Wikipedia talk:How to revert a page to an earlier version/Poll: Revert wars considered harmful

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

See Also:

Poll: revert wars considered harmful[edit]

Wikipedians who support[edit]

Wikipedians who support the guideline: "do not revert the same page more than three times in the same day" include:

  1. Jimbo Wales
  2. Martin (and preferably less)
  3. Angela
  4. Viajero
  5. Andrewa
  6. RickK
  7. Bcorr
  8. Cimon Avaro on a pogostick
  9. JamesDay (you'd do it as many times as that?)
  10. Baldhur
  11. Stan Shebs (Wik's opposition convinced me)
  12. Kosebamse
  13. Uncle Ed Once per day is enough for me.
  14. Alexandros
  15. mav 18:28, 14 Feb 2004 (UTC) (3 revert rule - more than that is vandalism and should result in a 24 hour block)
  16. Jor
  17. Metasquares There is no reason to revert a page more than 3 times in a single day. If there are that many, the page should be temporarily protected and sorted out by a sysop.
  18. till we *)
  19. Dori - drugs are bad, m'kay...I mean reverts are bad :)
  20. Hephæstos|§
  21. Anthony DiPierro In theory, though there are exceptions.
  22. Tannin Note: as a guideline, people seem to be claiming it is a rule. To my astonishment, I find myself agreeing with Anthony DiPierro, who writes: "In theory, though there are exceptions."
  23. Fennec - if it's a problem with vandalism, then get someone else to revert it. If it's an edit war, mere reversion is pointless. If someone else breaks the rule, get them censured and/or banned, as necessary, and have someone else revert it. :)
  24. →Raul654 05:19, Feb 16, 2004 (UTC) (Wow. Lir, Wik, and Anthony all agreed on something. Scary.)
  25. —Eloquence (if consistently enforced)
  26. Tuf-Kat <Insert witty caption here>
  27. Infrogmation
  28. Bryan though it will be necessary to make allowances for dealing with people who simply ignore this guideline, since otherwise it puts the "law-abiding" Wikipedians at a significant disadvantage.
  29. Cyan
  30. Sam Spade 04:24, 23 Feb 2004 (UTC) But only if there are NO exceptions for ANYONE (no matter how big their ego is ;). There needs to be a better version of article disputes that is similar to the Conflict resolution process.
  31. Dandrake My naive question: if it's in a triple-revert war, isn't it time to stop messing around? I.e., Metasquares is right.
  32. DanKeshet
  34. Danny
  35. Seth Ilys. This is a good idea only if it is strictly enforced.
  36. User:Fred Bauder
  37. Ruhrjung 13:49, 23 Feb 2004 (UTC) - three a day is far too much already
  38. RoseParks 18:18, 23 Feb 2004 (UTC)
  39. Ams80 18:34, 23 Feb 2004 (UTC)
  40. Flockmeal 20:50, Feb 23, 2004 (UTC)
  41. Nohat 21:09, 2004 Feb 23 (UTC)
  42. silsor 02:38, Feb 24, 2004 (UTC)
  43. James F. (talk) 03:13, 24 Feb 2004 (UTC)
  44. Meelar 13:06, 24 Feb 2004 (UTC). Recent experience has changed my thinking--the cost of not having this policy is a little too high.
  45. Catherine 19:29, 25 Feb 2004 (UTC)
  46. denny vrandečić 20:15, Mar 5, 2004 (UTC)
  47. Ryan_Cable 14:59, 7 Mar 2004 (UTC)
  48. Delirium 00:34, Mar 8, 2004 (UTC)
  49. Kingturtle 06:29, 10 Mar 2004 (UTC)
  50. Robert Merkel 11:32, 10 Mar 2004 (UTC)
  51. FirmLittleFluffyThing definitly as a guideline, not decided as a rule.
  52. Stewart Adcock 23:41, 10 Mar 2004 (UTC)
  53. Sean 08:18, 12 Mar 2004 (UTC)
  54. +sj+ (and preferably less than 3; little alarm bells should start to go off as soon as a page is reverted twice in 24 hours by any one person... more than thrice should be actively avoided (ignoring reversions of anonymous edits))
  55. Nephelin (in mind: sysops are normal users)
  56. d8uv (t) 02:28, 2004 Mar 27 (UTC) guideline
  57. Cribnotes 18:31, 4 Apr 2004 (UTC)
  58. Jonathan Grynspan (yeah, I can't for the life of me figure out how to insert a time without manually typing it in)
  59. Dark Tichondrias 06:23, 4 July 2006 (UTC)[reply]
  60. --Ye Olde Luke 02:33, 30 October 2007 (UTC)[reply]

Wikipedians who oppose[edit]

Wikipedians who oppose the guideline: "do not revert the same page more than three times in the same day" include:

  1. Daniel C. Boyer
  2. Taku
  3. Daniel Quinlan
  4. Wik 13:27, Nov 18, 2003 (UTC)
  5. NetEsq 00:06, 21 Nov 2003 (UTC)
  6. Lirath Q. Pynnor If you don't like reverting -- then protect the page in question
    Anthony DiPierro Not sure how useful this is, but I'm not opposed to it as a guideline.
  7. Ævar Arnfjörð Bjarmason 01:28, 2004 May 14 (UTC): I never did like arbitery limits, and even though it is a very good general rule there might be some cases where it is needed, though second opinion should be sought as always.
  8. msh210. I do not like revert wars, but I've found it necessary to revert a page twice in a day; see, e.g., DOTA, 29 Mar 2005, and I can easily imagine some persistent vandal making me revert the same article thrice or even, um, fries in a day.msh210 16:30, 14 Apr 2005 (UTC)

Wikipedians who think it pointless[edit]

Wikipedians who think its pointless to declare an "end to revert wars" without advocating another means of solving disputes:

  • Lirath Q. Pynnor
  • Anthony DiPierro I don't see what's wrong with revert wars in the first place. They're a symptom of a much more difficult problem: getting thousands of people across the world to agree on a single version of a page. Stopping people from reverting an article more than a certain number of times a day isn't going to solve that.
  • Tannin (With reservations: i.e., I respect the point but recognise that we have to be more proactive with persistent troublemakers.)
  • Cyan

In rare cases, one might consider having more than one version of the page. Sounds radical, but might resolve some edit wars and since Wikipedia is not a paper encyclopaedia, it doesn't have to be exactly like a paper encyclopaedia.

Having more than one version should be routine.

  • User:Fred Bauder There is more than one valid way of looking at many things.
  • ugen64 - to my experience, there's two kinds of revert wars: anti-vandalism, and a "troll" being reverted repeatedly for no reason at all (Wik once corrected a spelling error and removed extraneous info, then was repeatedly reverted by none other than Tim Starling).


I don't know about you lot, but I'm going to follow this rule for myself, as much as I can - and I hope others will follow it too. Revert wars on Wikipedia are typically an indication that two or more people need smacking around the head with a clue stick. Martin 21:18, 14 Nov 2003 (UTC)

...or one person is an unrepentant vandal or spammer, as with the person who tried it on the various belly dance related articles today. --Modemac 21:27, 14 Nov 2003 (UTC)
Looks to me like you didn't revert belly dance or other articles more than once. Eg Raqs al Sharqi - you reverted once, and Mav reverted once.
You're right: there are times when this guideline is inappropriate, but I think they're surprisingly rare. Martin 21:33, 14 Nov 2003 (UTC)

Two issues come to mind:

  • What about vandalism? Perhaps an exception needs to be made?
  • What about this problem
Person 2 makes edit
Person 1 reverts - 1
Person 2 reverts - 1
Person 1 reverts - 2
Person 2 reverts - 2

So now Person 2 will always "win" this revert war, no matter what limit is alowed... Dysprosia 22:15, 14 Nov 2003 (UTC)

Person 1 may make another edit to an alternative third version. Now person 2 may not revert without going against our guidelines on numbers of reverts. Person 3 may come along. In any case, Person 2 will only "win" for a day - the following morning, person 1 may, if they still believe they are correct, revert again. Martin 22:35, 14 Nov 2003 (UTC)
Oh right, I see now, so this limit applies to reversions of the same version, not absolute reversions... What about vandalism, still? Will this be excluded? Dysprosia 22:46, 14 Nov 2003 (UTC)
Some of LibertarianAnarchist's edits/vandalism has been reverted many more times than that. This doesn't work with vandalism :( Secretlondon 22:49, Nov 14, 2003 (UTC)

It's a guideline - in some cases of vandalism you might want to make an exception - but a quick scan of wikipedia:vandalism in progress suggests that with vandalism, a revert war is rarely necessary - most vandals make a few dodgy edits, and then leave.

LA is obviously a difficult case, and I don't want to be simplistic about him, because there are no easy answers to such problems. I think the spirit of the guideline is clear enough, even if attempting to apply it too rigidly could be a mistake. Also, I note that LA is very definately going against this guideline, both in spirit and in letter. It's another thing to point him to when explaining to him why his behaviour is inappropriate... and that may even lead to a ban, as a last resort. Martin 23:17, 14 Nov 2003 (UTC)

I mean the problem is the prevalence of relentless edit wars that may weed out valuable contributors in the worst case, but not the number of reverting. A kind of imposing numerial limitation is analogous to protecting the article and it cannot be the answer. -- Taku 23:48, Nov 14, 2003 (UTC)

Surely if the problem is that the edit was are relentless, then this would be a good solution?
One difference between this guideline, and protecting the article, is that this guideline is soft security, whereas getting a sysop to protect an article is hard security. Thus they are very different things. Martin 00:25, 15 Nov 2003 (UTC)
If the guidline is that soft then what is difference between it and just saying don't engage in edit wars. I personally don't revert someones' edits several times aside from vandal reasons because I don't want to go to the edit war. -- Taku
It's not a soft guideline: it's "soft security". To quote MeatballWiki: That's not the same as weak security. The idea is to protect the system and its users from harm, in gentle and unobtrusive ways. Soft security is like water. It bends under attack, only to rush in from all directions to fill the gaps. It's strong over time yet adaptable to any shape.
I could say "don't engage in edit wars" (and, elsewhere, I say something similar) but on this page I'd like to strongly discourage revert wars, which I feel are a particularly counter-productive instance of the edit war species. Martin 00:47, 15 Nov 2003 (UTC)

I moved a following section from the main page. It is not a consensus at all. -- Taku 00:08, Nov 15, 2003 (UTC)

That's cool - I reworked it back above. Martin 00:22, 15 Nov 2003 (UTC)

I think it's important to point out that it doesn't have to be a fight between two people. If person 1 is right, then the chances are person 3 will come in to support them, and if person 1 and 3 have both reverted three times, then it's probably time to protect the page. This happened tonight on Freemasonry and Knights Templar where I decided to stop reverting after the third one. Reverts were also made Cyan, Cimon avaro and Hephaestos and then the pages were protected by Minesweeper and Cimon avaro. I think the fact that there was more than one person involved in reverting may have helped Chiramabi to understand what was happening, rather than thinking it was just one person who didn't agree with him. So, even though Freemasonry was reverted 15 times, no one apart from Chiramabi had to revert it more than three times in the same day. Angela 05:10, 16 Nov 2003 (UTC)

Well put, Angela! Perhaps you could alter the proposed text to make that more explicit? Martin 13:37, 16 Nov 2003 (UTC)
I haven't changed it much. I think the "If the edit really needs reverting that much, somebody else will do it" covers it really. Angela

I worry this proposed change will only benefit people who don't play nice, I would prefer to have the ability for sysops to block/unblock pages for specific problem users. Saying three reverts indicates a problem is unfair to typical editors who run into unreasonable people. Daniel Quinlan

But whether you are reasonable or not is just your point of view. By sticking to the policy, you can see whether other people support you and whether you really are reasonable or not. Everyone in an edit war claims they are the reasonable one. Angela

If you need to revert more than once for something other than vandalism, it's time to get the community involved, IMO. That sends a much better message to those doing the problematic things. Look over Wikipedia:WikiProjects, seek Wikipedia:Peer review, ask for others to assist in the Wikipedia:Village pump or do other things instead of participating in escalation. JamesDay 12:25, 17 Nov 2003 (UTC)

The problem with edit wars is not that the Recent Changes are cluttered up. Forcing a reduction in the frequency is not solving the underlying problem of how to settle a content dispute where a consensus can not be reached by discussion. I see only two options there, either a voting procedure or a decision by a special authority to be created for that purpose. In any case there would have to be a final decision which would then be binding on both sides in the dispute. --Wik 13:27, Nov 18, 2003 (UTC)

Well continuing to revert more than three times isn't going to solve a content dispute either. Angela
A consensus on the text of Wikipedia articles can always be reached by discussion, where both parties are prepared to discuss. The aim should not be to "settle" the dispute, judging one side right, and the other side wrong, but to create a text with which neither side disagrees. That is the definition of a neutral point of view. Martin 18:59, 18 Nov 2003 (UTC)
That is quite obviously not true in practice. --Wik 19:10, Nov 18, 2003 (UTC)
Counter-example? Martin 19:23, 18 Nov 2003 (UTC)
Republic of Macedonia, Mother Teresa, Silesia, Adolf Hitler, ... --Wik 19:37, Nov 18, 2003 (UTC)
Republic of Macedonia is chiefly a dispute on the title of a page, not the text of the page. The dispute on the text of the page is basically solved - it's at the tweak stage. Certainly, I agree that in some cases it's not possible to get a unanimous consensus on the best title for a page. On the other hand, a series of move wars won't help either, as Angela points out.
Mother Teresa is being solved, by discussion. This will take time and effort, for it is a controversial subject close to people's hearts. I didn't say that consensus can be reached instantly, or without effort - merely that it can be reached. See this post for evidence that discussion was producing.
On Silesia, we currently have one party who is refusing to discuss: Caius2ga. Prior to that, we had someone else refusing to discuss. However, note that the conflict between Nico and szopen was solved, and it was solved by discussion, once both were prepared to discuss. This is evidence for my approach, not yours.
Adolf Hitler is much closer to consensus than it was prior to discussion, because the alternate solution of having a spelling note at the end of the article (rather than in the introduction) has been suggested and implemented. People seem sufficiently happy with the current version of that endnote, which was arrived at through discussion. Again, evidence for my approach, not yours.
Reaching a consensus requires effort. It requires research. It requires communication. All in all it involves a bigger investment in Wikipedia than a series of random revert wars. But once it produces, it produces solutions that last. Revert wars only ever get as far as an uneasy ceasefire - broken as soon as the balance of power twitches momentarilly. Think in terms of the Long Now. Martin 21:46, 18 Nov 2003 (UTC)
I don't think consensus is always going to be possible. What about September 11, 2001? I don't think there will ever be a consensus as to whether or not to use the word terrorist in certain situations. Sometimes you've just gotta vote, it seems. Hopefully you can make the votes broad enough so that they'll cover lots of different situations, but that's the best you can do. Anthony DiPierro 05:56, 16 Feb 2004 (UTC)
I agree. Polling wars are usually left as far too personal an issue, in my opinion. Once identified, they should be open to polling and larger debate. If we can keep the debates channeled into the relevant discussion pages, it should calm things down a bit and converge towards a solution. --boiled_elephant 10:13, 7 February 2006 (UTC)[reply]