Wikipedia talk:Three-revert rule/Archive 5

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Archive 4 | Archive 5 | Archive 6


undiscussed addition

On Nov 18, Dmcdevit without proposing and discussion amended this policy page by this edit. This addresses not the policy itself but a case by case interpretation of the policy and this particular wording is rather arrogant. This is rather unwarranted edit in my view. A policy is one thing. How it is interpreted is quite another and it is important to not have this confused. Besides, having such an arrogant statement a part of the policy page does not make a good impression. Administrator's discretion is a tricky and delicate thing and should be threaded with more caution. I would like to see this discussed and consensus reached before the page being altered like that. Until then, I am restoring the original wording. --Irpen 23:47, 19 February 2007 (UTC)
moved from user talk:Irpen

Hi Irpen. I just saw the recent activity at WP:3RR. You seem to be under the misconception that policy is prescriptive, when in fact, it is descriptive. I haven't "proposed" anything because this simple sentence ("Many administrators give less leniency to users who have been blocked before, and may block such users for any edit warring, even if they do not exceed four reverts on a page in 24 hours.") is a statement of current practice with widespread acceptance. You've interpreted it as a change in policy, but it isn't, so I'm replacing it. When you find yourself reverting three people in the space of a day without talking about it first, you might want to question who is trying to change policy. The sentence does not mandate its use, but note its acceptance according to administrators' preference. In the future, when you have an objection to an edit, many people (myself included) consider it disrespectful to revert first without any attempt at communication, and especially without any communication after the fact either. It won't kill anyone to wait a day to get to talk to someone about an edit you disagreed with instead of reverting immediately. The nature of the project is that cooperation and trust are key; it's good to treat others with latitude. Dmcdevit·t 00:03, 20 February 2007 (UTC)

Dmcdevit, in the future when you see a change you make to the page being reverted, many people (myself included) consider it disrespectful to reinsert without any attempt at communication, and especially without any communication after the fact either. It won't kill anyone to wait a day to get to talk to someone about an edit someone disagreed with instead of restoring immediately. The nature of the project is that cooperation and trust are key; it's good to treat others with latitude. --Irpen 00:09, 20 February 2007 (UTC)
No, Irpen, you've got it backwards. This is a policy page, and changes like you made that are actually substantive should always be discussed first. At least two reverts of your edit included a call for discussion of the issue. As to the substance of the issue, I totally agree with Dmcdevit that this is the way things work, and the page should be descriptive. It may not be totally necessary to bring this one point up, because such blocks can always be justified under WP:POINT or WP:DE or such, but as to what the policy is, there's no question this is how things are done. Mangojuicetalk 14:32, 20 February 2007 (UTC)
Mango, just check the facts please, will you. The substantive change was made not by me but by Dmcdevit[1] and I undid his change because he made a substantive change to the policy page without discussion and not the other way around. --Irpen 19:15, 20 February 2007 (UTC)
Dmcdevit is actually following the flowchart at Wikipedia:Consensus correctly. We've also automatically gotten to WP:BRD (though I've been told the latter page needs tidying). So in short, we're all in the clear, procedure-wise. --Kim Bruning 22:16, 20 February 2007 (UTC)

Why Block Access to Talk Pages?

Something I've been wondering: Why, when people are blocked for revert warring, does the block include access to talk pages? Has this been extensively discussed? I see several problems:

  1. Seems counter to the purpose of encouraging talk rather than revert warring
  2. Is particularly frustrating to someone who gets blocked
  3. Seems punitive
  4. Contributes to schadenfreude
  5. Doesn't seem to serve any purpose

The only reason I can think of is that we don't want people running across multiple talk pages to complain when they just got blocked. But couldn't that be dealt with some other way? Further problems with this basis:

  1. Forces people to use frustrating email appeal system (Is anybody listening? Better email everybody.)
  2. Disruptive talk page behavior could always subsequently be blocked
  3. If people could still use talk pages, they might be less adamant about getting unblocked (clearly they'd know they couldn't edit the page again for 24 hours anyway).
  4. The rule is still somewhat punitive by preventing people from editing any article for 24 hours, even those where they did not edit war.
  5. This blocking in regard to all articles, rather than simply the one at issue, already has punitive force.

I guess the other concern might be that it's not punitive enough, and people would feel free to use their four and then get blocked to no big loss. But then, wouldn't escalating blocks and the stigma of dirty block logs be able to deal with that? Curious what people think. Mackan79 20:11, 21 February 2007 (UTC)

I don't believe that blocks can be made that granular. The only exception to the rule that a block is for all pages is the user's own talk page (to appeal a block); this is built into the software (I believe).
In other words, you're proposing a change to the software; this isn't something that admins can do on their own. (That doesn't address the issue of whether such a change would be a good one, but it does - to some extent - answer your question of "why".) -- John Broughton (♫♫) 03:38, 22 February 2007 (UTC)
That's correct, it is not possible in the software to block a user only from certain namespaces or certain pages. I am aware of people who are at least considering implementing things like per-page blocking in extensions, but there hasn't been much enthusiasm from the devs on this front.
As to the desirability of such a feature, I don't really think it would be an improvement. The 3RR is like the proverbial rolled-up newspaper that one thwacks one's dog on the nose with to teach them that they've been bad: the idea is to encourage the behaviour of going to the talk page - and involving other editors - as a first resort, not simply after one has run out of reverts. If a user employs the wrong behaviour (edit warring) they get sent to the corner to think about what they've done, and so do all the other children who've been edit warring with them. --bainer (talk) 07:01, 22 February 2007 (UTC)
Basically, the only way to implement this kind of block from an article only would be to protect the article page, which is done on occasion. Talk pages are not protected when article pages are. I don't think I would want a software change to allow for extremely granular blocks; it sounds hard to implement, hard to use, and not that useful. Mangojuicetalk 12:55, 22 February 2007 (UTC)

Interesting, I was wondering if it was tech-related. I don't know, though, it would seem like a potentially big improvement to me. Per Bainer's comment, I see several interests being balanced by 3RR: protection, deterrance, keeping people guessing so they don't get too legalistic. Still, there seems to be some frustration with the continuing edit warring. As far as I can see, the problem partly relates to the conflict between those who want more punitive measures to really stamp it out vs. others who want more tolerance and fairness and presumptions of innocense, right? That said, I'd think one benefit of separating talk and article blocks would be to allow more vigilant ending of edit wars, while not blocking out their total involvement. This would say if you violate 3RR on a page once, you get blocked from editing all articles for 24 hrs. Twice, 72 hrs. Thrice, a week, etc. But we won't prevent you from participating in talk.

Of course we don't want to over-complicate the system. Speaking personally, though, I have to say I was pretty surprised upon first being blocked for a 3RR violation (which like everybody, I guess, I didn't think I'd committed) and finding that I wasn't even able to edit talk pages. I later got used to the idea, but isn't it a bit strange? We're saying people should stop edit warring and get back to talk, yet we then block them from using the talk pages. Simply from an outside perspective, I would never have guessed people would be blocked from editing talk pages, unless they were actually vandalizing them or trolling, etc.

So is the fear, then, that blocking people from articles wouldn't be frustrating enough to actually deter edit warring? So people would constantly tread the line and violate it without any real concern? I'd just think there must be better ways to deal with this, for instance longer time periods, than blocking access to talk pages. Mackan79 17:04, 22 February 2007 (UTC)

I have to say I was pretty surprised upon first being blocked for a 3RR violation (which like everybody, I guess, I didn't think I'd committed) - I was under the impression that admins generally wouldn't block for a 3RR violation unless a user had been specifically warned (and then did another revert, or did a 3RR violation at a later date/time).
As far as blocking everything, my sense is that people here have what might be called a longer view - that what matters is what happens in a week or a month or longer to an article, rather than that it must be accurate 24/7, 365 days a year. If one editor can't participate for a day, it's no big deal - editors have real life commitments that often keep them away for weeks if not months. Nothing (well, very little) ever disappears from Wikipedia; you can always revert to a day-old or week-old version of an article (not, generally, a good idea) or (better) compare the current version of the article to what it was when you saw it last.
Finally, consider one benefit of a block - it's a sort of forced wikibreak, giving an editor a chance to calm down and (possibly) put the importance of a Wikipedia article (or two) in perspective (say, compared to having a heart attack caused by excessive stress). If the editor could continue arguing on talk pages, it wouldn't be that much of a chance to calm down. -- John Broughton (♫♫) 22:17, 22 February 2007 (UTC)
Well, that's true in a way. Are we really so sure that getting blocked (and muzzled) tends to give people greater perspective, though? It strikes me as pretty speculative, for a rule trying to do the least possible to protect WP. My thought would be that WP would try to treat people like they simply didn't understand the rules (like we spend so much time treating vandals), not that they need perspective or a whack on the nose (for people who are actually presumably trying to improve WP). Of course, we do give warnings, but I think we also forget how complex WP's rules are, in addition to people's tendency to make mistakes. Isn't it a bit odd to start by assuming that what people need is a moral/life lesson? Maybe they really just need to be told "Yes, that violated the rules, which means you can't edit WP articles until this time tomorrow," and the message would get across. Mackan79 15:21, 23 February 2007 (UTC)

Also, I just wonder: if somebody is so stressed it causes them to break 3RR, and so they then get blocked while the other guy presumably doesn't, does that really reduce their stress? I have to say, if there have been wiki-heart-attacks, I'd actually be willing to bet that was the situation...Mackan79 15:33, 23 February 2007 (UTC)

Enlisting others to circumvent 3RR?

Here's an interesting twist on 3RR circumvention I saw mention of on someone's user talk page today; say User A is on the verge of violating 3RR if they make one more revert, so they decide to contact friendly User B to go make the revert for them. I suppose it'd be hard to prove collusion in such a case unless they blatantly talked about it openly here, but I was just curious if this has ever come up before? Tarc 21:24, 22 February 2007 (UTC)

Explicitly permitted. Though better would be not to edit-war at all. See also Revert only when nescesary --Kim Bruning 23:12, 22 February 2007 (UTC)
I'll quote from the very first version of this page: "If the edit really needs reverting that much, somebody else will probably do it – and that will serve the vital purpose of showing that the community at large is in agreement over which of two competing versions is correct." What's the difference between "collusion" and "cooperation" in this context, and more importantly, how would you make that difference clear in the policy? See also #Proposal: 3 Reverts of CONTENT, not 3 Reverts by user and #Are cartels allowed? above for recent examples where the idea of making the rule apply to particular content or to "sides" in a dispute, rather than individuals, has been discussed.
The 3RR is only to deal with individuals who make too many reverts. Other problems can be dealt with elsewhere; recruiting people to participate in an edit war, especially when canvassing is involved, is likely to be considered disruption, for example. There's no need to introduce something here to counter that which would have the downside of making it easier for "a single editor [to] hold community consensus to ransom", as Agathoclea put it. --bainer (talk) 00:22, 23 February 2007 (UTC)

What constitutes a revert?

This question has come up via unblock-en-l, regarding Crum375's block of User:Melonbarmonster for edits to Japanese people. Discussion includes User talk:Melonbarmonster, User talk:Crum375, User talk:Georgewilliamherbert#Your note about 3RR primarily.

Precis of question: does a first recent edit which partially restores and partially modifies detailed wording which an editor had originally created but last edited a month previously count as an initial "revert" for purposes of 3RR?

The specifics in this case come down to a one-word disagreement between two editors, over "forced" versus "compulsory". There are other phrasings involved, but those have evolved over time.

There had been a number of intermediate edits, including one by the subsequently 3RR blocked editor, which left the section unmodified after the last change/revert (once each) cycle.

My understanding of the usage of 3RR is that its intent is to prevent edit warring and in particular sterile edit warring. In this case, both parties stopped at what they interpreted 3RR to cover, and were commenting on the talk page appropriately.

Crum375 and SlimVirgin have both indicated that they feel that the first Feb 28th edit counts as a first revert. I disagreed, but I pose the question here for comment to see if this is me interpreting the guideline in a nonstandard manner against other admins' consensus, and to seek clarification on how rigidly people are enforcing first-edits of this nature.

Comments from all (including clarifications from Crum375 and SlimVirgin) welcome. Georgewilliamherbert 22:54, 1 March 2007 (UTC)

This is a clear 3RR violation, showing four reverts in 68 minutes by an editor who had already been blocked twice during the preceding seven days for personal attacks. The edits were:
  • Version reverted to: 18:04 24 Jan, Melonbarmonster adds the word "forced."
  • 1st revert: 17:15, 28 Feb, Melonbarmonster restores the word "forced."
  • 2nd revert: 17:26, 28 Feb, Melonbarmonster restores the word "forced."
  • 3rd revert: 17:36, 28 Feb, Melonbarmonster restores the word "forced."
  • 4th revert: 18:23, 28 Feb, Melonbarmonster restores the word "forced."
George feels it's a "broad interpretation of 3RR," as he wrote on his talk page, to call the first edit a revert, given that the previous version is from 35 days earlier. I might agree if the previous version had been created by a different editor, so that the editor accused of the violation couldn't be expected to know about it. But in this case, it was the same edit made by the same editor, and he almost certainly did realize that he'd already inserted this into the article, and that it had been rejected. He then proceeded to revert to it four times in 68 minutes, ignoring a warning that he was about to violate 3RR. [2] It seems to me that this is exactly the kind of editing the 3RR rule is designed to prevent. SlimVirgin (talk) 02:52, 2 March 2007 (UTC)

If both parties have stopped reverting and are engaged in discussion on the talkpage, there is no need to block anyone regardless of whether it's arguable there was a fourth revert. Enforcement of the 3RR is a means to an end, and not an end in itself. (General observation, I haven't reviewed this specific case.) Newyorkbrad 02:58, 2 March 2007 (UTC)

Brad, it's almost always the case that blocks take place several hours after the reverting has stopped, simply because of the time it takes for the report to be made and acted upon. That doesn't make the block invalid; on the contrary, it'll hopefully prevent Melonbarmonster from violating it in the future. SlimVirgin (talk) 03:05, 2 March 2007 (UTC)
I think I'm out of sync with the general enforcement policy on 3RR. The one night I was there I wound up giving out a lot of warnings rather than blocks, and my talkpage was beseiged with complaints of "I'm entitled to have my opponent blocked, why didn't you do that?!" It would be interesting if we could do some sort of controlled experiment to see whether 24-hour blocks add more in deterrence than they take away of editing (including editing of articles unrelated to the one being edit-warred on). In the meatime, I think I will leave 3RR enforcement to admins willing to be stricter than I seem to be. Newyorkbrad 03:11, 2 March 2007 (UTC)
It's always preferable to warn editors before they violate, or before they're reported, but in this case Melonbar was warned, but he ignored it and reverted again anyway. SlimVirgin (talk) 03:20, 2 March 2007 (UTC)
I want contribute to this discussion if I may. I left this comment on my own talk page in response to George but I feel it's relevant to this discussion.
While I had no problem just waiting out the block, I'm still trying to bring attention to this and get more feedback in hopes of some sort of rasonable understanding bc I feel that Slimvirgin and Crud573 are not correct on the facts of the situation.
From what I can gather Crum is still representing my first edit as a reversion to an edit that I proposed back in January(mind you he had no idea about the January version when he first blocked me!!!). Those two versions contain similar elements(such as "force") but they are not the same! There are many compromised elements present in my recent edit that were not in the January version which is why it was supported by a third editor who had rejected my January edit proposal. Slimvirgin and Crum are wrongly ignoring the evolution between those two versions(documented in the History and Talk page of "Japanese people"). I've explained this over and over but Crum has yet to respond on this point and is stuck on the fact that "force" was added in both the January edit and my recent edit and slimvirgin(and perhaps other admins) seems to have just taken Crum at his word.
My 2 cents on the policy question is that I think there has to be some issues that need to be clarified:
1. There has to be a policy that enables administrators and editors to distinguish a REVERT and a normal EDIT when the edits involve same words which are being reworked, reused in a normal editing process by editors working out disagreements. My case presents an example when simply looking for same edit elements(such as looking for addition of "force" in place of "compulsory") misses the natural progression of edits that may use, delete, reuse the same edit elements in different variation. E.g., just bc "force" was proposed in multiple edits doesn't mean that each edit containing that word is automatically a REVERT. Admins should be capable of looking beyond reappearance of a single "edit" element and distinguish REVERTS from instances when editors are engaged in normal editing with edit proposals that contain different variations of the same edit element. To ignore the existence of this distinction in policy creates a ridiculous situation in which editors can't propose different and original uses of edit element(s) from previous edit versions ecause it would be considered a REVERT. This is not what the 3RR and wiki policy against RW are trying to prevent.
2. A clear, classic example of a revert is when an edit changes the text back to a previous version in whole or in part. However, it must be noted in policy that this doesn't preclude editors from reusing elements of previous edits in novel ways or in a normal edit process to trying to resolve differences. You want editors to do more of this: to rework and propose compromised versions of previously disagreed elements in new arrangements and compromise and be flexible. This is how you prevent edit wars.
3. If RW'ing is deemed enough for a block, the block must be also placed on the other party engaged in the same level of RW'ing. It makes no sense to block only the reported party(reported by the RW opponent) for RW while taking no action against the reciprocating party.
4. The process of review of Unblock requests must be conducted by third party admins. Admins who placed the Block becomes a biased party as soon as review is requested and are more likely than neutral admins to be unwilling to overturn their own decision. It is unfair to make blocked editors who may have been wrongly blocked to have the burden of convincing the admin that they were wrong. This defeats the purpose of review and is bad policy. melonbarmonster 03:46, 2 March 2007 (UTC)
Melonbar, you've made 193 edits to articles, and in the space of seven days have been blocked three times, including for this remark, which clearly shows you counting reverts; and for calling someone an "asswipe". Then you repeatedly remove the various admin warnings and block template from your talk page, [3] to the point where the page has to be protected. [4] You continue edit warring, you're warned about 3RR, you ignore the warning, and so you're blocked. You then bombard the blocking admin with nine e-mails; you involve George from the unblock list; and you post 35 times about it on your talk page, mine, and here, even though it's a straightforward and obvious violation (no matter what you say, it is). All in all, this is not a good track record. My advice to you is to read our policies, particularly the content policies WP:A and WP:NPOV, and 3RR, and WP:CIV, and get on with increasing your edit count to the actual encyclopedia rather than to talk pages. SlimVirgin (talk) 06:56, 2 March 2007 (UTC)
For some background on this having dealt with this user myself I'd say that a 3rr block doesn't surprise me. I had first encountered the user when he was in a revert war with another user on the Karate article. I had warned the user about our 3rr policy and he responded by leaving personal attacks on my talk page telling me to "take your POV crap elsewhere" [5] Subsequently I blocked the user for personal attacks, I reported the block on AN/I because it was an attack on myself and I wanted it double checked by other admins. Some administrators thought it was to early to block and another administrator User:FayssalF had also disagreed and unblocked. After this same user continued to make attacks against an IP editor and the same unblocking administrator Fayssal blocked him again for personal attacks. After that he proceeded to continue to remove warnings and block templates from his talk page. He refused to stop and I had no option but to protect the page. Myself and Fayssal set up a "deal" where we told him to archive everything with block messages and warnings and try and start over. I was hoping this advice would be taken seriously, unfortunately it seems it has not. After reviewing this I think this particular block was a correct one. The user was edit warring on that page, the version he was reverting to is not relevant to the discussion. Furthermore it is quite obvious from previous occasions that this user does not heed warnings against edit warring. Let's not let wikilawyering by frequently blocked users tie the hands of our administrators.--Jersey Devil 12:24, 2 March 2007 (UTC)
Melonballmonster didn't raise the policy question, I did, because two admins I respect disagree with what I thought the policy was saying in practice, and that worries me.
Regardless of the particulars, I'd like to discuss disambiguating the policy, because as it's being stated above, I think I've violated it a few times, and I've seen a whole lot of others do the same (including other respected admins...), and that leaves me more than a bit worried.
Part of the problem with legislation-by-precedent is that the cases which reach the discussion stage or appeal stage are generally terrible cases in some manner. Melonballmonster's case is terrible due to the other stuff, but the point remains - where do we draw the line on reversions starting? Thanks. Georgewilliamherbert 19:49, 2 March 2007 (UTC)
George, can you cite examples of where you or other admins have violated the policy because of the ambiguity of the time-frame issue? SlimVirgin (talk) 20:51, 2 March 2007 (UTC)
I would have to go looking for a specific example, and it's not going to be anything recent, but I can recall situations where there were contentous things and moderate edit warring with ongoing discussions, where I did three clear reverts and stopped, and had done other non-immediate-revert edits prior to that. All the examples I can find recently involved apparent vandalism and don't exactly count, but I bet if you go back more than 6 months I did something along those lines. Georgewilliamherbert 21:50, 2 March 2007 (UTC)
I'm not quite sure what your main point is, but if it's the issue of timeframe (that the version reverted to was from 35 days earlier by the same editor), that's a matter for the judgment of the blocking admin. 3RR violations aren't totally objective things as some people believe: there are often nuances that the admin has to evaluate, the timing of the version reverted to being one of them. In this case, it was clear cut: the blocked editor was the one who had made the edit, and it was the same edit that he reverted to 35 days later, after being warned not to. In other cases, it might be less clear cut: if the version reverted to had been from six months ago and a different editor had inserted it, then it might be something the admin wouldn't want to take into account. SlimVirgin (talk) 22:21, 2 March 2007 (UTC)
If I may, I'd like to reiterate my understanding of the 3RR 'version reverted to', that I also made on George's Talk page. As I understand it, the intent of the 'version reverted to' is simply to show the editor's preferred version - it itself is not part of an 'edit war'. IOW, assuming his/her opinion about things is fairly stable, that opinion could easily remain the same for months, if not years. It is not strictly needed per WP:3RR, but it is required on the WP:AN3 page to make the reports as clear cut and as conclusive as possible. AFAIK, once you can show the above report, that an editor inserted his favored word (as demonstrated by an older 'version reverted to') into an article four times within 24 hours, each time being reverted by others, it is clear and unambiguous violation of 3RR. The important issue here is to decide if the first 'revert' of the four is an actual reversion, and the way to determine that, IMO, is to compare it with the version reverted to from the past, demonstrating the editor's position on that specific wording or issue. Crum375 22:40, 2 March 2007 (UTC)
Ladies and/or Gentlemen, with all due respect I did not leave these comments to prove that I had a steller wiki record or to ask for the wikian of the month award or to claim I've made a million edits, etc.. Please do not equivocate the actual issues that lead to the block in question which is the topic in question. The facts and arguments that deal with the substance of this incident still awaits a substantive discussion.
Sainjust, I must point out that you were simply wrong and I still question the NPOV-ness of your involvement in the incident which you have brought up. You have twisted the facts in your summary and I could give a lengthy rebutt but this is not the right place for that which is a point that has passed you.melonbarmonster 17:28, 2 March 2007 (UTC)
Melonbar, this is not the place to discuss your block any further. SlimVirgin (talk) 20:51, 2 March 2007 (UTC)
Nor my wiki record, experience, etc.. I just wanted to leave my 2 cents in good faith without any disrespect since I thought this was a discussion about possible policy ambiguities which arose out of my unblock request and review.
Question posed by George still remains for discussion: "does a first recent edit which partially restores and partially modifies detailed wording which an editor had originally created but last edited a month previously count as an initial 'revert' for purposes of 3RR?"melonbarmonster 22:15, 2 March 2007 (UTC)

It's a clear violation, and there's no "statue of limitations" on the version reverted to. Jayjg (talk) 18:19, 4 March 2007 (UTC)

Yah. I recently had an edit warrior revert a specific piece of text to something they insisted on two years ago. Some edit warriors have long memories, they want that "version reverted to" to be stale, they think it gives them 4RR. It exacerbates the first mover problem to ignore it. SchmuckyTheCat 03:39, 5 March 2007 (UTC)
The time-frame isn't the issue here. If the same editor makes the same edit, that's a clear revert without question even if it's months. However, the question being posed is when does an edit becomes a revert when partial restoration of previous versions are made? George posed it well above. melonbarmonster 04:28, 5 March 2007 (UTC)
Sometimes reverts hinge around restoration of one particular fact. If an editor repeatedly inserts that Kyoto is the Capital of Japan, while making other unique constructive edits it is a revert. SchmuckyTheCat 06:03, 5 March 2007 (UTC)
"Sometimes" is where the issue lies. In what cases is it a revert(e.g. your example with Kyoto being capital of Japan) and in what cases is it not a revert? Clearly, not ALL reuse of a word(as was the case here) or an edit element in a new variation is not a revert. If that were the case, it would create a ridiculous situation in which editors can only try using a word or propose an edit element once. You want editors to heed each others' disagreements and concerns by proposing novel compromises and proposing new variations of past edit versions.melonbarmonster 01:37, 6 March 2007 (UTC)
I'd say that any time you are reverting to an edit three or four times in a 24 hour period, you are at risk of being in precisely the sort of edit war that 3RR was designed to prevent. If your point is that the policy allowing flexibility on the part of the blocking admin makes it harder for people to edit war, and may make them consider their actions more carefully, my response would be "Yes. Exactly. We want that." Nandesuka 01:48, 6 March 2007 (UTC)

To add a minority report. I disagree with the analysis made by Crum375, Jayjg and SchmuckyTheCat, I think that time frame is relevant. If the original edit was more than a month ago (picking an arbitrary statute of limitations) then I would not consider the first edit to be revert but a new edit. The whole point of this policy was to give the blood a chance to cool. If an editor still thinks that they were correct and enough time has passed that this is a cold blooded change, then I do not think that it can be called a revert to introduce the change. --Philip Baird Shearer 20:15, 20 March 2007 (UTC)

Summary: I was wrong

I think I will summarize the above as "my interpretation was not consistent with admin standards as a whole" and overly-liberal. Information noted, thanks everyone. Georgewilliamherbert 02:41, 6 March 2007 (UTC)

Persistent once-a-day reversions?

Is there a rule against persistently reverting once a day, thus resulting in a (slow moving) edit war? I refer to [6] where as you may note there is a very slow revert war between one user (User:Certified.Gangsta) and many other users (including myself) stretching back, I don't know, months... where User:Certified.Gangsta would revert once a day, then get reverted on the same day by some other editor, then repeat the same process the next day. --Sumple (Talk) 01:51, 8 March 2007 (UTC)

There doesn't have to be an automatic procedure for everything. There's plenty of latitude in 3rr for an admin to block this user for that. Or you could start a user-conduct RfC. --Trovatore 02:11, 8 March 2007 (UTC)
Or, to put it another way, WP:DE can lead to blocking even if the 3RR isn't being violated... as long as a block is really necessary. Mangojuicetalk 02:47, 8 March 2007 (UTC)
Thanks for the pointers. --Sumple (Talk) 04:55, 9 March 2007 (UTC)

3RR, non-identical revisions, and other joyous things

Current link, perma link. Anyone care to weigh in on this? Essentially, it looks like this user violated 3RR as he wholly or partially undid the edits of other users more than three times within 24 hours. There's debate over whether the undoings constitute "reverts", though. -- Consumed Crustacean (talk) 04:16, 9 March 2007 (UTC)

I think the issue is when does an edit become a 'revert'? Or what distinguishes an edit from a revert? Crum375 04:27, 9 March 2007 (UTC)
And I still hold that these are reverts. They're undoing other editors edits (not randomly, either), and are pulling the whole or part of the page back to a previous revision. -- Consumed Crustacean (talk) 04:46, 9 March 2007 (UTC)
Note: considering the user has pretty much completely stopped the warring and such, I'm not looking for a block. It's only a clarification / solidification of the definition that I'm seeking here. -- Consumed Crustacean (talk) 05:01, 9 March 2007 (UTC)
Three of the four are unquestionably reverts, they each are identical to earlier versions of the page. The fourth is a revert with two words changed from earlier versions, and the section in question is identical to versions before that. --Milo H Minderbinder 14:56, 9 March 2007 (UTC)
I think the question here is more generic; what defines a 'revert'? If the sole criterion is that each edit somehow roughly matches an earlier one (but all are different), especially if we are talking about single word changes and/or punctutations, we are getting very close to just regular (possibly tendentious) edits. If we start flagging regular edits as reverts, we'd be blurring the clearness of regular reverts, where the editor seems to have a clear agenda, always re-introducing the same old version that has been rejected by others. Crum375 16:38, 9 March 2007 (UTC)
Again, and again, and again. The definition on WP:3RR is that a revert is an edit that wholly or partially undoes the edits of another user. It doesn't matter if the revision reverted to is identical. These edits were all intended to undo any effort to join two sentences, so even by your definition they seem to be reverts. -- Consumed Crustacean (talk) 17:03, 9 March 2007 (UTC)
We clearly also include as 'revert' the addition of a word (for example), so we would almost include any edit as a revert, potentially. So where do we draw the line? It seems to me that the only reasonable distinction would be when there is a clear 'version reverted to'. Having different versions for near-single word 'reverts', that only approximately match the edits, seems to me to be a stretch, and very close to regular edits. I welcome other opinions here. Crum375 17:10, 9 March 2007 (UTC)
A revert is an undoing. If different people repeatedly remove a word, the undoing of those edits (i.e. by readding the word) constitutes reverting. Hence your comparison doesn't seem to make any sense. If these were original ways of approaching the problems, such as completely rewording the paragraph in a new way each time, they would not be reverts. They, however, are not; the user is distinctly restoring a previous version. -- Consumed Crustacean (talk) 17:26, 9 March 2007 (UTC)

(undent, multiple edit conflicts) The addition of a word is a revert if that addition has happened previously and been reverted. An addition is an edit if it's adding a word for the first time. And as 3RR policy specifically says, reverts don't all have to be to a single version. Have you looked at the diffs? Three of the reverts are identical to earlier versions, not "only approximately match". Here: [7] [8] [9]

"A revert means undoing the actions of another editor, whether involving the same or different material each time." That's what the page defines as a revert. The difference between a "regular edit" and a revert is that a regular edit creates a new version of the content in question while a revert takes it back to an older version. In this particular case, there's no question that three of the edits took the page back exactly to an earlier version (not "roughly matches" but a revert to the exact same version). Can we at least agree that those three are reverts? The fourth doesn't take the entire page exactly back to an earlier version of the full page (because other edits have been made in the meantime which weren't reverted), but it does take the paragraph in question back to exactly the same version it had been in previously ("same old version that has been rejected by others"). I'd argue that's a revert as well (although probably a partial revert, which still counts towards a 3RR violation). --Milo H Minderbinder 17:27, 9 March 2007 (UTC)

I would agree that if someone prepared a well presented 3RR report, showing clearly that each alleged revert takes us back to a previous version that was previously rejected by others, it could possibly fly. In this specific case, FWIW, no such effort was made, at least not initially. Additionally, as you note, the 4th revert is not even a perfect match against any previous version. The question is how much effort should be expended on one specific report - I would argue that if the edit pattern is complex, we should shoot for a disruption report and not try to shoe-horn an ill-fitting case into the supposedly straight-forward WP:AN3 page. It seems to me that ill-fitting cases (or otherwise improperly prepared reports) that are filed there result in a lot of time and effort invested, that could be better applied elsewhere. Crum375 20:30, 9 March 2007 (UTC)
Indeed it was poorly prepared, though I did eventually get around to properly describing it. The discussion left that and went on to the validity of labelling the edits as 'reverts' and such though, which is the only reason I brought it here. -- Consumed Crustacean (talk) 23:31, 9 March 2007 (UTC)
  • I don't think we need a strict formal definition of "revert" because that only encourages wikilawyering. The idea behind the rule is clear. >Radiant< 12:12, 13 March 2007 (UTC)

Is this intentional?

A user can remove a link, and I restore it because it was useful. He picks another link and I restore it. Repeat twice more. Am I correct in that he has not made any reverts, and I have made four? --NE2 10:30, 11 March 2007 (UTC)

A revert is any action that undoes the action(s) of another contributor (or other contributors). Someone obviously took the action of adding the link in the first place; simply removing it is undoing that work. --bainer (talk) 13:06, 11 March 2007 (UTC)
However, the link was there from the time that section was written. Are you saying that Indon made 10 reverts on Myrtle Avenue Line (surface)? --NE2 18:44, 11 March 2007 (UTC)
Removing or re-adding a link is a revert if it undoes the actions of another editor. If the user adds a different link each time, the it may not be a revert, but it still may be disruptive. HighInBC(Need help? Ask me) 18:47, 11 March 2007 (UTC)
Even re-adding a different link each time, as long as each link has already been specifically rejected in the past by another editor, would still constitute a 'revert'. Not all reversions must be to the same revision, so in theory one could have 4 different added links, each previously rejected, adding up to a WP:3RR violation if done within 24 hours. Crum375 18:55, 11 March 2007 (UTC)
Indon was removing links based on (what I believe to be) an incorrect interpretation of the manual of style. Was each of his removals a revert? --NE2 19:05, 11 March 2007 (UTC)

MONGO RfA Exception to 3RR

Wikipedia:Requests_for_arbitration/MONGO#Links_to_attack_site gives an additional exception that should be added to this policy. --Ronz 21:40, 14 March 2007 (UTC)

I assume that would fall under the category of "reversion of vandalism," which is already mentioned, but I'm not sure...--Vox Rationis (Talk | contribs) 22:25, 14 March 2007 (UTC)
Pretty much, yes. >Radiant< 09:16, 15 March 2007 (UTC)

3RR violations possible even if no one responds on the discussion page?

Recently, I was warned not to run afoul of 3RR and apparently skated close to the edge of being blocked. This seemed a little odd to me since, up to that point, I had never recieved a warning of any sort on Wikipedia, and have never had a dispute on WP that I was not more than willing to resolve through discussion, even if it meant dropping a legitimate concern just to prove good faith.

But then came this:


This matter is a little troublesome for the following reasons:

  • The entire "conflict" related to an issue that was actually initiated in article talk by other contributors besides me (and with whom I had no prior relationship);
  • I thought they made a good point, and saw fit to address the issue, meticulously documenting my rationale every step of the way (since it is a controversial article);
  • I made (what I considered to be) a reasonable NPOV edit, and requested discussion from anyone who might have found fault with it;
  • Every single edit summary I made included a request for comments on Talk;
  • I made only two "undo-style" reverts (because my edits were undone without discussion, and with vague edit summaries that didn't even seem to address the issue initially raised)
  • The very first "discussion" of any sort I recieved was a 3RR warning (on my user talk page)
  • All but two of the "reverts" I was accused of were what I genuinely considered good-faith efforts to resolve the (up till then) entirely vague complaints that "overruled" me in edit summaries, (I *thought* I was properly addressing their summaries instead of racking up a toll on the 3RR meter)
  • I was later informed that introducing *any* prior wording constitutes a "revert" (I was under the assumption that only the "UNDO"s counted as reverts)

This seems manifestly inconsistent with the spirit of WP policy regarding consensus and good faith. Under facts like this, users of a particular inclination could overrule someone's edits without even bothering to look at the discussion page, let alone refute their rationale, or dispute their credibility, or explain why they should stop editing (for example, a current effort at consensus is underway, or the article is undergoing substantial revision).

Yes, I am only one person, but do "drive-by" undiscussed deleletions with garden-variety buzzword-only edit summaries ("POV not allowed", "weasel words violating WP:OR") rise to the level of the sort of content intended to be protected by 3RR? Shouldn't contributors at least be required to participate in a pre-existing discussion before the 3RR meter starts ticking against a user?

This is especially of concern since 3RR violations apparently entail a requirement not to introduce *any* prior language, which could easily be "accidentally" violated (whereas an "UNDO" is much more obviously avoidable).dr.ef.tymac 23:01, 29 March 2007 (UTC)

NOTE: In no way do I impugn the good faith of any contributors involved in the aforementioned matter. Reasonable people can see things differently, and any concerns I raise here are strictly matters related to interpretation of existing WP policy alone.

It's bad when editors don't discuss on the talk page, but 3RR isn't the place to address that, nor is refusal to discuss an exemption from 3RR. If you have problems with editors doing that, I'd recommend finding another way to get other editors or admins to look at the situation, such as WP:RFC. --Minderbinder 14:07, 30 March 2007 (UTC)
Your response makes sense, although it suggests Wikipedia:Edit war, WP:HEC, WP:DR (and perhaps others) could use substantial clarification, since "taking it to the discussion page" is the first-phase 'antidote' to "edit warring" (exactly what WP:3RR is in place to prevent).
  Failure to pursue discussion in good faith shows that you are trying to escalate the 
  dispute instead of resolving it. This will make people less sympathetic to your 
  position and may prevent you from effectively using later stages in dispute resolution.
(excerpted from WP:DR). If good faith pursuit of discussion is not a sufficient basis for disclaiming an accusation of "edit warring," and failure to pursue discussion does not prevent a party from invoking 3RR against someone who does pursue discussion, it would seem some clarification is indeed in order. Just some thoughts ... thanks for your response. dr.ef.tymac 16:36, 30 March 2007 (UTC)
To cut to the heart of the matter, it's good that you were trying to encourage discussion. Nonetheless, when edits get to the point of 3RR violation, merely calling for discussion isn't enough. Sometimes, you just have to let the article stay in the state you don't like until things really get resolved: just because there has been discussion, or you are willing to discuss, doesn't justify quick flip-flop reverting after a certain point. As for what to do in the situation, I'd say pursue discussion more if early attempts don't work out. Go to a user's user talk page (people don't always look at edit summaries), start discussion on the talk page. If the situation is like one lone editor trying to hold consensus hostage, take it to WP:ANI; if it's more complex, start dispute resolution. Mangojuicetalk 12:56, 31 March 2007 (UTC)
Duly noted. Thanks for your response and for taking the time to review my question. Regards. dr.ef.tymac 13:52, 31 March 2007 (UTC)

Suggestion for a Policy Revision

The current policy is sort of rough. Based upon the way it is written a person can get a block even if they are not really edit warring. They might simply wander on a page and make 4 edits that they think are ok but that remove four previous edits. This is really too rough.

Furthermore, a block from the whole project might be too much, particularly for the first offense or rare offenses.

Why not a ban from that particular article... say for a week, instead of a block from the whole project for a day or two? --Blue Tie 23:34, 14 April 2007 (UTC)

Just making edits don't necessarily count as reverts. Also, if the edits are in proximity or consecutive, they count as one revert. There are plenty of caveats that allow good-faithed editors to make good contributions. ~ UBeR 01:03, 15 April 2007 (UTC)
Well, if your edits are deletes of something that someone (or a variety of people) have just added, you are reverting. Yes, its true. I think I nearly got a block for something like that not too long ago!
But anyway... I still think a ban from an article (you might get heated on one article but not on all of them) is a better way to go than a block from the project. You could go and do good editing elsewhere.
But I can see others have proposed similar or other things here and none of them have seen any sun or air, so this will die too. Wikipedia has no good process for proposing changes to policies or making them happen. Its ridiculous. --Blue Tie 02:00, 15 April 2007 (UTC)
That's not true. Lots of policies have evolved productively over time. This one is rather entrenched, but I think there are some useful ideas out there. I had mentioned taking a look at this whole area in my RfA statement, so let me think about this for a bit longer and I may make a proposal. Newyorkbrad 02:05, 15 April 2007 (UTC)
Agreed: Although I withhold comment on the point about the policy change procedure (that's a totally separate issue), I think Blue Tie raises some legitimate concerns about 3RR. It seems some WP contributors make very expansive application of what constitutes a "revert" and this expansive application, combined with a strategy of "block first, ask questions later (or never)" can result in prejudicial outcomes against contributors who are entirely acting in good faith, willing to discuss and substantiate their actions, and reach consensus. Proposed blocks against well-meaning editors who are not otherwise disruptive, intransigent, self-promoting or otherwise abusing WP resources should at least be met with the highest degree of scrutiny and caution. dr.ef.tymac 03:09, 15 April 2007 (UTC)
It may be worth remembering that admins are not jerks, and have a measure of common sense. If someone is clearly not revert-warring, they won't be blocked under the 3RR. It may be worth mentioning on the page more strongly. Currently, we have "Since the rule is intended to prevent edit warring, reverts which are clearly not such will not breach the rule." which does capture the basic point, but could be worded more strongly. Also, there could be more emphasis that breaking the rule may lead to a block (as opposed to "does" or "must" or "should", et cetera). Mangojuicetalk 02:53, 16 April 2007 (UTC)
And let me add that currently, there is not the technology in place to automatically block a user from a single article but not from editing generally. There are such things as bans but those have to be manually enforced. Mangojuicetalk 02:55, 16 April 2007 (UTC)

Arbitration Committee precedents

Proposed section:

Statements of principle

  • Edit wars or revert wars are usually considered harmful, because they cause ill-will between users and negatively destabilize articles. Editors are encouraged to explore alternate methods of dispute resolution, such as negotiation, surveys, requests for comment, mediation, or arbitration. When disagreements arise, users are expected to adhere to the three-revert rule and discuss their differences rationally rather than reverting ad nauseum. "Slow revert wars," where an editor persistently reverts an article but technically adheres to the three-revert rule are also strongly discouraged and are unlikely to constitute working properly with others.
  • The three-revert rule prohibits editors from reverting an article more than three times in any 24-hour period, except in cases of simple vandalism.
  • This rule should not be construed as an entitlement or inalienable right to three reverts, nor does it endorse reverts as an editing technique.
  • The rule applies only to individuals, not groups.
  • The term "revert" as used in Wikipedia policies and guidelines is intended to include both absolute reverts (where versions differ not at all) as well as de facto reverts (where versions are only very slightly different). Attempting to avoid being accused of reversion by making very minor edits that are then edited out again is in bad faith and against Wikipedia policies and guidelines.
  • Sockpuppets and anonymous editor accounts may not be used to evade the three-revert rule.
  • It is expected that editors, when reverting, will provide an explanation for doing so in the edit summary.

Previous penalties relating to principle

  • In cases of edit-warring, revert limitations are applied in which reverting is restricted and violators can be blocked for specified periods.
Cases involving this principle

Added section to further explain the way WP:3RR is enforced/interpreted by the Arbcom. Anynobody 00:33, 23 April 2007 (UTC)

The Arbitration Committee refers to policy, not vice versa. Jayjg (talk) 01:27, 23 April 2007 (UTC)

Correct me if I'm wrong, but in a dispute over interpretations of how WP:3RR should be applied, the committee makes the final interpretation.

The precedents further explain the policy. Anynobody 02:01, 23 April 2007 (UTC)

I agree with Jayjg. If we include ArbCom decisions inside policies, and then ArbCom uses the policies in its decisions, we get into a circular argument, which is logically flawed. When Congress makes laws, it doesn't cite court cases - court cases cite Congress. The same should hold here. Crum375 02:08, 23 April 2007 (UTC)
We can certainly benefit from their wisdom though. Not sure about the specifics of this case though, have not looked. HighInBC(Need help? Ask me) 02:12, 23 April 2007 (UTC)
What we can benefit from is their turns of phrase and eloquent expression. The ArbCom is regularly very adept at expressing policies in good ways, and we should certainly update the policy pages if the expression of the policy can be improved. But we mustn't forget that the ArbCom doesn't make policy.
Note that the policy as it is currently worded covers pretty much everything you've listed there. It doesn't list all of the types of dispute resolution, rather it links to WP:DR. Really all that's there that isn't in the policy is the recommendation to use edit summaries when reverting. --bainer (talk) 07:23, 23 April 2007 (UTC)

It's like a judge in the US making a ruling and citing a precedent judgment in a similar case. For example one of the cases clarified that whether the revert is full or partial if it involves information in an edit war it counts toward this policy. Anynobody 02:45, 23 April 2007 (UTC)

Precedents can be broad or narrow, and it's not up to you to decide "case law" in this matter; rather, the Arbitration Committee will decide which precedents are relevant and which are not. By the way, when judges cite previous rulings, they do just that; however, legislators don't then go and change the law based on the legal rulings, or citing them. Jayjg (talk) 03:17, 23 April 2007 (UTC)

Jayjg (talk) I noticed that you devoted no space in your reply to explain why it would be a bad idea to show the cases I mentioned. I just posted the disputed section on this talk page and I think I understand why you don't want it out there:

*HistoryBuffEr and Jayig

Is this why you are against posting the precendents? Anynobody 03:39, 23 April 2007 (UTC)

The information is relevant and useful. Perhaps it can be incorporated in a fashion better than just tacking it on the end though. HighInBC(Need help? Ask me) 03:41, 23 April 2007 (UTC)

I agree, but I want to find out if Jayjg (talk) is embarrassed or uncomfortable with his case being mentioned. (I think the bullet points should go in either way, but the cases can be probably be skipped. Anynobody 03:47, 23 April 2007 (UTC)

ArbCom precedents may be useful for Talk page discussion, but in general they should not be relied upon or referenced inside policy pages. As mentioned above, ArbCom interpret policies to reach their decisions, so we cannot reference them inside the actual policies, or we'll have circular arguments. If you have some specific change you'd like to propose, you can use any rationale you want, including ArbCom precedents, to justify it, but the ArbCom precedents don't belong in the policy itself, and their 'principles' should be considered on an individual basis. Crum375 12:52, 23 April 2007 (UTC)
Exactly so. As I said before, legislators don't go about making laws referring to the previous decisions of judges. Jayjg (talk) 16:37, 23 April 2007 (UTC)

Respectfully, I disagree with the assertion that they do not belong in the article. Would anyone mind if I posted a question about this to the Arbitration Committee talk page to resolve this? Anynobody 00:26, 24 April 2007 (UTC)

Go ahead. Remember that it is the community that ultimately decides what goes into policies, but ArbCom is certainly part of the community. Crum375 00:34, 24 April 2007 (UTC)

I have asked for their advice, here is a link to the post:Wikipedia talk:Arbitration Committee#Role of AC relative to policies like 3RR. I fully understand that this is a consensus run community, however I have difficulty reconciling saving arbcom decisions if they are irrelevant. Anynobody 03:00, 24 April 2007 (UTC)

Anynobody, you are incorrect in concluding that ArbCom decisions are 'irrelevant'. They are, in fact, very relevant and most important, and they can and should help us in shaping our policies and guidelines. But the proper place for their use is on the Talk page, where they can be considered and discussed by the community, as an important (but not exclusive) input into the community's consensus. It that consensus which goes into the policy. Crum375 03:08, 24 April 2007 (UTC)

I'm not saying they are irrelevant, I think they are most relevant which is why I'm trying to get them included in the policy page as further explanation. Anynobody 03:13, 24 April 2007 (UTC)

Before you add precedents to this page, it might be worth pointing out that the Arbitration committee is not a court of law, and does not set precedents. They specifically reserve the right to contradict themselves in later decisions, if it turns out their earlier ideas were wrong. Just thought you'd like to know that ;-) --Kim Bruning 03:16, 24 April 2007 (UTC)

Yes, and that's one more reason why policies and guidelines should be able to stand on their own legs. Crum375 03:19, 24 April 2007 (UTC)

I'm also not saying they are set in stone, no two situations are identical of course. However, the decisions give guidance on questions like "what exactly is a revert?". Anynobody 03:27, 24 April 2007 (UTC)

ArbCom decisions are a most valuable input into our own decision making process. But their use should be on the Talk page, as one more input for our discussion. The policy page should reflect our consensus and other relevant policies, not references to ArbCom decisions. Crum375 03:41, 24 April 2007 (UTC)

Perhaps an example of what I mean would explain my intentions better, WP:3RR states:

A revert, in this context, means undoing, in whole or in part, the actions of another editor or of other editors. This can include undoing edits to a page, undoing page moves (sometimes called "move warring"), undoing administrative actions (sometimes called "wheel warring"), or recreating a page.

An editor does not have to perform the same revert on a page more than three times to breach this rule; all reverts made by an editor on a particular page within a 24 hour period are counted.

Note that consecutive reverts by one editor are often treated as one revert for the purposes of this rule.

"Do minor changes count as a revert done through the history page?" This question is not addressed above, but the arbcom said that:

The term "revert" as used in Wikipedia policies and guidelines is intended to include both absolute reverts (where versions differ not at all) as well as de facto reverts (where versions are only very slightly different). Attempting to avoid being accused of reversion by making very minor edits that are then edited out again is in bad faith and against Wikipedia policies and guidelines.

Based on what the committee said above, a revert can be full or partial to count toward the WP:3RR policy. Anynobody 22:49, 24 April 2007 (UTC)
I am copying my response from Wikipedia_talk:Arbitration_Committee#Role_of_AC_relative_to_policies_like_3RR, but we really should keep this in one place:
Using ArbCom's decisions and interpretations as an input to productive Talk page discussions, and to help tighten or clarify the policies, is highly encouraged. But it should not be the sole input, nor should ArbCom decisions be quoted or referred to from the policy itself. Crum375 23:29, 24 April 2007 (UTC)

Disruptive editing and the 3RR rule

I've got a highly disruptive user who keeps insisting on moving an infobox from the top of a page to the bottom of a page, Can I take it that I am protected from the 3RR rule if I revert such an edit.

The page in question is Anomalistics

perfectblue 13:57, 3 May 2007 (UTC)

There is no excuse for violating 3RR except when dealing with persistent vandalism. --ScienceApologist 14:10, 3 May 2007 (UTC)
It seems fairly clear that this is a dispute simply about where the box should go. This is definitely not protected from 3RR. Calling SA "highly disruptive" is not promising for peace and harmony - try to be more neutral. Meanwhile, the best procedure would be to try to establish the policy re box positions - if the is one - via the talk page William M. Connolley 14:27, 3 May 2007 (UTC)

Attempting to clarify 3RR with a hypothetical

These questions are based on a hypothetical. Any similarity to existing users or Wikipedia content is unintentional. The fake article in question is "Kinessar theory" which is a minority viewpoint that has been discredited, but is still notable for a WP article (compare "Flat Earth").

Consider the following hypothetical edit history (earliest edit at bottom):

* Echo5     (Talk|) ( Revert: Alice, you are in violation of 3RR, stop now ) 
* Alice     (Talk|) ( re-add: the entire *article* is about the minority view![see talk page!])
* Delta4ce  (Talk|) ( Revert: Undue Weight; eyanthon is minority view ) 
* Alice     (Talk|) ( re-add w/NYT wording: "the first sporenol of kinessar theory was eyanthon" )
* Charlie33 (Talk|) ( Revert: POV; NYTimes says nothing about "widely-favored" ) 
* Alice     (Talk|) ( re-add with alternate cite to NYTimes [again, please see talk page.] )
* Baker22   (Talk|) ( Revert: delete POV pushing with blog as unreliable source see [WP:SPS] )
* Alice     (Talk|) ( re-add with cite to FooBlog [please see discussion page for rationale] )
* Baker22   (Talk|) ( Revert: weasel worded POV pushing in violation of [WP:OR] )
* Alice     (Talk|) ( added "eyanthon was the most widely-favored sporenol of kinessar theory"  )

Note: Under this hypothetical, assume the issue has yet to be addressed on the talk page (making the matter of "pre-existing consensus" ambiguous); assume that Alice has made each change in a good-faith effort to directly remedy the problems mentioned in the edit-histories; assume that Alice is the only one who is on-record on the discussion page; assume all Alice has to go on is the edit summaries of the reverting editors.

Question: Has Alice violated 3RR? If so, is a block justified? Comments and feedback welcome. dr.ef.tymac 15:45, 3 May 2007 (UTC)

Assuming the same material has been added each time, it seems plain that Alice has violated 3RR; so much so that its not clear why you've brought this up William M. Connolley 15:59, 3 May 2007 (UTC)
@William M. Connolley Clarification: each edit she does is a direct attempt to correct the deficiencies stated in the edit summaries.[1] Assuming you still hold to your "yes" after my clarification, What about the second part of my question? dr.ef.tymac 16:11, 3 May 2007 (UTC)
Ah, my mistake. If each time the revert is "text string (ref)" but the ref varies each time, then each one is a partial revert: changing the ref each time doesn't save you William M. Connolley 16:33, 3 May 2007 (UTC)
Ok, so then your answer is yes to "3RR violation" and yes to "block is justified"[2] Duly noted. More comments from others welcome. dr.ef.tymac 16:50, 3 May 2007 (UTC)
Yes, unless there are Special Circumstances. I'm curious what you are referring to really, though William M. Connolley 19:30, 3 May 2007 (UTC)
The hypothetical pretty much says it all. As far as my own opinion, it seems rather odd (to say the least) that someone should be blocked from WP for adding a cite to content that was previously removed for not having a cite, or for adding a more authoritative cite to content that was previously removed for not having a reliable source. This is why I asked for clarification. dr.ef.tymac 21:40, 3 May 2007 (UTC)
Looks like a violation to me, any action that undoes another editors edit is a revert. HighInBC(Need help? Ask me) 21:37, 6 May 2007 (UTC)
So, essentially, even if an editor removes content on the grounds that it is uncited original research, a respondent is not entitled to demonstrate that the assertion "uncited" is incorrect by providing a citation; at least, not if they are close to the 3RR limitation and wish to avoid being blocked for addressing the claim and improving a percieved deficiency. What then is the purpose of providing edit summaries when removing other's contributions at all? dr.ef.tymac 01:30, 7 May 2007 (UTC)
  1. ^ 1) First time she adds content [C1]; 2) next she adds a cite [C1+citeB] (after someone complains WP:OR); 3) next she substitutes with a more authoritative cite [C1+citeN]; 4) next she rewords the content to exactly match the cite [C2+citeN]; 5) next she re-adds the contribution from (4) after disputing an unsubstantiated claim of "Undue Weight".
  2. ^ If I am misinterpreting, please feel free to clarify.

Clarification added to exemptions section

I have added a clarification to the exemptions section, which makes it clear that users reverting in their own userspace are not exempt from the removal of copyvios, libel, etc. If someone can phrase it more clearly than I did please go ahead, but reverting me would make no sense as this is already common practice. (And policy should ideally codify common practice.) Picaroon (Talk) 21:32, 6 May 2007 (UTC)

Makes sense to me. HighInBC(Need help? Ask me) 21:35, 6 May 2007 (UTC)
Dreftymac, I see you have reverted this asking for an explanation on the talk page. Which part did you not understand? HighInBC(Need help? Ask me) 01:37, 7 May 2007 (UTC)
Clarify: as the original contributor noted if someone can phrase it more clearly ... I agree with him that this is a relevant consideration, and believe the ambiguity of the added wording not only merited consideration, but merited removal pending the precise "more clear" phrasing. dr.ef.tymac 01:52, 7 May 2007 (UTC)
Is this more clear? Picaroon (Talk) 01:58, 7 May 2007 (UTC)
The point I was addressing is relatively minor. I sense a disturbance in the force, so I will dissipate it and get back to you in a sec, thanks. dr.ef.tymac 02:01, 7 May 2007 (UTC)
Follow-up: Just to note, I'm pretty sure the intent of the original contribution was clear to anyone following along here, and it's probably safe to assert the intent was entirely consistent with the existing policy, and therefore entirely uncontroversial. The relatively minor issue was that of phrasing (as the original contrubitor indicated of his own initiative, and I agreed).
Although the issue in isolation is minor, I think the "cognitive discomfort" caused by the newly-added phrasing was a result of the incremental progress of the "Exceptions" section itself. We all have a justifiable reluctance in making pro-forma changes to WP policy, especially when such changes might mistakenly be interpreted as an attempt to change the substantive meaning of the policy, rather than merely improve its clarity.
Consequently, I have made a change that, I too, believe is consistent with the substance of the existing policy, and consistent with the intent of the previous contribution, but also attempts to address the "cognitive dissonance" that naturally results when attempting to draft (and read) an exception to an exception to the prohibition against tendentious reversion. My main concern was simply for new users who may come to this policy page with "fresh eyes".
To demonstrate good faith, and a full recognition of the appropriateness of the original contributor's intent, I have User:Dreftymac/Scratchpad002|edited the changes in my own user space, for review by any and all interested parties, prior to proposing they be added to the policy page.
See User:Dreftymac/Scratchpad002|here. Regards. dr.ef.tymac 03:01, 7 May 2007 (UTC)
Update: struck out links as they are no longer available externally, and instead are now integrated into the policy text itself. dr.ef.tymac 19:52, 29 May 2007 (UTC)
The "Exceptions" section was significantly consolidated in my rewrite of this page in January this year, see here for the diff and here for the version prior. The current version is thus dramatically less verbose, with perhaps a loss of subtlety as a small tradeoff. I wonder whether it's really necessary to mention userspace explicitly at all, and have these problems of expression, given that reverting one's own actions doesn't really count as reverting. --bainer (talk) 06:11, 7 May 2007 (UTC)
The consolidation and rewrite absolutely represent an improvement. It's good to have that reminder for additional context. In any event, unless preferences and opinions suggest the contrary, some additional minor clarifications to the consolidated section seem appropriate. Please review the upcoming edit to the article that should reasonably clarify the section, while still retaining the caveat about user space edits. This is intended to be entirely consistent with existing policy, as well as the purpose of Picaroon's original modification. Thanks. dr.ef.tymac 12:04, 7 May 2007 (UTC)
Follow-up: I've incorporated the proposed change to the main policy page. The rough draft is now deleted from my user space. Comments and feedback of course still welcome. dr.ef.tymac 21:16, 7 May 2007 (UTC)

3RR too vague

I have a number of problems with 3RR. The first is that I don't think it should count as a revert at all if you are removing unsourced material from a page. Other policies state that removing of unsourced material is not only allowed, but encouraged, but 3RR as it stands gives editors no support for making such removals.

Secondly, I think there's a case to be made that material that is added which is clearly in violation of policy should also be subject to removal without triggering 3RR. For example, stuff that is sourced to places that clearly violate WP:RS.

The onus should be on the person adding material to come up with reliable sources, and if he cannot do so, anyone should be able to remove his edits without violating 3RR, while the person attempting to add such material will himself violate 3RR by continuing to add it.

Thirdly, it seems to me there should be more exceptions to 3RR. For example, as the page stands, even fixing a spelling mistake could be construed as a 'revert' and get you a ban. I'm also inclined to agree with an earlier user that reverting material which was added more than 24 hours ago should not be subject to 3RR, for the simple reason that 3RR is there to stop edit warring and it can scarcely be called edit warring when you are reverting something that is more than 24 hours old. Gatoclass 05:52, 9 May 2007 (UTC)

None of this sounds like a good idea. The usual counter argument applies: if what you are doing by reverting is clearly sensible, then you should have no problem finding someone else to do it, rather than breaking 3RR yourself William M. Connolley 08:43, 9 May 2007 (UTC)
I edit pages relating to the Israel-Palestine conflict. If you think the system as it stands works fine, you should try editing these same pages sometime. There are often several editors on both sides revert warring at the same time.
What I am arguing for is a clarification of the 3RR rule to make for less edit warring. There is no reason whatever I can see why it should be considered a revert to remove unsourced material. Nor should it be a problem to remove material referenced to clearly invalid sources. Clarifying these matters at least would help somewhat in preventing unnecessary edit wars, and it would also help maintain article quality. Gatoclass 14:38, 9 May 2007 (UTC)
The three revert rule is intended to prevent edit warring, and overly literal intepretations of the words here are not useful if they ignore this. To take your spelling example, merely correcting spelling mistakes is not edit warring, though a dispute about spelling which involved users reverting each other would be edit warring. --bainer (talk) 09:34, 9 May 2007 (UTC)
Yes but you could be blocked just for doing some rephrasing or improvement of someone else's work who is not involved in the edit war at all. That's crazy. Surely 3RR can be a little better refined than it currently is. Gatoclass 14:54, 9 May 2007 (UTC)
I was just thinking there should be less exceptions. There are other means of resolution than just reverting again. HighInBC(Need help? Ask me) 14:41, 9 May 2007 (UTC)
I agree with HBC and William. It doesn't matter if one side thinks they have a good reason, communication about it is still necessary, or it just leads to an escalation of the conflict. Mangojuicetalk 14:46, 9 May 2007 (UTC)
And what happens when one or more parties simply keep adding unsourced or badly sourced material? What if one party simply doesn't edit in good faith? I edit on the Israel-Palestine pages where emotions run hot, and anyone who has edited these pages regularly can tell you about the abuse and the gaming that goes on. There needs to be a clearer standard so that responsible editors are not penalized for removing irresponsible edits. Gatoclass 14:54, 9 May 2007 (UTC)
If they keep doing it then it is a 3RR violation. Problem solved, unless you are the only person reverting the guy. HighInBC(Need help? Ask me) 15:07, 9 May 2007 (UTC)
No, unfortunately it's not "problem solved". If someone adds some unsourced material, and you revert and then get into an edit war, guess who violates 3RR first? The one who removed the unsourced edit in the first place. This means the responsible editor gets penalized, while the irresponsible editor gets his way.
The irresponsible editor can then run to page protection to get his version of the page "frozen" for a week because of the "edit war". It's this kind of gaming that drives good editors away from Wiki and leaves the bad ones in charge, I've seen this sort of thing happen numerous times. Gatoclass 15:22, 9 May 2007 (UTC)

Hmm. The fact tags seem appropriate for unsourced material, so I'm not sure why that should be deleted. If absolutely no sources can be found regarding that particular line, then it would be appropriate to delete.

I'd also like to point out that there is very rarely a "clear" violation of WP:RS. The reliable source policy is (rightly so), quite vague. .V. [Talk|Email] 15:17, 9 May 2007 (UTC)

Policy says that unsourced material can be deleted at any time. But that is not reflected in 3RR. I'm just arguing for some consistency. If it's legitimate to delete unsourced material at any time, then such deletions should not count toward 3RR. Gatoclass 15:29, 9 May 2007 (UTC)
To answer your question: if someone persistently posts unsourced material, it becomes a different kind of problem. After discussing the issue with them, move on to warning them, and then bring their behavior issue to some wider forum. Ultimately, it will be dealt with properly. The reason endless reverts are a bad way to do this is that (1) you can't fight this kind of fight all by yourself anyway, and (2) if the other party can be reasoned with, they will need to see that other independent users don't agree with them, which means bringing the issue to a wider venue, and (3) with continuing reverts, it's hard for others to edit the article and in the meantime, the version posted keeps flip-flopping. The 3RR is there to make sure you pursue a more useful way of resolving the problem, namely, ask for outside input from admins or otherwise. Mangojuicetalk 15:47, 9 May 2007 (UTC)
The problem is that other means can often be very awkward and time consuming, and meanwhile you have a substandard page which lowers the quality of the project as a whole. If editors knew they couldn't win a 3RR war just by continually adding their unsourced edits, they would be less likely to engage in such wars in the first place, which would mean less edit wars and flip-flopping, not more. Gatoclass 16:35, 9 May 2007 (UTC)
You are describing a situation indistinguishable from most good-faith edit conflicts: User A thinks they are right and policy backs them up, and so does User B. They get into a revert war. The 3RR is there so that the revert war doesn't get too bad. The proper response for those two users, if they can't work it out directly by discussing with each other, is to get outside input so consensus can be found. That's a good way to resolve the conflict. Another way to resolve the conflict would be for an admin to come in and make an immediate judgement about who's following policy and who isn't, and declare a winner. That's a bad way: it doesn't resolve anything, it leads to hurt feelings on the part of the losing side (who will probably feel their loss is arbitrary), it doesn't make any process towards a resolution, it's just a temporary fix. Getting outside input or intervention may be "awkward" and "time consuming" but it is the only way conflicts can really get resolved when there's a deep, good-faith disagreement. So, no, thinking you are right about policy (which is, of course, indistinguishable from being right about policy to the person who thinks so) should not give you license to revert freely, or else revert wars would break out all over the place. Mangojuicetalk 17:13, 9 May 2007 (UTC)
(ec)Well said. HighInBC(Need help? Ask me) 17:36, 9 May 2007 (UTC)
@Gatoclass: You've expressed your concerns and others have addressed the merits and alternatives. Although I myself withhold comment on the merits of your position, a bit of review is in order:
  1. under the current terms of this policy, removal of unsourced controversial claims regarding living persons is expressly permitted, you seem to be claiming that is not enough;
  2. you've asserted issues with wording in other policies and guidelines, but have not asserted why this policy must be changed as the only means of addressing the matter;
  3. you've asserted "akwardness" of other remedies, but given no rationale for why your proposal is the only way to minimize this akwardness, or why this "akwardness" is unduly restricting; and
  4. you've not addressed why adding {{fact}} or {{dubious}} tags represents an inadequate solution to the underlying problems you pose: (i) retention of good editors, (ii) addressing article deficiencies without exceeding a revert limit, (iii) preventing "flip-flopping", (iv) maintaining consistency between the different policies and guidelines.
Because of the scope of the problems you pose, if you really feel strongly about this, one might suggest that your issue extends beyond this specific policy page, and would be better suited for a wider forum. If your primary purpose is to focus exclusively on 3RR as a policy, one might be more inclined to view this purpose more favorably if you can show you've made a concerted and dilligent attempt to exhaust the full range of other options that are more readily available. dr.ef.tymac 17:35, 9 May 2007 (UTC)
under the current terms of this policy, removal of unsourced controversial claims regarding living persons is expressly permitted, you seem to be claiming that is not enough - Drefymac
Yes, exactly. Have there been any problems about the stricter 3RR policy in relation to WP:BLP? Apparently not. So why restrict this particular 3RR exception only to BLP's?
Just a reminder of what other policy and guideline pages say about the matter. WP:RS - Unsourced or poorly sourced edits may be challenged and removed at any time. WP:V - Editors should provide a reliable source for quotations and for any material that is challenged or is likely to be challenged, or it may be removed. WP:A - Editors should provide attribution for quotations and for any material that is challenged or likely to be challenged, or it may be removed. The burden of evidence lies with the editor wishing to add or retain the material.
That is, we have three separate policy or guideline pages explicity stating that unsourced material may be removed. We even have a quote from Wales on WP:V which explicitly states the following:There seems to be a terrible bias among some editors that some sort of random speculative 'I heard it somewhere' pseudo information is to be tagged with a 'needs a cite' tag. Wrong. It should be removed, aggressively, unless it can be sourced. This is true of all information, but it is particularly true of negative information about living persons.
So what exactly is the big deal about reflecting these policies in 3RR? It apparently works fine already for BLP's, and it's only making 3RR consistent with other Wiki policies. And it seems to me it would be only be providing support to responsible editors who are trying to remove substandard material. Gatoclass 18:35, 9 May 2007 (UTC)
To answer you further would involve repetition. HighInBC(Need help? Ask me) 18:36, 9 May 2007 (UTC)
Did someone address why we make an exception for BLP but can't for this rule? My concern would be that any such exception could be abused. A person says disruptively "you haven't provided a reliable source for that," and the other responds "I have too," and then the person trying to exclude information thinks he's entitled to keep deleting ad nauseum. Unfortunately almost all rules can be used for good or for disruption. Another tension here is the assumption/belief of many admins that people should not edit war (by which they mean reverting any good faith edit), ever, at all. Thus, the chess match you're describing isn't supposed to happen. I know that's not really a satisfactory answer in the Middle East section where edit warring continues so prevalently, but I think it's what you're up against. Mackan79 18:57, 9 May 2007 (UTC)
I've found that a good guideline is the so-called "1RR". Revert once, if it's restored, discuss it on the page. It works well in 80%+ of cases. .V. [Talk|Email] 22:05, 10 May 2007 (UTC)
That can also be found at WP:BRD Lsi john 12:47, 13 May 2007 (UTC)

As a general comment in this discussion, note that wikipedia policy works on most of our over 1M pages, but fails on some controversial pages (in the order of 1-3Kpages). It may well be that the 3RR fails to work in some circumstances. If some people have the time to devise a ruleset for controversial pages, that might be interesting. --Kim Bruning 14:18, 13 May 2007 (UTC)

Interesting proposal, thank you. In my short experience here, I have seen controversial pages go through a number of edits, with two or more editors making contributions, followed by a revert to a version which was days old. These leap-backward reverts often result in the loss of spelling and grammar corrections which were not involved in the controversy. I'd have to think about how to word it, but I would suggest/support wording which discouraged using UNDO to go back beyond a single editor's changes in a particular section. Clearly this could be abused by tag-team editing, however I'm suggesting 'discouraging' the practice, not proscribing it.
I don't believe that the reverting editors intend to lose the non-contentious edits, but hasty reverts result in just that. If the material needs to be reverted, in most cases it can be manually re-inserted which would at least encourage the editor to look at what was changed between the version (s)he wants and the latest version of the text.
I am not suggesting a change that prohibits reverting, only that some wording could be added which discourages blind-UNDO to older versions, rather than reverting by working-forward. UNDO to the most recently old version, is fine. UNDO to something 4 versions old, can be counter-productive and does not seem to be in keeping with the wiki-spirit of editing.
One specific case comes to mind, where an editor returned to an article, and did 4 hasty reverts. This particular editor was willing to look at their edits, gratiously conceeded that they had been hasty, and then went back and corrected their edits. However, not all such conflicts are so easily resolved and on many occasions good work is lost (or must be re-added) by hasty reverting. Lsi john 15:25, 13 May 2007 (UTC)

Question regarding reversions

If an editor has been blocked for breaking the 3RR, and then reverts the same content again after the block is lifted, is that one subsequent reversion grounds for a later block or not, given the fact of the previous block for the same behavior? I think this is a rational question, and it is one that I cannot find directly answered in the content of the page. John Carter 19:38, 9 May 2007 (UTC)

When I was doing 3RR, my rule was to stick fairly precisely to the letter of the rules. If you made 4R in 24h, you got blocked, but maybe not for the full 24h. If you *were* blocked for 24h, then the first revert after that couldn't be 3RR, because it would be the first in that 24h time frame. But if you were blocked for 8h, then immeadiately came back and reverted, you could well be blocked if that made another group of 4 edits in 24h. In other words, the 3RR block did not "clear" the previous reverts. None of this, of course, affects the possibility of being blocked on the grounds of edit warring William M. Connolley 21:26, 9 May 2007 (UTC)
If the 4th reversion were within 24 hours of "eligibility", not couting the time when the editor was blocked? John Carter 21:33, 9 May 2007 (UTC)
Blocked time counts. Its always done on real, wallclock time William M. Connolley 21:58, 9 May 2007 (UTC)
You guys may find this take interesting, an argument that if a 3RR report is dismissed, it "strikes off" reverts that have happened and the editor can continue reverting and counting at 1 again. [10] --Minderbinder 00:24, 11 May 2007 (UTC)
Users can be blocked for edit warring whether or not they have 4 reverts, when such a block is appropriate. CMummert · talk 01:11, 11 May 2007 (UTC)

Policy enforcement

Is "policy enforcement" other than the specific ones listed on this page an exemption from 3RR enforcement? Example here, it's a situation where there wasn't copyright violation and there was a disagreement over whether content violated WP:NONFREE. Editor was initially blocked but later unblocked per "policy enforcement". Is it OK for editors to keep reverting if it's "policy enforcement"? And if so, does it apply to all editors or just admins? --Minderbinder 12:17, 11 May 2007 (UTC)

I would say, no, it's not okay. So that unblocking might have been a bit rouge. Mangojuicetalk 12:28, 11 May 2007 (UTC)


Can you tell me please how you know if someone reports you for 3RR, or for that matter any ANI or RfC. Fainites 09:55, 18 May 2007 (UTC)

Fainites rules of etiquite suggest that anyone who reports you should notify you. You can also monitor the admin boards, however they tend to scroll rapidly and you'd have to check them regularly to be sure. Lsi john 11:47, 18 May 2007 (UTC)
Personal opinion: Ideally, the person who reported you (or someone else) should have first taken reasonble efforts to notify you in advance that your actions appeared to constitute "edit warring" ... unless the circumstances were exceptional or there were other reasons to suspect that such notification would not have been necessary or productive. dr.ef.tymac 14:42, 18 May 2007 (UTC)
Established users (i.e. not newbies) should know about 3RR and should be mindful of not editwarring. ≈ jossi ≈ (talk) 19:28, 29 May 2007 (UTC)
The issue is not simply "newby vs established" ... the issue is there are dramatically different interpretations of what constitutes edit warring to begin with. This very discussion page presents a wide range of practical interpretations of what differentiates "edit warring" from "good faith attempts to find consensus, address article deficiencies and make constructive contributions". Simplistic distinctions between "newby and established" simply do not capture the full nuance and complexity of many cases that require review under this policy. dr.ef.tymac 19:40, 29 May 2007 (UTC)
What I mean, Dreftymac, that an established user should know what editwarring is and that reverting more than 3 times in 24hrs is not acceptable. I do not see any fine lines here or differences of interpretations, just simply: 'do not editwar. If you do, you may lose your editing privileges temporarily. I newbie may no know that, and that is the reason the admins monitoring the noticeboard will not block a newbie unless the user was informed of this policy. ≈ jossi ≈ (talk) 20:27, 29 May 2007 (UTC)
I do not see any fine lines here or differences of interpretations ... yes, User:Jossi, and respectfully, that is precisely my point. A lot of people just assume that "edit warring" and "reverting" are intuitively obvious terms with blatantly unambiguous application in all cases. In the easy cases, they may be clear. The problem is, not every case is an "easy case". Even if the "close call" cases are few, few does not equal "non-existent".
For example, is it "warring" to re-add content with a citation if someone prevously removed it, claiming it lacked a cite? Some might say: "yes it's a revert" but "no it's not warring, but instead an attempt at resolution", whereas others might see it differently and say it's "warring". Others may make a determination based on the track record of the individual contributor. Still others may consider other factors. The point is, the notion of "obviousness" (in the abstract) seems like a house of cards when it comes time to apply specific consequences to specific actions.
The bottom line, it would seem useful to give people advance notice of potentially adverse outcomes, and not refrain from doing so, simply because we may think something "should be obvious". That about sums it up for me. dr.ef.tymac 15:25, 30 May 2007 (UTC)

reverted edit by Moreschi

User:Moreschi, I reverted your last edit marked with the edit summary "... this is slightly silly ..."

The modification you made to the policy presented (IMO) a potential for unduly mischaracterizing the risks that individual contributors face in 3RR situations. Indeed, a good faith reading of this very discussion page indicates the matter is not silly at all. Substantial advanced notice should be given whenever and wherever possible, to provide contributors with a realistic depiction of the potential problems associated with this type of situation. dr.ef.tymac 19:31, 29 May 2007 (UTC)

Actually, as dr.ef.tymac just illustrated by example, you should 'discuss' on the talk page after you revert, and not wait until after your self-revert due to 4RR. If you are applying WP:BRD after someone was bold, then you would explain why you reverted, and not wait for them to ask. Lsi john 20:17, 29 May 2007 (UTC)

Help Requested

I'm glad I found this page. Baseball Bugs has been causing some major problems. I am relatively new here and he's called me a sock and been reverting edits I've made. The two problems are on the shoeless joe jackson page and the will ferrell page. Any advice would be greatly appreciated. El redactor

  • Until the case is resolved, it would be improper to assume bad faith and act on that assumption. Additionally, socks (or not) aren't generally relevant to the 3RR rule. If the information being added is from a reliable source and is significant and relevant to the article, and is verifiable, reverting it would not be recommended. If the information does not meet Wikipedia standards, and someone continues to insert it, then there are methods to handle that, other than repeatedly reverting them.
If you need an uninvolved 3rd party to get involved, you're welcome to post a note on my talkpage. I'll be glad to help out if I can. Lsi john 17:24, 12 June 2007 (UTC)

The complaint against Tecmobowl (as well as El redactor) on the Shoeless Joe Jackson page has primarily to do with his/their continued insistence on posting his/their personal baseball card sales website as an external link. Another admin has already remarked that it's a fansite and should be kept out. That subject is the key to the sockpuppet investigation. Regarding the Will Ferrell birthdate citation, it's not that important, it's just a symptom of Tecmobowl and/or El redactor's general insistence that his/their way of doing things is the only way to do things, which has aroused the ire of various editors. The original editor explained his reasoning for the citation, so I restored it. If El redactor still don't like the citation, he should discuss it with that editor directly. Baseball Bugs 18:04, 12 June 2007 (UTC)

As I said above, none of that really relates to 3RR. There are other venues of dispute resolution which should be pursued. Third-Opinion, Administrators Noticeboard Incidents, Conflict of interest, requests for mediation are some of them.
Though it may not be a reliable source, I would disagree that the single link to the Joe Jackson page is spam. There is no advertising on the page, and it does seem to contain useful information. Is there some reason you doubt the facts on the page? WP:IAR can always apply to things which are outside the guidelines or rules. Again, it would be better to continue this conversation on the article talkpage. Lsi john 18:34, 12 June 2007 (UTC)

This isn't the place for this discussion: this is for discussion about the 3RR rule. Put 3RR reports on WP:AN3 and discuss possible ones there. Not here William M. Connolley 18:40, 12 June 2007 (UTC)

El redactor has now been permanently blocked for being a sockpuppet of Tecmobowl during Tecmobowl's block. To whom it may concern: You may delete this section from this talk page as you see fit. Baseball Bugs 19:12, 12 June 2007 (UTC)

Recommended additional exception

Over the last 1.5 years, I've been conducting removal of fair use images from places outside of the main article namespace per Wikipedia:Non-free content criteria item #9. The policy has been repeatedly upheld and no exceptions to it have been permitted for userspace or templates, where these removals most often happen. Considerable debate has been conducted previously regarding such use on portals and to a lesser extent on wikiprojects, and even considerable debate regarding the main Wikipedia page. The outcome of all debates on this subject has been that item #9 has been upheld in every case to mean no fair use display on any pages other than actual articles.

Over the last 1.5 years, I've made roughly 5000 such removals. Recently, User:Eagle 101 has been running a bot called User:Gnome (Bot) that has been conducting thousands upon thousands of these removals in a considerably smaller time span. At times during my removals and in helping to support Gnome bot, some users have tried engaging in revert wars to prevent application of this policy. A case example is a minor revert war that erupted today regarding a fair use image on a wikiproject (see [11] history actions of 19/20 June 2007).

Since every objection to the application of this policy has been rejected and the policy upheld in every case, and since it is commonly the case that willful violations of this policy are considered vandalism, I'd like to add an exception to the 3RR policy that would say:

  • reverts to remove fair use tagged imagery from non-mainspace pages (see WP:NFCC item #9) provided that the editor removing such imagery contacts the editors reverting in opposition.

Comments? If no objections, I'll add this in a week's time. --Durin 13:46, 20 June 2007 (UTC)

  • I think this is a no-brainer, and support it completely. Rebecca 13:49, 20 June 2007 (UTC)
  • Yup. Let's close the door to gaming the system; seems to me this is a special case of copyright violation, so should be appended to the second bullet. Guy (Help!) 15:00, 20 June 2007 (UTC)

Request: This makes sense, but, respectfully, this seems redundant. Can you please explain why your proposal is not already adequately addressed by the following (already stated in the policy exceptions):

  • reverts to remove clear copyright violations or clearly libelous material;
  • reverts done by a user within his or her own user space, provided that such reverts do not restore copyright violations

The proposal makes sense, but it seems to be already supported by the current language of the policy. The language of this section can become confusing to inexperienced contributors, so additional care is called for here. At very least, the text of the proposed addition should be re-written for clarity. dr.ef.tymac 15:24, 20 June 2007 (UTC)

Counter-proposal: It also seems appropriate (if clarification consistent with this proposal turns out to be warranted) to simply add this proposal as a footnote to the already-included mention of copyright violation, as a case-in-point example. dr.ef.tymac 15:30, 20 June 2007 (UTC)

  • The fair use violations in this case are a matter of policy, not necessarily of copyright. Thus, the clarification is needed. People have and do dispute that such use is a copyright issue. This clarifies the matter, and makes it clear that reverting re-insertion of fair use images repeatedly is not a 3RR violation if the person conducting the removals has attempted to discuss the issue with and educate the editor in question. --Durin 15:34, 20 June 2007 (UTC)
  • reverts to remove clear copyright violations or clearly libelous material or other material contrary to policy;
  • reverts done by a user within his or her own user space, provided that such reverts do not restore any of the above
Agathoclea 15:41, 20 June 2007 (UTC)
Ok, Durin, I think you make a good point, there is sufficient ambiguity here to justify a clarification.
Please, however, work on a wording that you think clearly and concisely conveys this clarification. It takes considerable effort to keep the "Exceptions" section from getting too confusing and unreadable for inexperienced users. The fewer excuses people have for not carefully reading this policy in advance ... the better. I'll be happy to try and help if needed. dr.ef.tymac 15:47, 20 June 2007 (UTC)
  • I'm not averse to Agathoclea's suggested addition. I do think my original wording is concise and to the point. We're getting too many edit wars from this. --Durin 15:52, 20 June 2007 (UTC)
Counter-proposal: Here's an alternate that should be both useful for the "edit war" problem you've expressed, as well as still readable to the "general audience"
(please see User:Dreftymac/Scratchpad003 for the entire context). dr.ef.tymac 16:06, 20 June 2007 (UTC)
  • The problem with that wording is no suggestion of discussing the matter with the user. Many users are completely uneducated about this issue and revert on sight, thinking it's vandalism. Continued removals without discussion is not conducive to resolution. I know this is discussed elsewhere on 3RR, but I think it's especially important in this case, because it's a little understood policy. --Durin 16:11, 20 June 2007 (UTC)
Alternative: * # reverts to remove non-free content that has been appropriately discussed and identified as such by Wikipedia contributors; (please note also that the text will have direct links to the relevant policies, as well as a footnote referring to WP:NFCC). The links are included in the scratchpad version of this counter-proposal. It's essentially the same thing you proposed with less in-line replication of content from other policies. dr.ef.tymac 16:15, 20 June 2007 (UTC)
Update: Scratchpad proposal removed. This matter is hopefully resolved by bainer's simple modification to the policy wording [12] here. HTH. dr.ef.tymac 16:40, 20 June 2007 (UTC)
  • I think the additional exception is reasonable, and have no particular opinion on the wording thereof. >Radiant< 16:08, 20 June 2007 (UTC)
  • I clarified the bit about copyvios to mention violations of the non-free content criteria too, it was always intended to cover that anyway. As to the issue of telling people first, well that's just good practice and should be done anytime someone seeks to rely on one of these exceptions to justify their reverting. --bainer (talk) 16:30, 20 June 2007 (UTC)

I'm all for it, provided the wording is made very clear as to what this applies to. ^demon[omg plz] 18:38, 20 June 2007 (UTC)

When reverting to different past versions, which edit is 3RR?

Since reversion is "undoing, in whole or in part, the actions of another editor or of other editors," when two editors are going back to different past versions of a page (or part of a page), which one must stop first to observe WP:3RR? For example see this: [13] Dhaluza 13:06, 24 June 2007 (UTC)

I'm no expert, but in such a case, what existed in the past shouldn't matter as far as 3RR. Each party is reverting the other. Now in terms of a discussion on Talk, and ultimate inclusion of the point, then if it was removed without good reason, one would be hard-pressed to defend keeping it out. Cheers, TewfikTalk 20:22, 25 June 2007 (UTC)

exceptions section

If you feel the need to add to the Wikipedia:Three-revert_rule#Exceptions section, please discuss first. Otherwise this is destined to require its own separate page. Thank you. dr.ef.tymac 17:06, 24 June 2007 (UTC)

Unrelated edits

Forgive me if this is already dealt with, but I've seen situations where editors are reluctant to revert edits that are clearly nonconsensus from outside IPs etc. for fear of traversing 3RR when there is a separate content dispute going on. The result is that an anti US editor, who would still normally revert an edit like "the US is a terrible country lacking in freedom", would let such an edit stand so as not to cross the line, even though it is clear that reverting an edit whose sentiments she might agree with is something to be commended. Are such issues left entirely to the judgement of the board administrators, or is a clarification of policy that would prevent such a situation feasible/desirable? Let me know, TewfikTalk 20:16, 25 June 2007 (UTC)

Honestly? "The US is a terrible country lacking in freedom" is blatant vandalism, which is already an exception. Something more subtle and potentially usable would be a different story. If you're in such a situation, the best thing to do is, if you feel your revert should be an exception (and you know the difference between vandalism and edits you disagree with), note in your edit summary that you are reverting vandalism. Careful, though, while this would be just fine if the edit really is vandalism, it could be viewed even more unfavorably if the edit isn't vandalism. If you're not sure if something will be viewed by an arbitrary admin as vandalism or not, assume it's not vandalism: vandalism cases are really clear. Mangojuicetalk 20:31, 25 June 2007 (UTC)

Well, I suppose my example could use some work, but the point that I was getting at was that the edit to be reverted would be something that the "other side" of the content dispute would surely want to be removed on content grounds, so that while you might be expected to "agree" with the view, your removing it would only be a demonstration of neutrality if anything. Now that there is a separate content dispute approaching 3RR, you might allow the edit to there any way we could recognise such a case, or is it best left unwritten and subject to whoever is dealing with AN3 at the moment? TewfikTalk 04:35, 27 June 2007 (UTC)

The 3RR is a speed limit. It purposely doesn't apply to only one dispute at a time, because reverting on two or three or many different disputes at once is just as harmful as reverting a lot on a single dispute. Mangojuicetalk 20:26, 10 July 2007 (UTC)


I've got a question about if there is some type of exception to the 3 reverts in 24 hour rule if the reverting is done over the course of days. Because the article Joan Chittister appears to be in a really slow, drawn out edit war (theres barely 2 edits a day) and would like to know if you can handle this like a normal 3RR violation (or if there is some special way of handling this). --​​​​Dtbohrer​​​talkcontribs 03:36, 7 July 2007 (UTC)

While 3RR applies to general edit warring as well as rapid reverts, it's best to direct the editors involved to discuss their differences on the article talk page. If the reverting continues, you might want to consider page protection. --Ronz 03:46, 7 July 2007 (UTC)
Two more things, does it matter that the editors engaged in reverting each others edits are same two. And if the page is to be protected should the article be reverted to the way it was before the war began. Thanks. --​​​​Dtbohrer​​​talkcontribs 03:56, 7 July 2007 (UTC)
When a page is protected, it is always meta:the wrong version... meaning, the protecting admin will just protect the version that's there, unless it seriously violates policy (e.g. copyvio or something). If there's an edit war, protection is justified: doesn't matter if it's two users or ten. However, when it's just two users, there might be some hope of settling the issue via WP:DR. Mangojuicetalk 20:26, 10 July 2007 (UTC)


If one editor violates 3RR, other editors should not be bound by it in reverting him/her. Currently, anyone may prevail in any content dispute without consequence by blowing far past 3RR with an edit-warring anon: the time it takes to report 3RR and for the report to be responded to is too long relative to the time it takes to violate it (i.e. hardly any.)Proabivouac 10:27, 10 July 2007 (UTC)

Strongly disagree. The 3RR is a speed limit and shouldn't apply unevenly. I agree, it would be nice if 3RR violations were resolved faster. Remember not to be too overly concerned with the current state of the article. If the article is vandalized blatantly, go ahead and revert. If it's more subtle, but still clearly policy-violating, consider complaining about the behavior at WP:AIV instead of reverting: AIV is much faster than AN3. If it's just something you disagree with, let it stand until the 3RR report goes through. Mangojuicetalk 20:26, 10 July 2007 (UTC)
"Remember not to be too overly concerned with the current state of the article."
That's good advice, thank you. Similarly, your suggestion to take such matters to WP:AIV.Proabivouac 22:22, 10 July 2007 (UTC)

Equal treatment?

What's this about having to treat 3RR violators 'equally'? I don't think that's a good point for enforcement. Does it means block lengths must be equal? That would be very poor, in some cases. I believe that admins should have full discretion here. Charles Matthews 15:42, 16 July 2007 (UTC)

I think it means that an established user's first 3RR offense should result in the same length of a block as a new user's first offense and similarly for their further offenses. Sancho 15:44, 16 July 2007 (UTC)
I've generally interpreted this as meaning make it fair based on each user's history (you might block the one with a long 3RR-related block log, while only warning the one who also violated but has never been blocked nor warned). It may be better to word it in a way that makes this clearer. Heimstern Läufer (talk) 09:25, 17 July 2007 (UTC)
I've thought of it as an admonition to not excuse one party from their violation based on the merits of their change. I don't think it means that block durations have to be even. The text could maybe stand some editing there. Mangojuicetalk 17:45, 17 July 2007 (UTC)

Persistent use of reverts (edit warring)

I've noticed confusion surrounding the 3RR rule on many occasions and have seen many arguments about interpretation, particularly among relatively new WP users but certainly not limited to relatively new users. I'd like to recommend thinking about an additional behavioral standard that might apply specifically to edit warring, which might compliment the 3RR rule at some conceivable stage in the future. At present 3RR is the rule, and "No edit warring" is a byproduct, almost an afterthought that requires administrative discretion and interpretation of the circumstances. To the extent that this is the case, why not begin talking about officially calling it what it is rather than tacking it onto 3RR? ... Kenosis 01:19, 19 July 2007 (UTC)

Question re: reverts

I've been removing an external link at Edward James Olmos that I feel does not qualify under the EL policy. It is being repeatedly restored by the same editor, despite explanations. Having approached the 3RR limit today, I do not intend to go beyond that; I have advised the other editor of this, as well as cautioned them in an attempt to avoid a similar event. (This is not something that requires administrative intervention at this point, and my goal is to ensure that it stays that way.) That aside, does anyone here have suggestions as to an area on Wikipedia where editors can pose questions re: the suitability of content? That would be helpful - instead of reverting, I could just direct a question to that area, and then have a proper answer. (I don't care if I turn out to be right or wrong, I'd just like to know for sure either way.) Thoughts? --Ckatzchatspy 20:36, 24 July 2007 (UTC)

There are at least three approaches you can take. First, you can post a note at Talk:Edward James Olmos; that's standard when editors disagree on an article. Second, you can post a note on the talk page of the relevant policy or guideline, particularly appropriate if the issue is more related to a technical point than to article content. In this case, that would be Wikipedia talk:External links, I think. Third, you can make a general request for additional editors to get involved, assuming that only two editors are involved in this disagreement, by posting at Wikipedia:Third opinion. -- John Broughton (♫♫) 13:28, 29 July 2007 (UTC)

How ?

How to inform (or warn ) someone that he/she has reached the WP:3RR ? Ammar (Talk - Don't Talk) 07:27, 29 July 2007 (UTC)

The standard warning can be found at Template:uw-3rr; this is to be posted on the user talk page of the person who has reached (or even gone over, but without being previously warned) the 3RR limit. You don't really have to use the standard warning, but if you do use your own words, it's pretty much required that you provide a wikilink to Wikipedia:Three-revert rule, because new editors aren't going to be able to find the policy without such a link.
Also, once the warning is posted, don't worry if the user deletes it from his/her talk page; deletion is taken as acknowledgment that the posting has been read, and of course the posting is still part of the history of the user talk page. (A user has an absolute right to delete postings from other editors from his/her user talk page, although this is a much misunderstood point.)
And, finally, a user only needs to be warned once in his/her editing career - so if a valid 3RR warning (i.e., one linking to the policy) is already visible on the user talk page, don't post another one (some people would regard that as harassment). Similarly, if the user has been blocked once (or more) for a 3RR violation, it's absolutely unnecessary to post a (new) warning; he/she can be presumed to be quite familiar with the policy.
P.S. You need to fix your signature - you shouldn't be using larger-than-normal font (which is what shows for me, at least: Windows XP, Firefox browser). -- John Broughton (♫♫) 13:20, 29 July 2007 (UTC)
Thanx boss . In fact , im planning to change my signature too. thanx Ammar (Talk - Don't Talk) 21:36, 31 July 2007 (UTC)

User space?

I changed "User space" to "User page and subpages" in one of the exceptions. My understanding is that people are generally not blocked for 3RR violations on their User page or subpages because those are kind of like "their pages." But this does not apply to the user talk page. I would not have taken "User space" to refer to the user talk page, but according to the link, it does. (I would have taken "User space" to refer to pages in the User: namespace). Anyway, so I changed it, to clarify. Mangojuicetalk 18:02, 31 July 2007 (UTC)

Something I've noticed about the way some people handle the 3-Revert Rule

First, although people will deny it, there seems to be a little loophole in this rule. When someone makes an edit (edit 1 by User One) and it gets reverted by someone (revert 1 of User Two), User One has the upper hand in getting their way if they choose to ignore Wikipedia policy of going to the talk page to discuss compromise. Should User One be stubborn and revert the article back (revert 1 of User One) and this is not appreciated by User two and they revert again (revert 2 of User Two), all User One has to do is revert 1 more time (revert 2 of User One) and if User Two reverts again (revert 3 of User Two) he has reached his limit, and User One can now revert the article again (revert 3 of User One) and get the article to reflect his edit. Both are now at there limit. However at this point since User One has broken Wiki policy about editing in good faith, if User Two is stubborn as well and decides to revert again (revert 4 of User Two) they then have broken the 3RR and User O then reports User Two. I call this WRANGLING (it should be a wiki term as far as I am concerned) If User Two is a conscious editor and does not revert the 4th time User One gets their way by default. It happens all the time and is apparently thought to be sanctioned by Wikipedia in general as there is no mention on this behavior on the 3RR page. This may or may not be a conscious behavior but has the same result. I know Admin is smart enough to see this when it does happen (they are not idiots) but it does create a lot of conflict and is the basic start to nearly all edit wars. Shouldn't this be added to the 3RR page?

Also shouldn't that first edit be official considered as a revert after the edit of the next user is then reversed against wiki policy on consensus by the first user. That would be a great way to stop this behavior. It allows User Two to warn User One that they are going against not only consensus but the 3RR themselves. It keeps a person from picking an edit fight as they will not be able to gain the upper hand in that manner. --Amadscientist 12:07, 1 August 2007 (UTC)

The technicalities of the rule don't matter that much. What you are describing is a revert war. I might only warn the users the first time around that they were edit warring and need to stop, but if they were to continue without technically breaking the 3RR, I would block them anyway. This kind of situation is somewhat routine. So your change wouldn't have much impact, but it does sound like it could encourage ownership issues. Mangojuicetalk 13:42, 1 August 2007 (UTC)
What you just wrote is technically correct, and proves that Admin can see through a problem such as this-however it does not stop this from occurring or even answer to why it is not mentioned in the 3RR page. Many of the problems that Wikipedia is currently facing are interpretation issues of members pushing what they feel is the correct way that a policy should be interpreted, even when it flat goes against the written policy. More details should be added to this policy page to make it far more clear.
You can never stop a true vandal until they are blocked or banned, but explaining further will stop misinterpretation or a misguided user--Amadscientist 23:04, 1 August 2007 (UTC)
Also you do not talk about what I brought up at all about intention of those revert wars. Many revert wars go unseen on pages where nothing is reported until someone gets past 3 reverts. You may be able to see through it when it does get reported and may even not block the Second User who has technically broken the 3RR, but that doesnt deter the war from beginning in the hope that that second user will stop at 3 reverts. The premise that someone else will make the change if it is important is not always true. This then encourages the editor to do it as a regular practice.--Amadscientist 23:20, 1 August 2007 (UTC)

3RR can harm productive editing

There is a problem in the operation of 3RR. After nearly a year of blemish-free record (at least in this regard), I've suddenly been tripped up by it. Let me call myself a "productive editor". I was doing a lot of general work in one large article Battle of Jenin, including new writing. The article has/had multiple problems, bad references, factual errors and a muddled layout.

I was suffering reverts, but refusing to edit-war. At the same time as moving things around and putting in references, I was writing in a whole new section to the article - this work was also suffering reverts. One of the latter was for alleged copyvio, one as "irrelevant", one appeared to be a mistake. And so I got caught out. I'd re-written my section almost completely (I think half the title is all that remains), but my offence is (or can be) still classified as a 3RR. The net result is horrible, and cannot be a good thing for the encyclopedia. The current system:

1) Lends itself to collusion against a productive, working editor in the middle of doing a lot of work on a complex article.

2) Is damaging to cooperation between editors. Even a true first offender such as myself wasn't entitled to a warning - or even to be told I faced a block!

3) Encourages deceitful summaries, disguising the reasons for the edits being made. (eg copyvio, references problems, duplicate material etc).

4) Is misleading - I didn't revert anyone, it was me that was reverted/blanked (3 times, leading me to re-write 3 times).

I once spotted another editor who I know to be productive get caught out in a similar fashion (except I couldn't follow the details of what had happened). It cannot be good for articles for it to happen.

I don't know what the answer is, other than to warn admins here of what can happen. Someone can be adding information to an article, and extensively re-writing it each time ..... a 3RR report in such cases is worthless. The blocking admin in my case has apparently volunteered that he approves my unblocking ........ but I think more people need to know. PalestineRemembered 19:42, 8 August 2007 (UTC)

Yes, it is not uncommon for otherwise productive contributors to get caught up as you've described. The best strategy is probably to try to involve more people and request feedback when you notice this type of circumstance arising. You may have to pull back a bit and restrict your contributions to the discussion page for a while. dr.ef.tymac 20:40, 8 August 2007 (UTC)

Chronic Fatigue

user orangemarlin has reverted twice my edit within a matter of hours here. without addressing my comments on discussion talk page, out of the blue within minutes another editor who has never contributed to the page reverts the third time and threatens me with 3rr here] with no attempt to address discussion page Anything I can do Jagra 01:33, 10 August 2007 (UTC)

See WP:AN/3RR. I don't think Filll (talk · contribs) is a sockpuppet. JFW | T@lk 06:52, 10 August 2007 (UTC)
Thanks for comments jfd this is a new area for me probably the wrong page! from what I can find, agree Fill not a sockpuppet, but seems to be an association of 'usual suspects' on Homeopathy talk page, so surrogate is probably better explanation. Still do not think it in spirit of Wiki to turm up at an article, revert an edit without discussion then declare an edit war and warn only one party, not a neutral or advisory role as it pretends. Will try his talk page for response Jagra 01:22, 11 August 2007 (UTC)

Replacing citation needed tag with a source

Does replacing {{fact}} with a source count as a revert? If an editor has specifically requested replacing the tag with a source, which is what the tag requests, then can fulfilling that request count as undoing the "action" that made the request? I believe all would agree that the request implicitly includes removing the tag. Doesn't the "action" of the editor who placed the tag live on by the fact that the source was added when it otherwise might not have been? ←BenB4 08:21, 12 August 2007 (UTC)

The Gdansk Exception

If anybody wants to bother reading Talk:Gdansk/Vote, the outcomes of that discussion are quite clear. In fact, it's one of the few instances in the history of the project that the community took matters into its own hands and settled a content dispute. As far as I'm aware those outcomes were never repudiated, and I know of at least one case where that exception was specifically (and successfully) invoked. This is an unusual exception, but it's quite wrong to dismiss it as "undiscussed craziness." Best, Mackensen (talk) 15:29, 15 August 2007 (UTC)

  • Frankly, that it's "discussed craziness" is probably far more disturbing. But if nobody else is interesting in keeping the 3RR viable for long, I'll be happy to drop it. Frankly, edit warring is already a blockable offence, and if anything this was just creating license to revert three times a day more than it was stopping anything. WilyD 15:35, 17 August 2007 (UTC)

3RR hurts Wikipedia

User:Alexia_Death/Accusations_of_collaboration:_3RR_hurts_Wikipedia please read and discuss this essay... I would love your feedback. --Alexia Death the Grey 13:52, 18 August 2007 (UTC)

I like it in principle, but in practice, counting reverts "per edit not per editor" would be really very hard. ←BenB4 09:15, 20 August 2007 (UTC)
I would love it if you could illustrate your concern with an example in the essays talk...--Alexia Death the Grey 09:45, 20 August 2007 (UTC)

You have to understand that 3RR is really just an especially lenient one-revert rule. Reverting back and forth accomplishes nothing, whether once or twice or five times. — Omegatron 17:02, 22 September 2007 (UTC)

Speed limit

I thought it might be good to make an analogy to a speed limit. They aren't always enforced, but they always can be. Doesn't matter why the admin chooses to enforce it in a specific instance, just that you broke the rule and they enforced it. And the fact that someone else was also revert warring at the time doesn't let you off the hook. — Omegatron 16:59, 22 September 2007 (UTC)

What does "consecutive reverts by one editor are often treated as one" mean?

If this is a rule that is going to be enforced on a very frequent basis, as it apparently is, then shouldn't it be as specific as possible to avoid ill-will against the enforcers? This part is abundantly and absurdly vague:

"consecutive reverts by one editor are often treated as one revert" [emphasis added]

Exactly how are people supposed to know how many uninterrupted changes they are allowed to make?

I have read that administrators are often accused of unfairness or misconduct, and that this is often upsetting to them. I respectfully submit that for allowing such a ridiculously vague rule to stand, and endorsing it by participating in its enforcement, such accusations and discomfort are not entirely undeserved.

Is there any objection to removing the word "often" from that phrase? Acct4 04:52, 29 September 2007 (UTC)

Why, have you seen someone blocked as if multiple consecutive reverts were not treated as one revert? I find it hard to believe that would happen. Mangojuicetalk 05:11, 29 September 2007 (UTC)
Frankly, I have no idea, but rules that are not unreasonably applied can still be unreasonably vague. If it has never happened then everyone will support removing "often". Acct4 06:23, 29 September 2007 (UTC)
I support it; I just wanted to see if there was some actual example where someone found an exception. The way I figure it, we don't need to hedge that rule because (1) I can't think why we'd need to make an exception to it, and (2) we always have WP:IAR. Mangojuicetalk 15:09, 29 September 2007 (UTC)

Well, it's been a couple of days, I'm going to boldly delete that "often". 04:09, 1 October 2007 (UTC)

I put it back; there are instances where back to back reverts (for instance, to different pieces of content in the same article) clearly do count as separate reverts. --- tqbf 18:23, 5 January 2008 (UTC)
I can not bring myself to believe this. The only way back-to-back reverts are not counted as distinct is if they are to the same passage? Consecutive reverts are almost always to different places. Can you give a specific example where any blocking admin has considered consecutive reverts to count as separate? MB83 (talk) 19:55, 9 January 2008 (UTC)
I can't, but I don't see why this is hard to imagine; all you have to envision is two simultaneous edit wars with a shared participant on one page. Look at the edit history for Ron Paul presidential campaign, 2008. The word "often" is innocuous and was the status quo; the change may be incorrect, and that's why I reverted. --- tqbf 22:20, 9 January 2008 (UTC)
We seem to be in a WP:BRD. I've reverted. I'm happy to help with alternative wording (though I don't feel qualified to). But the wording you have on the page excuses editors who are in simultaneous edit wars on the same page. WP has pages that are both controversial and large enough for that to happen. An editor that reverts two different editor's different content on the same page is making two reverts, not one, and we should not have a policy they can wikilawyer their way around that with. --- tqbf 00:25, 10 January 2008 (UTC)
It has never been the case that "an editor that reverts two different editor's different content on the same page is making two reverts" -- why do you say that? It doesn't matter how many edits there have been to different places disputed by any number of editors. If it is made in a single edit, it is by definition a single revert. And as far as I know, no blocking admin has ever considered consecutive reverts to count as more than one. If you want to insert the word "often," that makes the rule vague. But you are unable to clarify it because there has never been enforcement that has considered consecutive reverts to be more than one. You are doing readers of this policy a grave disservice. MB83 (talk) 03:15, 10 January 2008 (UTC)
"Grave disservice"? You must be right; I don't feel anywhere near this strongly about it. I'll revert it back in. The policy already says you can be hit with 3RR without making 3 reverts, which covers the edit warrior who is batching multiple different reverts up to slide in through this "loophole". --- tqbf 05:56, 10 January 2008 (UTC)

Meh., if unsure, just apply the plain vanilla edit warring guidance instead? --Kim Bruning (talk) 02:04, 10 January 2008 (UTC)

This is a rule for which several people in heated disputes are blocked daily, and it depends on counting. If people don't know how the counting is done, there is no way they can be sure they are in compliance with the rule. If it is going to be enforced literally, then it needs to not be vague. MB83 (talk) 03:18, 10 January 2008 (UTC)
If you are counting, you are not in compliance with the spirit or the letter of the wikipedia rules. You are not supposed to actually game the 3RR. ;-) --Kim Bruning (talk) 20:51, 10 January 2008 (UTC)
So in the spirit of Wikipedia rules, surely this should be the NRR, where N is left entirely to the discretion of the blocking administrator, should it not? What could possibly go wrong? MB83 (talk) 22:12, 10 January 2008 (UTC)
Mu :-) --Kim Bruning (talk) 23:01, 10 January 2008 (UTC)

Danger Will Robinson

3RR is in fact the 3 rules of robotics. We must therefore have this be speedied since it is a copyvio. Oh no, I'm about to forget my sig! SineBot Help! —Preceding unsigned comment added by (talk) 07:16, 8 October 2007 (UTC)

<3 --Kim Bruning (talk) 02:04, 10 January 2008 (UTC)

BLP discussion regarding conflicting wording

Please note a new discussion on 3RR as it applies to BLP, over at Wikipedia talk:Biographies of living persons#3RR exclusion. Depending on the outcome of that discussion we may change the description of the BLP 3RR on this project page. Wikidemo 15:19, 12 October 2007 (UTC)

Currently, this policy reads, regarding the BLP exception:

Reverts to remove clearly libelous material, or unsourced or poorly sourced controversial material about living persons (emphasis added)

while the Wikipedia:Biographies of living persons policy reads:

Editors should remove any contentious material about living persons that is unsourced (or) relies upon sources that do not meet standards specified in Wikipedia:Verifiability ... The three-revert rule does not apply to such removals if the information is derogatory. (emphasis added)

So the 3RR policy, as currently written, appears to be in conflict because it covers all unsourced/poorly sourced information (negative or positive), whereas WP:BLP limits the 3RR exception to negative/derogatory information.
Opinions of other editors on how to make the two policies consistent would be appreciated. -- John Broughton (♫♫) 14:09, 15 October 2007 (UTC)

Proposed addition

I wrote a crazy little proposal about a 5 revert rule, see Wikipedia:Don't violate consensus, concerning the case when a single editor is reverted by five different editors. I think it has some chance in reducing WikiStress, but on the other hand it is doubtful that adding more rules will help. I'm not going to defend that proposal, but I would appreciate, if some policy gurus here gave the idea some thought. I just think it might actually help in some extreme cases. --Merzul 18:37, 16 October 2007 (UTC)

Paragraph that makes no sense

Note that in the case of vandalism, blocking editors who have engaged in vandalism or protecting the page in question will often be better than reverting. Similarly, blocking or page protection will often be preferable in the case of repeated addition of copyrighted material.

Huh? If a page has been vandalized, how is blocking a user or protecting the page "better than reverting"? That may stop additional vandalism, but it won't get rid of the vandalism that's already there... you've still got to revert it – Gurch 09:05, 19 October 2007 (UTC)

What it means is that blocking or protecting is a better way to stop the problem than to merely revert. If I insert "merely" will that take care of it? Mangojuicetalk 11:59, 19 October 2007 (UTC)

User talk pages

User:Taeedxy, The list of exceptions should explicitly list user talk pages in order to clarify that users should, within reason, be able to control what gets displayed on their own page. I see no difference between someone's user page and user talk page in this regard, and it became an issue of contention in a recent case at WP:3RR. Can you discuss why you took this out? Ronnotel 16:48, 23 October 2007 (UTC)

It should remain out because editors will use it to Wikilawyer their right to control their user and talk page per this policy. On occasion clueful admins or experienced editors will ask an editor to remove edits to their talk and user page for the benefit of the project. If an editor refuses then on occasion it may need to be done despite their objection. FloNight♥♥♥ 19:00, 23 October 2007 (UTC)
Well, how about adding a link to Wikipedia:Don't readd removed comments then? I've had WP:3RR cases where one user reported another because they kept removing hostile comments from their own talk page. Ronnotel 19:31, 23 October 2007 (UTC)
This rule, as currently written, does not apply to "reverts done by a user within his or her own user page, user subpages" per this exception. "User talk page" does not qualify for this exception, or this exception would read "reverts done by a user within his or her own user space". Am I misinterpreting? Thanks!   — Jeff G. (talk|contribs) 20:08, 18 November 2007 (UTC)
That exception was very recently added, and a discussion was started in the section 'user space' below about it. There doesn't seem to be consensus there in favor of it, after plenty of time for people to discuss it, so I removed the exception just now. — Carl (CBM · talk) 20:38, 18 November 2007 (UTC)
I was somewhat confused about this. FloNight very recently proposed removing that section [14], and the conversation in the section 'user space' below shows that several others agreed. — Carl (CBM · talk) 22:29, 18 November 2007 (UTC)
Thank you! Please see Wikipedia:Administrators' noticeboard/3RR#User:, a discussion in which this change is relevant. Thanks again!   — Jeff G. (talk|contribs) 21:37, 18 November 2007 (UTC)


I'd like a clarification of the following situation. A user violates 3rr, and admits knowing about the policy. But it is claimed that he/she violated 3rr by "mistake". Should the user be blocked?

What if the user goes back to wholescale reverting after the 24-hour period in which he/she made reverts expires?

Such a situation happened here. While I'm not disputing the decision made by the admin (as a non-admin it is not plac to do that) I want further clarification for such a rule. When is it applied, when not? How do we decide if a user has made a mistake?Bless sins 23:48, 23 October 2007 (UTC)

It's fine to dispute decisions made by admins; it's more productive to do it politely, like this, than to rant on their talk page. Then you can point the admin to the discussion, and everyone can discuss what's going on.
Here's my general interpretation of how 3RR is implemented. If the reverts were in the past, say a couple days earlier, then we generally let the issue drop, with just a note. If the user doesn't have a history of 3RR, and just got caught up in the moment, some admins are very generous about letting them off with just a note.
If the reverts are very recent, the user realizes they made a mistake, and they revert their own last edit before being blocked, that is usually accepted instead of a block. Some admins will alert users to this option and give them a short period of time to revert themself before being blocked. But this is a courtesy, not a requirement. If a user has already made this "mistake" before and does it again and again, that's a different matter. — Carl (CBM · talk) 20:41, 24 October 2007 (UTC)

User space

Mardetanha added a section without any conversation that might be worth discussing first because it's a fairly substantial change to a base policy. I encourage some eyes on the proposed change he added (I've linked to the diff above) so any issues have a chance for redress before it turns into an official lightning bolt. - CHAIRBOY () 19:25, 24 October 2007 (UTC)

Followup, I guess this has been an ongoing issue, I see a history of adds & deleted, but little recent conversation. Before I revert myself, I'd still love to request some discussion about whatever open issues are causing folks to volley this back and forth. I'm leaving some messages to recent editors of that section to try and wrangle 'em over here for a talk. - CHAIRBOY () 19:29, 24 October 2007 (UTC)
I support Mardetanha's edit as I believe a user should be aware that they have some control over their own page, within reason of course. As per my comment above in the section User talk pages, how just providing a link to Wikipedia:Don't readd removed comments? I think that would satisfy FloNight's concerns as well as mine. Ronnotel 19:30, 24 October 2007 (UTC)
As a general rule, editors can remove comments and warning from their talk page beyond the 3RR limit without being blocked. By removing them we know that they have been read and inquiring minds can look in the history to see them. Someone edit warring with an user to make them stay is often upset with them and might report them for a 3RR violation. The best course of action would be for an experienced user to step in and sort it out by getting every one to calm down. Important to remember that admin are never required to enforce 3RR violations with a block. (Personally, I rarely ever block when users 3rr, instead I remind them of the rule, I ask them to revert, and start a discussion. They usually agree. This is usually much more productive in the long run.)
That said I do not think changing the wording here is best. Usually users have control over their user and talk page. But users need to understand that when asked by a clueful administrator or knowledgeable editor to change their talk or user page to conform to our standards in a particular situation then they need to do so. Telling them that they can not get a block for edit warring over their user page is ripe for wikilawyering and will make them feel that they have rights to their user page that they do not really have. That is why I object to the change in wording. FloNight♥♥♥ 20:23, 24 October 2007 (UTC)
But what about my proposed compromise? Are you opposed to that as well - i.e. adding a link that re-adding comment to a user's page should be avoided. Ronnotel 20:32, 24 October 2007 (UTC)
I generally agree with FloNight that it's better not to encourage users to edit war on their user pages by giving an exception to 3RR for that. I'm neutral about a comment regarding not reinserting warnings; I don't see that this is the right place for that - it should go in the guideline about warnings. — Carl (CBM · talk) 20:45, 24 October 2007 (UTC)
as i see and believe everyone' talk page in his or her own territory and we should respect it .even by violating 3rr .though i prefer by not doing this --mardetanha 22:18, 24 October 2007 (UTC)
It's more than possible to recognise that, generally speaking, users can control what is in their userspace, but we don't need to make an exception to 3RR to do so. Wikipedia:User page covers the ground on this issue adequately already. --bainer (talk) 00:33, 25 October 2007 (UTC)
but in fawiki we got some probelm one of editor's is going to be blocked by sysop for violating i think it MUST be in 3rr page --mardetanha 11:46, 25 October 2007 (UTC)
Perhaps what is needed is a change from a list of 'exceptions' to a list of possible exceptions. Violating 3RR to keep a 'vandalism' warning over a content dispute off your user talk page shouldn't result in a block... but violating 3RR to keep a nasty personal attack on another user displayed on your user page may. Rather than trying to identify and list every specific instance where 3RR does/does not apply it might be better to provide some examples of the types of things admins may take into account when deciding whether to block or not. Most of the listed exceptions still involve judgment calls... if someone exceeds 3RR removing 'vandalism' and an admin thinks it is a content dispute they may still be blocked, if supposed 'BLP violations' turn out to be non-controversial everyday info supported by the references 3RR may still be blocked, et cetera. As noted above, stating 'this is an absolute exception to 3RR' inevitably invites abuse and wiki-lawyering. --CBD 12:48, 25 October 2007 (UTC)
I would agree that there must be a way to express this concern while acknowledging that the right to control one's page is not absolute. Ronnotel 15:58, 25 October 2007 (UTC)

Reverting to undisputed versions

After reading this article I was under the impression that it was acceptable to revert to keep newly-added text that is being disputed on the talk page from an article, and also to defend the inclusion of tags notifying readers of the dispute. But when I tried to do this, I was criticised for engaging in an edit war.

This really ought to be made clearer on this page, as it's not unreasonable to expect that removal of new and controversial text that is still under discussion would leave you exempt from the 3RR. --Tom Edwards 19:51, 29 October 2007 (UTC)

Reverting back to an "undisputed version" does not exempt one from running afoul of 3RR unless one is reverting obvious vandalism or a BLP violation. The fact that people are willing to revert between one version and another indicates that neither version is undisputed and as such neither can claim to be the "undisputed version". If the version that you are trying to "enforce" is the version that contains the new and controversial text, just provide a link from the article's edit history of this version on the talk page so that other editors can see it. --Bobblehead (rants) 22:07, 29 October 2007 (UTC)
I can't imagine where you would have gotten the idea that "it was acceptable to revert to keep newly-added text that is being dispute on the talk page from an article" from this policy. It's not inherently acceptable or not acceptable - what matters is that you should not be repeatedly reverting. The practice you describe is not exactly the norm, but neither is it a bad idea.. but if you find yourself being the only one reverting out the new changes while others are putting them back in, you should stop. If several people object to the change existing in the article, it can be removed. But flip-flopping back and forth does no one any good. Mangojuicetalk 14:16, 30 October 2007 (UTC)
Probably he got the idea from WP:V#Burden_of_evidence or WP:Citing_sources#Unsourced_material. It may be natural for users to assume that it cannot violate policy to remove material that violates policy and that policy says to remove. For anyone interested, I've summarized the proper procedure to win an edit war at User:Jemmy Button/Legal. What the article should be more clear on is what to do instead of reverting. —Jemmytc 23:00, 14 December 2007 (UTC)
I just think that what you're saying there should be made explicit.
What happens if you're reverting purely to defend the addition of dispute tags (when a dispute is ongoing, of course)? Would that be simple vandalism on the other user's part? --Tom Edwards 14:23, 1 November 2007 (UTC)
Edit warring over dispute tags is particularly silly. Everyone should avoid it. The presence or absence of the tag doesn't affect whether there is actually a dispute. If someone removes a dispute tag and you feel there is still a dispute, the right course of action is to find wider attention for the issue. — Carl (CBM · talk) 14:33, 1 November 2007 (UTC)
I completely agree... I have seen people both remove the tags because they think the concern is a non-issue when they shouldn't, and I've also seen people place tags over non-issues or resolved issues when they shouldn't. Remove the tag if you think there's no dispute left; add the tag if you think there is, but for crying out loud don't edit war over it. It makes the dispute worse that way. Mangojuicetalk 16:34, 1 November 2007 (UTC)
Could this advice be added to the article, then? --Tom Edwards 10:07, 3 November 2007 (UTC)
I think it's a little hyperspecific. Better to avoid WP:CREEP. Mangojuicetalk 14:21, 5 November 2007 (UTC)


The three-revert rule is a good way to measure edit warring. But it alone cannot determine when a user is edit warring. This policy page acknowledges this to an extent when it points our the possibility of gaming the system. Nonetheless, I feel that this page does not adequately emphasize the idea that the the offense is edit warring, not breaking the arbitrary limitation of three reverts per 24 hours. The Arbitration Committee has repeatedly reaffirmed the idea that edit warring is considered harmful and has sanctioned users for this reason: edit warring, not for repeatedly reverting four times in 24 hours.

The emphasis on three reverts in 24 hours results in us missing the point sometimes. For example, a mistake I made often as a newer admin was to block a user who had four reverts in 24 hours, while not blocking his/her opponent who had only three. Both sides were edit warring, yet I blocked only one because she/he had crossed an arbitrary threshold. I don't think I'm the only one to have made this mistake.

Also a problem is that sysops come under attack when they do use their judgment. For example, when I've refused to block an editor who technically violated 3RR whilst other editors were also edit warring without technically violating 3RR, I've been accused of dereliction of my duty. Likewise, sysops who block for edit warring that is not technically a 3RR vio tend to come under attack ("admin abuse!").

Ideally, it would be good for us to abandon 3RR completely in favor of a policy that simply forbids edit warring, without some arbitrary limitation. At this time, though, I think our best move would be to keep this policy, but put greater emphasis on the idea that edit warring is the real prohibited behaviour, and that 3RR is an arbitrary (but often effective) way to keep it in check. Heimstern Läufer (talk) 07:05, 3 November 2007 (UTC)

Any specific wording changes you want to bring up? I think this is kinda what this policy is already doing, but maybe it could be doing it better. Mangojuicetalk 14:26, 5 November 2007 (UTC)
Edit warring is already blockable behaviour. The three-revert rule is a subset of the more general prohibition on edit warring; but because edit warring can be hard to clearly define, this was created to be a nice, unambiguous line that noone could dispute they had crossed. That's been undermined a little by the continuous desire to add exceptions to this, but that overall purpose remains. The 3RR is a subset of edit warring. --bainer (talk) 00:05, 6 November 2007 (UTC)
Yes, that's really the point I'm trying to get across here. That's why I've been working on improving the policy page on edit warring (OK, it was mostly Dmcdevit, not me). Improving that page will do a lot toward helping us move our focus in the right direction.
I recently made one change which was to add a section concerning the sysop's discretion in deciding whether to block for 3RR violations [15]. The idea behind this is to dispel the notion that users are entitled to have their opponent in a content dispute blocked if they report them. Any thoughts on this? Heimstern Läufer (talk) 07:51, 7 November 2007 (UTC)

Clarification on 3RR

FYI, on reading this page on the 3RR section of WP:Revert they seem slightly out of sync. Nothing huge but it makes things confusing.

Also, I am unclear on the exact definition of 3RR. Are the reverts counted strictly on an individual page basis regardless of what is being reverted? In other words, if a given page has 4 independent sections would one revert of content inside of each section be a violation of 3RR? I can understand the point of trying to prevent people from just reverting things endlessly back and forth so it is a good rule, I am just trying to understand where the boundary actually is.

"A revert, in this context, means undoing, in whole or in part, the actions of another editor or of other editors."

When I read this description it sounds like the intent is to prevent warring over specific content on a given page, not just the page itself. When it talks about "the actions of another editor or of other editors" it sounds like it means specific actions (i.e. edits) with on a specific piece of content. This is why I want clarification.

For example, if I am working on a piece of content (a given section for example) on a page and after 1 round of restoring my work in response to someone else's revert I say OK, lets put our comments on the talk page and then move on to something else can that something else be a different part of the page which is unrelated to the aforementioned dispute? And if for whatever reason some different user decides to start reverting my work in that new section would restoring my work in this case constitute a second revert under the rule?

I ask not because I intend to be a bull in a china shop, but just so that I clearly understand the rules. I understand that we should be discussing these differences on the talk pages but some users are very protective of what they consider to be their turf and refuse to allow any changes, so sometimes reverting is unavoidable.

In reading the comments above I think that my issue, while not exactly the same, is similar in spirit to the proposed addition above.

--GoRight 20:10, 12 November 2007 (UTC)

Yes, that would count as a second revert. The policy currently states " A revert means undoing the actions of another editor, whether involving the same or different material each time." — Carl (CBM · talk) 20:13, 12 November 2007 (UTC)
This wording is vague, then, since in my example there were two independent and unrelated editors involved. I was not warring with the same editor each time. "another editor" makes it sound as though you mean the same person to me.
So I take it to mean that any revert on a given page for any portion of that page and regardless of whose changes are being reverted (excepting my own) counts as a revert, with consecutive reverts counting as a single revert. Does that sound correct? --GoRight 20:59, 12 November 2007 (UTC)
Well vandalism and such don't count as reversions. I wouldn't enforce it unless it's been for reversions of the same edit - as in a revert war. the_undertow talk 21:14, 12 November 2007 (UTC)
OK, I forgot the vandalism and BLP violation exceptions. Understood. It seems that the safest course of action is to assume the above. As you say, some admins might take the circumstances into account more than others which is fine. Thanks for the response. --GoRight 21:25, 12 November 2007 (UTC)
The more relevant policy as far as editwarring is WP:EW. If folks revert instead of talking, one or both parties are liable to be blocked for disruption. WP:3RR is merely an upper limit on allowable disruption, if parties are editing and reverting sections in good faith with real improvement happening to the article as a result of the reversions, (ie a diff from when the editing started to when the editing ended actually shows some changes..), then it is more likely ok. —— Eagle101Need help? 02:59, 13 November 2007 (UTC)


I suggest we merge the entire policy into Wikipedia:Edit war It would serve to eliminate a whole lot of rules lawyering while keeping or current practice. Mercury 02:49, 13 November 2007 (UTC)

Could you elaborate on how it would eliminate wikilawyering? —bbatsell ¿? 03:05, 13 November 2007 (UTC)
For example... I only reverted three times, in 23 hours. Regards, Mercury 03:08, 13 November 2007 (UTC)
Keep the bright line, the fence, but lets not emphasize it. Mercury 03:09, 13 November 2007 (UTC)

I do not see the reason to merge this long-standing policy with another page. ≈ jossi ≈ (talk) 03:26, 13 November 2007 (UTC)

What do you think about my above reasoning? Mercury 03:27, 13 November 2007 (UTC)
I'd rather replace all these guidelines with "act like a jerk and you'll be held accountable," but 3RR has generally been useful to the community. WP:3RR already contains multiple provisos that 3RR isn't a right and so on. Raymond Arritt 03:32, 13 November 2007 (UTC)
This policy was created to address the issue of edit warring, that is right, but I do not see merit in merging a long-standing policy which is referred massively across the namespace as WP:3RR. Also noe that edit warring discusses the subject, while this policy is a remedy designed by the community to address these behaviors. ≈ jossi ≈ (talk) 03:33, 13 November 2007 (UTC)
Mercury, revert three times in 23 hours and I will be glad to enforce this policy. Read it. ≈ jossi ≈ (talk) 03:34, 13 November 2007 (UTC)~~
I know what it says.  :) While I understand that 3RR is not a right, I know... I was using an example. Additionally, both pages address the same subject. While 3RR is a long standing policy, the idea is now forked. A good deal can be gained by merging the two policies into one policy. Mercury 03:36, 13 November 2007 (UTC)
As I said, one describes the behaviors related to edit warring, and the other describes a specific remedy designed to address a specific one. You can edit war without breaching 3RR., and you can get dinged for it, regardless if you breach 3RR or not. ≈ jossi ≈ (talk)

I am in support of de-emphasizing 3RR in favor of the more fundamental issue of edit warring. It is a useful way to identify edit warring, but little more than that. Merging this page into Wikipedia:Edit warring would be one good step in that direction. If the community is not prepared to do this yet, it would be best for us to de-emphaize it in other ways, for example, referring to edit warring rather than 3RR when warning and blocking users. Heimstern Läufer (talk) 03:39, 13 November 2007 (UTC)

But there is less ambiguity with 3RR, which is helpful to new users. the_undertow talk 03:40, 13 November 2007 (UTC)
I don't believe the proposal is to eliminate 3RR. The proposal is to merge the policy documents, including the three revert rule, into one document. — Carl (CBM · talk) 03:43, 13 November 2007 (UTC)
Yes, Carl has my pulse. This is what I'm suggesting. I don't want to cut anything. I'd like to merge the two documents. Mercury 03:45, 13 November 2007 (UTC)
I was responding to the de-emphasis remark by Heimstern. However, I don't see a need to merge them as 3RR seems like a logical offshoot of edit war. the_undertow talk 03:49, 13 November 2007 (UTC)
By de-emphasis I mean primarily that we leave behind the long-standing notion that it's exceeding three reverts per day that is the problematic behavior, when really it is edit warring itself. The reason 3RR is useful is that it catches a great deal of the most egregious edit warring: If you revert more than three times a day, it shows that the edit war has clearly gotten to the point that admin intervention in needed. But the emphasis should rightly be on the greater problem of edit warring of all types. I'm not suggesting deprecating 3RR, but making it clearer that it is a subset of our edit warring policies. Ideally, we should have one edit warring policy that covers 3RR and other forms of edit warring. Heimstern Läufer (talk) 04:02, 13 November 2007 (UTC)

Sorry, folks I don't think this is a happening thing. WP:3RR was the first policy to be agreed in a large discussion by the community at large. To make that proposed merge you will need more that a discussion in talk.... ≈ jossi ≈ (talk) 03:55, 13 November 2007 (UTC)

Are you suggesting another venue? Mercury 03:57, 13 November 2007 (UTC)

I have posted a note at the WP:AN/3RR talk page, as this merge proposal would affect that noticeboard. If there is traction for this proposal, a note at the Village pump would be necessary. ≈ jossi ≈ (talk) 04:00, 13 November 2007 (UTC)

How exactly would this merge affect that? The proposal is not to eliminate the three revert rule, just to merge some policy pages. — Carl (CBM · talk) 04:02, 13 November 2007 (UTC)
It will affect it because 3RR is a specific policy related to a remedy, and WP:EW is another policy with a wider appeal. ≈ jossi ≈ (talk) 04:04, 13 November 2007 (UTC)
If the texts of WP:3RR and WP:EW were merged into WP:PQRS, it would still be possible to enforce the three revert rule. The name of the policy document does not determine whether its contents are policy. Nobody is suggesting that the three-revert rule would be eliminated. — Carl (CBM · talk) 04:08, 13 November 2007 (UTC)
(ed conf) ::: What I am saying is that there is a difference between WP:3RR and WP:EW. Merging these two will eliminate these distinctions, and that is why I object to the merge. ≈ jossi ≈ (talk) 04:14, 13 November 2007 (UTC)
What is the distinction, exactly? 3RR is an arbitrary measure that allows people to be blocked for edit warring without further discussion. — Carl (CBM · talk) 04:17, 13 November 2007 (UTC)

I see the above points, but would oppose a merge. 3RR provides clarity, particularly for newcomers -- and the fact that it's a separate policy lends it strength, clarity, and conspicuousness (if that's a word). Obviously it could be merged, but given its usefulness and centrality, I don't think it should be. --TheOtherBob 04:12, 13 November 2007 (UTC)

( See dict:conspicuousness) ≈ jossi ≈ (talk) 04:16, 13 November 2007 (UTC)
3RR is not the central issue, though - edit warring is the central issue. 3RR is an arbitrary measure that is used to simplify blocking people for edit warring, but the reason for the block is because edit warring is bad, not because 3 reverts are worse than 2. — Carl (CBM · talk) 04:17, 13 November 2007 (UTC)
Totally arbitrary, but relatively clear -- that's the point, I think. --TheOtherBob 04:34, 13 November 2007 (UTC)

Oppose I think that merging this policy with others would be serious mistake. It is perhaps the clearest behavior policy we have, that has real teeth; let's keep it that way. Crum375 04:17, 13 November 2007 (UTC)

The teeth we have put into this policy have come at the expense of teeth that could be in a policy against edit warring. Our heavy emphasis on "no more than three reverts in 24 hours" has led many admins to mechanically block for 3RR and only for 3RR, ignoring the greater question of who is edit warring and how this issue can be best resolved. A unified edit warring policy with proper teeth, emphasizing 3RR as one (but only one) way to measure edit warring, would be a vast improvement. Heimstern Läufer (talk) 04:28, 13 November 2007 (UTC)
Example of how emphasizing only 3RR is a problem: the classic situation with two warring factions where one reverts four times, the other three. Two edit warring users, yet only one is blocked. Why is this? Because we emphasize one element of edit warring, the three-revert rule, and leave behind the idea of preventing all edit warring. Heimstern Läufer (talk) 04:31, 13 November 2007 (UTC)
Yes -- it's the classic problem of rules versus discretion. Discretion is more accurate, rules are both over- and under-inclusive. But rules are predictable, easier to administer, and easier to explain. --TheOtherBob 04:33, 13 November 2007 (UTC)
Two edit warring users, yet only one is blocked. I think you are mistaken. Check WP:AN/3RR and see how the policy is applied. ≈ jossi ≈ (talk) 04:37, 13 November 2007 (UTC)
Certainly, it isn't always applied the way I've suggested (in fact, probably more often than not it isn't). Still, it is the result of the mechanical way of enforcing 3RR that is sometimes taken (especially by newer admins). What I'm most looking for is a way to emphasize that 3RR should be enforced as a subset of a greater policy against edit warring, emphasizing the need for admin discretion. Anyway, I'm getting a bit off the merge topic at this point; this is more a general issue of how we should deal with edit warriors in a more efficient manner than we do now. More thoughts later, possibly elsewhere. Heimstern Läufer (talk) 04:50, 13 November 2007 (UTC)
I certainly appreciate Crum's concern -- 3RR is a great success, a firm, no-nonsense policy with real teeth that has had some great successes. Merging the text of these two into Wikipedia talk:Edit war/Merged (one draft, feel free to tweak or make another), I definitely noticed that one was much softer than the other. But, as Heimstern says, 3RR's teeth come at the expense of a more comprehensive policy to deal with edit warring in general, especially in cases where users are experienced enough to game the mechanics. I'd like to think that we're keeping the teeth, just angling them in a slightly different direction. By all means, keep the electric fence, it's done a good job. Guess you could say I just want to add some landmines around the fence, so to speak. – Luna Santin (talk) 22:00, 14 November 2007 (UTC)
My, such violence, Luna! ;-) Heimstern Läufer (talk) 22:05, 14 November 2007 (UTC)
  • The 3RR represents a bastardisation of our core tenent that blocks are preventative and not punitive but was created for a very good reason - to provide a tangible line where edit warriors know enough is enough. It has been extremely successful frim this point of view. I would oppose any merge for fear that it would dilute the effect of the 3RR. Perhaps, if there is a problem with EW, we simply rewrite that to have more teeth in it. Spartaz Humbug! 06:32, 13 November 2007 (UTC)

Fear is the mind killer.... no. A merge will not dilute 3RR and will not change EW. I'm not sure what you mean here. Mercury 13:27, 13 November 2007 (UTC)

  • Oppose - I don't think there is a lot of overlap (but, unless I've missed something, the merge proposal lacks a draft to show what the merged policy would look like), which is about the only argument that I would find at all persuasive. As it is now, we have a policy (edit war) that makes for an easy wikilink, for discussing the need for compromise, in general, and a specific policy, 3RR, that makes for an easy link to warn an editor. If we combine them, the first link (about not edit warring in general) goes to a very long policy with much about 3RR detail (not helpful) and the second link (3RR warning) goes to a policy that takes about edit warring in general. I think keeping them separate is best. I fail to see how merging two policies is going to change editor behavior in any way (less wikilawyering?!?), and it clearly is going to damage quick references to WP:EW and WP:3RR, even if the latter shortcut drops one into the middle of a combined problem. In short, this seems to be a solution searching for a problem. -- John Broughton (♫♫) 13:48, 13 November 2007 (UTC)
  • I oppose this as well. If the point is not to scrap the 3RR in favor of something else, we will still have the 3RR as a policy, and therefore should have a separate page explaining it. It's not like jamming the two pages together will preserve storage space, and I'd also like to point out that Wikipedia:Edit war isn't accepted as policy (it was first marked policy only a few days ago, and has been de-marked several times since then). Mangojuicetalk 14:16, 13 November 2007 (UTC)
  • Oppose. The prohibition on edit warring is a general principle we should all strive for while the 3RR is a hard and fast rule with no shades of gray and exceptions in only the rarest of cases. No need to muddy things by merging the two. Gamaliel (Angry Mastodon! Run!) 18:22, 13 November 2007 (UTC)
So you would prefer that we confuse newcomers with blatantly redundant policy pages? ;) The proposal isn't that we close up the 3RR shop, only that we admit what is already and long has been the case: 3RR is an electric fence, but not the only path to a block. Edit warring is bad. – Luna Santin (talk) 04:40, 14 November 2007 (UTC)
Hear, hear. Heimstern Läufer (talk) 05:22, 14 November 2007 (UTC)
I would prefer that we not confuse newcomers with overly complicated multi-purpose policy pages. Sometimes an actual screwdriver is preferable to a leatherman. Mangojuicetalk 06:09, 14 November 2007 (UTC)
That assumes that 3RR and EW address distinct issues, when it seems to me they address the same issue. Going with your screwdriver metaphor, you seem to be suggesting we should keep distinct pages to describe slot screwdrivers and also slot screwdrivers which happen to be three inches long, or seperate pages on cleaning up vandalism in general and penis vandalism in particular... they seem like pretty much the same thing, at least to me. Given that the express purpose of 3RR is to prevent edit warring, I don't see the basis for the implied claim that they're not directly related -- not to accuse anybody of bad faith, at all, I just don't see the basis. I don't see anything confusing about saying "edit warring is frowned upon, and can get you blocked; 3RR is one common measure of edit warring... (and so on)." – Luna Santin (talk) 08:29, 14 November 2007 (UTC)
  • Approve, though this needs a lot of discussion and would take considerable work in implementation. Beyond the question of where the policy cits, we should not confuse newbies, oldsters, and ourselves by inconsistent administrative actions, by which users may sometimes go up to 3RR without blocking, people are blocked for 3RR even when following the listed exceptions, and occasionally people are blocked and pages protected even though they are not up to 3RR. There needs to be a single, consistent rule. If you don't give people a reasonable set of criteria to know whether their behavior falls inside or outside the range of accepted conduct, and a good place where they know to find it, the efforts will be arbitrary and ineffective. Wikidemo 06:36, 14 November 2007 (UTC)
  • Strongly Oppose for the reasons aready stated. Brimba 16:07, 14 November 2007 (UTC)
  • Support As others have said above, 3RR is a remedy but edit warring is the root illness. In some ways the 3RR system has become something of a crutch and a hindrance towards tackling the truly problematic behaviors on Wikipedia. How many times have we seen someone make Revert #3 at hour #25 and so forth? Worse yet it encourages the use of sock and meat puppets to further help game the system. A stronger policy with teeth would tackle the heart of the issue-edit warring and not just one of the means that folks edit war. WP:EW is a step in the right direction and while 3RR should not be abandoned, I think a collaborative endeavor to merge these two pages would help strengthen the community's overall stance against edit warring. AgneCheese/Wine 18:42, 14 November 2007 (UTC)
  • Mild Support I've raised a concern on Edit Warring talk about tag teaming which I think should be in there. It is one of the reasons 3RR has fallen into disrepute in my personal view. Having been complaining about edit warring and had the minutia of 3RR thrown in my face by admins, even when I've complained about the spirit, to me this has to be the right direction. I'd like to think that 3RR could wither and die, simply there as a backstop for peculiar situations. Spenny 19:04, 14 November 2007 (UTC)

Too quick??

While I support the idea of merging the two policies into one, I think we may be just a bit premature at this point. The edit warring policy has only fairly recently moved from guideline to policy and seems to still be in just a bit of flux. I would suggest waiting until after the first of the year and then merging the two. I do feel that edit waring is a fundamental concept which will likely be applied more often than 3RR. We ought keep both, in one document, but at the right time. JodyB Roll, Tide, Roll 13:18, 13 November 2007 (UTC)

How about if someone adds suitable 3RR language into WP:EW, then we can have a discussion (once WP:EW itself is stable as you said) on whether a separate page for WP:3RR is necessary? —Random832 14:11, 13 November 2007 (UTC)
It is moving quickly, yes, but I don't see any reason it would be a bad idea. Edit warring is bad, and can lead to a block -- 3RR is just a particular case where you're very likely to be blocked. It's certainly worth a section or two in any merged policy. Keeping the pages separate seems more likely to cause confusion, in my eyes. – Luna Santin (talk) 04:44, 14 November 2007 (UTC)
I am not saying it's a bad idea. In fact I like the idea - in time. It's just that there needs to be some time for the two stand together before 3RR is depreciated. Edit warring is bad and always will be. 3RR is a little confusing but does contain the tools to deal with the problem. I just think we should give a little more time to allow the community and the administrators to become aware of, and more comfortable with, the new policy. Of course, if the community chooses to move ahead, that's certainly fine. JodyB Roll, Tide, Roll 12:54, 14 November 2007 (UTC)


While I continue to oppose the concept in principal, I think it would move the discussion forward if someone would actually create a draft (on a subpage) of what the combined policy would look like. That way, editors could compare what exists today to what is proposed to replace it, rather than comparing existing policies to some hypothetical page that actually may not be the same in everyone's mind. -- John Broughton (♫♫) 13:37, 14 November 2007 (UTC)

  • I can write a draft later tonight, I'll subpage it here... unless someone beats me to it. :) Mercury 20:21, 14 November 2007 (UTC)
I like the way you blocked it out, and I like the way it doesn't have so much commentary as the new policy page. Just as a caution, proposing a draft at this point is useful to illustrate what it can look like, but it is a content spork. Now we have three pages that say the same thing - 3RR, edit war, and the draft. I just did a heavy copy-edit to the new policy, which I thought was weak in language and organization. Wikidemo 02:05, 15 November 2007 (UTC)

Different direction

(Sorry for the section-header-cruft.) I'm afraid that merging 3RR into WP:EW will only serve to dilute the recent improvements. The essence of the policy on edit warring is this: edit warring is the specific prohibited behavior; 3RR is one possible metric we use for measuring edit warring, and it is a common one, but violations of 3RR are only blockworthy insofar as they constitute edit warring, while at the same time, behavior need not violate 3RR to be edit warring. As such, I think the direction we need to move is towards making Wikipedia:Edit war the primary policy in this area for which blocks are made and such, and snip this page down to a reasonable guideline on that useful 3RR metric that is commonly used to determine edit warring, with reference to how it fits in to the wider edit warring policy. Dmcdevit·t 03:29, 15 November 2007 (UTC)

  • Oppose, bad idea. Edit warring isn't allowed, 3RR defines when you are guaranteed to get blocked for it. They are both required. Stifle (talk) 17:04, 20 December 2007 (UTC)

Exception relating to reverts within user space

The exception relating to reverts within user space was added on 14 June 2005 by David Gerard: see here for the edit adding it. The precise wording has been changed since then but the exception is a long-standing one and I have therefore reverted the recent edits removing this exception in its entirety. Sam Blacketer (talk) 22:11, 18 November 2007 (UTC)

Thanks, I was confused and wrong. FloNight proposed removing it [16] and the discussion above in the section 'user space' shows that several others agree. Can you explain what purpose is served by allowing users to edit war in their userspace? — Carl (CBM · talk) 22:31, 18 November 2007 (UTC)
As I see it, the point isn't to allow users to edit war in their own space, but to prevent other editors from edit-warring with them. The general theory of WP:OWN doesn't apply in the same way in user space; normally one would leave it to the user whose page it was to decide what went there. Of course there are specific things which users put on their user pages which cause problems, and they normally go to MfD or get speedy deleted. But in general, we should allow users to control their own userspace and it would be wrong to put them in a position of being blocked for defending their way of writing pages in their own back yard. I'm not saying I would never block for edit-warring in user space; it can be disruptive. Sam Blacketer (talk) 22:40, 18 November 2007 (UTC)
I'm sorry, I believed this edit. In any case, it is my firm belief that the user page (and any subpages) associated with an account is there for the user to express their opinion (an IP Address does not get a user page because an IP Address is not associated with any one person), and the talk page associated with an account or IP Address (also called a user talk page due to the way MediaWiki works) is there for others to express their opinions about the user or IP Address, and of course for the user to respond in an appropriate manner, and later to pre-respond by adding some content at the top. Legitimate warnings, especially for vandalism, need to be kept on the user talk page, or an easily-visible archive of that page, in order for anti-vandalism efforts to succeed. It is also my belief that the current language for the exception, "reverts performed by a user within his or her own user page, user subpages" per this exception does not include "User talk page", the third component of "user space" on the linked Wikipedia:User page, or this exception would read "reverts done by a user within his or her own user space". What purpose is served by letting users edit war on the user talk pages allocated to their own accounts and IP Addresses? Thanks!   — Jeff G. (talk|contribs) 23:29, 18 November 2007 (UTC)

Recommended change, new exception

What about reverting bot edits that are clearly disruptive (i.e. the bot is "broken" or malicious), or obviously unneeded (many hyperspecific cases, things that were good at the time but bad now, also good faith edits by users)? Obviously, first you go to WP:ANI about bad bots, but what about cleaning up the mess? I know you should not have to make three reverts, but what if you already made three? --Thinboy00 talk/contribs @166, i.e. 02:58, 19 November 2007 (UTC)

I think, if you have already reverted three times and have to undo a bot mistake, you were probably reverting too much previously. But anyway, if you see a bot making a mistake, the first thing to do is to bring it to the attention of admins, because malfunctioning bots should be blocked. (Also, the bot may have a "shutoff button," and the bot operator may be online, so try dropping them a note.) If there's a simple and obvious bot mistake going on, someone will revert the change so you don't have to. Mangojuicetalk 03:45, 19 November 2007 (UTC)
Mango is right, the priority is to stop the bot. Cleaning up after it can wait until it is blocked. --bainer (talk) 03:56, 19 November 2007 (UTC)
In fact it's definitely better to wait: if the bot's malfunctioning, it could just redo the change it did. Mangojuicetalk 15:38, 19 November 2007 (UTC)

Reverting blatantly inappropriate usernames

I'm of the mind that a statement should be added that reversions of edits by users with blatantly inappropriate usernames are not subject to 3RR. The reasoning? If a user is not allowed to edit with an inappropriate username, it follows his or her edits can be reverted without the need to discuss them, much as is the case with reverting socks. Blueboy96 17:04, 10 December 2007 (UTC)

If a user has an inappropriate username, the proper thing to do is to report this at Wikipedia:Usernames for administrator attention. There is no need to start reverting their edits as long as they are not vandalism or otherwise inappropriate. Lova Falk (talk) 18:18, 10 December 2007 (UTC)
Oh, I do report them ... it just seems to follow that if a user can't edit with a blatantly inappropriate username, any edits made under that username can be reverted without discussion since they have no right to make the edits in the first place. Note that I emphasize blatantly ... like a person whose username contains profanity, is promotional, implies intent to cause trouble here, etc. Blueboy96 18:42, 10 December 2007 (UTC)
I understand what you say, but I still don't agree. The article is much more important than the editor. If the edit is okey I wouldn't revert it. Lova Falk (talk) 19:49, 10 December 2007 (UTC)
PS The difference between you and me is that I think everybody has the right to make edits, including people with inappropriate usernames - with the exception of people who are blocked. Lova Falk (talk) 20:17, 10 December 2007 (UTC)
Where on Earth did you get the idea that it's policy that the edits of username-blocked accounts can be automatically reverted? --bainer (talk) 01:15, 11 December 2007 (UTC)
I was under the impression that inappropriate usernames are like sockpuppets--they aren't allowed to edit, so they can be reverted without discussion. I'm willing to admit if I'm wrong, though. Blueboy96 13:54, 11 December 2007 (UTC)
If the edit isn't vandalism, there's no reason to treat the edit differently from any other content-related edit, and an inappropriate username is no reason to edit war. Adding this wouldn't make sense. Heimstern Läufer (talk) 03:00, 11 December 2007 (UTC)
Agree with many of the others. If someone has an inappropriate username and they make an edit, it might be a good reason to question it, maybe to revert it. However, edit warring is WAY more harmful, so this should be no reason to violate the 3RR if no other exception applies. Mangojuicetalk 14:15, 11 December 2007 (UTC)
  • (edit conflict)Agree with Thebainer and Heimstern, but I suspect that Blueboy96 and you all are talking past each other and in practice would likely do the same thing. An highly inappropriate user name suggests that we need to examine the edits closely to look for problems but it does not mean that all edits should be automatically reverted. If the majority of the first few edits examined are blatant vandalism type edits, then quickly reverting all of the edits made in that session of editing would likely not be seen as controversial, even though policy does not directly support it. I think most people would agree that it would be a waste of time for us to closely scrutinize the edits looking for good content. FloNight (talk) 14:20, 11 December 2007 (UTC)
Yes, if there's a problem with the edits then there's a problem with the edits. But it should be clear that that's solely to do with the nature of the edits themselves, and nothing to do with the account. --bainer (talk) 14:25, 11 December 2007 (UTC)
Yes, but reality on the ground is that classifying repeatedly removing edits of what is a self promoted vandalism only account (because that is what these folks are doing when they so name these account) as edit warring and sanctioning an editor for repeatedly removing a good edit or two (perhaps even planted to increase the drama) is not likely to happen. But I do not agree that the policy needs to state this reality as it is instruction creep. FloNight (talk) 14:51, 11 December 2007 (UTC)

"applies to all editors individually."

Why? All it does is encourage the development of revert-warring "camps" of editors. I think it's obvious and should be part of this guideline that in the case of multiple involved editors, still every revert after the third on each "side" will be treated as a 3RR violation. Otherwise it's madness. Can the parties in one camp really be regarded as independent editors? How sane is that assumption? Even if it may be so every once in a while, exclusion here factually encourages that kind of "camp" (or "side, or "clique", whatever you wanna call it) behaviour. And how exactly can it be justified to exclude the reverters who revert for the fourth time on each side from 3RR? They know all too well what they are doing, worse: they are Wikipedia:Gaming the system in an egregious way. Any responsible editor would never do the same revert for a fourth consecutive time, save for cases of obvious vandalism. Any responsible person would step back instead of continuing, ideally after the very first revert, but no later than three reverts have happened on each side. And why this excplicit exlusion of such multiple-editor-situations in the first place? I dorftrotteltalk I 04:52, December 14, 2007

Well, let's consider this situation: editor arrives and adds a fringe interpretation to an article, clearly giving it undue weight. Another editor reverts. First editor then reverts to his/her version. A third editor reverts. First editor reverts. A fourth editor reverts. And then the first editor reverts. So, do we consider the second, third and fourth editors a group? If so, that means no one can revert to the older version and it's locked with the fringe interpretation until the 24 hours pass.
The essential reason for not including groups in this rule is so that the general community can enforce consensus on the article. Put another way, it stops one editor from being able to foist his or her edits on the article and have just as much say as a group of editors. It is true that this does not stop users from ganging up to force their POV on an article, and the rule has never really handled this satisfactorily. (It's partly for this reason that I encourage admins not to enforce the rule mechanically, but to use careful judgment. Or better, to think of Wikipedia:Edit warring as the real policy and this as a mere yardstick to aid enforcement of that one.) Still, I think applying the rule to groups and individuals equally would cause more problems than it would solve. Heimstern Läufer (talk) 06:16, 14 December 2007 (UTC)
Excluding groups from 3RR doesn't enable the community to enforce consensus, but instead to force through a simple brute-force majority rule. Some admins mechanically enforce 3RR, as you put it, precisely to gain the upper hand or assist other, likeminded users. And those on the receiving end or 3RR warnings and blocks are not much less likely to be on the side of actual community consensus: the community isn't a tightly-organised group at all, and this is an unspoken premise in all our rules and closely related to the spirit of Wikipedia as a whole.
Unfortunately, the ideal that "consensus will eventually prevail" (by means of a community made up of fair-playing individuals, which encourages consensus-seeking and discourages gaming and brute-forcing) is no more than this, an ideal, an empty shell buzzword if it isn't enforced through policy. Instead, questioning and criticising this type of gang behaviour is discouraged as assumption of bad faith, calling people on it is near-impossible per NPA.
All in all, it's commendable that you "encourage admins not to enforce the rule mechanically", but in reality, this policy loophole (like other, similar loopholes) can be (and is frequently being) abused. In its current state, the policy actually enforces this type of behaviour.
As far as WP:EW goes: Too weak, not suitable to stop this kind of thing. I dorftrotteltalk I 08:29, December 14, 2007
Do you have a solution that avoids the problems I've suggested? The current system arguably allows groups to game the system. Changing it to include groups would allow individuals to game the system. Is there something you think we could do to avoid either? Heimstern Läufer (talk) 06:26, 16 December 2007 (UTC)
Counting the initial edit as the first revert would mean that groups of editors as well as individual editors could be forced to discuss the change and defend it with arguments; the first 3RR would occur when the initial change is re-instated for the second time. Yet even if that happens, the other side should not continue, and instead call for outside opinion/intervention. Let an uninvolved user do the revert to the initial version if necessary, to deny the option of "taunting" the other side. I dorftrotteltalk I 12:22, December 17, 2007
Who are you going to enforce the policy against? This proposal implies that all editors read edit histories before contributing. That's not a requirement, nor should it be. It should be impossible for an editor to accidentally "fall into" a 3RR violation on her first edit. --- tqbf 00:55, 9 January 2008 (UTC)

Scope question

Is the "all pages" referred to across the board or just within article space? I'm asking since I'm running into an instance of an non-free image getting constantly reverted to an over-sized version on the image page. - J Greb (talk) 19:21, 24 December 2007 (UTC)

It refers to all namespaces. Heimstern Läufer (talk) 21:46, 26 December 2007 (UTC)

How to avoid violating 3RR?

I'm having problems trying to figure out what to do to avoid violating 3RR. Say an editor adds a clearly NPOV, unsourced addition to an article (say...oooh, for example, here, I can revert that, they can re-add it, I can revert, they can re-add - if I revert again, I'm hosed. This page doesn't help me with what I should do at that point. Torc2 (talk) 21:43, 26 December 2007 (UTC)

You should not be repeatedly reverting to solve problems with pov editing. First attempt to solve the problem through discussion on the talk page. Involve other editors if there is an disagreement. An article RFC or mediation can used if the situation is not resolved. If user conduct is an issue, then an user conduct RFC might be needed if the situation can not be resolved by educating the user about Wikipedia policly. Hope that helps. FloNight (talk) 21:50, 26 December 2007 (UTC)
Thanks, it does, especially the Article RfC suggestion. I think it would be useful to include this information in the WP:3RR article itself. Torc2 (talk) 21:54, 26 December 2007 (UTC)
We do have a section on alternatives to edit warring at the edit warring policy page. Perhaps that should be included here, too. Heimstern Läufer (talk) 21:56, 26 December 2007 (UTC)
Even just a pointer there would be sufficient, especially if the information is just going to be duplicated. Thanks for your help. Torc2 (talk) 21:59, 26 December 2007 (UTC)
As FloNight says, the objective of the rule is to stop stale edit wars between small numbers of people by making it an attractive option for them to get more people involved in productive discussion. Wikipedia:Third opinion is another avenue to try, though that is a less trafficked path than Wikipedia:Requests for comment. You could also consider notifying WikiProjects that may have an interest in the area, although take care to express your message neutrally. --bainer (talk) 23:43, 26 December 2007 (UTC)

OK, I added a short section on this, mostly pointing to the information in other guidelines and policies. I hope it's OK. Thanks again for all the help on this.Torc2 (talk) 01:12, 27 December 2007 (UTC)