user within his or her own user page, user subpages, provided that such reverts do not restore copyright or non-free content criteria violations, libelous material or biased, unsourced, or poorly sourced controversial material about living persons. A user is allowed to blank his/her own page without violating 3RR. It does appear, however, that Gene Poole has violated 3RR on Adam233’s page. — Travis talk  02:37, 15 August 2008 (UTC)
 * I'm willing to accept that Gene also broke 3RR in this case, and I wasn't much help on that front either. However, "serious" accusations such as sock puppetry or being highly NPOV shouldn't be hidden away while an issue is open about it, but rather discussed with the interested parties. Simply reverting those seems like something that would violate 3RR. --coldacid (talk|contrib) 02:41, 15 August 2008 (UTC)
 * Again, Adam233 is in no danger of violating 3RR on User:Adam233. As other admins and users have stated, if you or another user has evidence of sockpuppetry, make your claim at WP:SSP rather than edit warring on the user page. — Travis talk  02:45, 15 August 2008 (UTC)
 * It'd be a lot easier to collect said evidence if user names didn't obscure IP addresses. But again this isn't the place for that discussion; I'm just saying that user warnings shouldn't be flat-out reverted. In this case it was a sock puppet accusation; tomorrow someone might accuse me of being highly NPOV and put a similar banner on my user page. If I got in an edit war over something like that on my user page, as far as I'm concerned, I'd be violating 3RR even though it's my user page. You see what I'm saying? This needs to be clarified in policy, which way such a scenario goes. This one incident is just a specialization of a general issue that I want clarification on. --coldacid (talk|contrib) 02:52, 15 August 2008 (UTC)


 * Reverts to a user’s page by the user are explicitly exempted in policy as I point out above. Unless his/her reverts violate some other policy, e.g. copyright violation, personal attacks, libel, or abuse of unblock, he/she is not violating 3RR. — Travis talk  02:57, 15 August 2008 (UTC)
 * Thank you, the straight answer I was looking for (finally). --coldacid (talk|contrib) 03:08, 15 August 2008 (UTC)

Rewrite
I have rewritten this to make it shorter and more concise. There is no (intentional) change to what the policy actually says -- Gurch (talk) 15:57, 27 August 2008 (UTC)
 * Nice work!--Kotniski (talk) 08:44, 28 August 2008 (UTC)


 * Regarding your change, it said "simple and obvious vandalism" before, and I left it like that because I didn't want to change the meaning. I don't have any opinion on whether that's what it should say -- Gurch (talk) 09:31, 28 August 2008 (UTC)