Forums: Index > Wiki discussion > Ambiguous guidelines - What is a "revert"?

I would like to address an issue in our guidelines regarding edit warring. My problem is that it is ambiguous in its definition of "reverting", which may lead to inconsistent admin intervention. In this discussion I hope to find a solution to the ambiguity.

For starters, the edit warring guidelines currently read as follows:

Do not edit war: Be ready to discuss your changes with others. If you disagree with another editor, discuss the issue either on user or article talk pages. Repeatedly reverting each other's changes ("edit warring") is bound to escalate the conflict instead of solving it. If you cannot reach a consensus, ask an uninvolved user to mediate. In general, if someone reverts an edit you made, you should not re-add it without reaching a consensus on the article's talk page. Consensus does not have to be reached in cases of disruptive editing.

The problemEdit

As it turns out, the guidelines are ambiguous in their definition of "reverting". In particular, it does not properly explain what does and what doesn't constitute a revert, which has led to a discussion on whether this revision counts as a revert. I argued that it is a revert because it essentially undoes the previous edits, while Agent C and Richie9999 argued that it is not a revert because "reverting" means "use of the undo tool [or the rollback tool]". (See Agent C's messages in #the-editorial-bullpen at 2018-06-25 23:09 (UTC).)

I think we should decide on a single definition of "reverting" for the sake of the guidelines and explicitly write it in the guidelines. With regards to this discussion, I think it is important that we agree that the definition should not be applied retroactively, and that this discussion is not about who is right, but about what is best for Nukapedia.

The positionsEdit

Strict definitionEdit

First I will present a strict definition that is objectively verifiable: If the undo/rollback tool has been used and another user then uses the undo/rollback tool on the other's revert, it's an edit war. It is not possible to debate whether something is an edit war, and detecting edit wars would be very easy. Furthermore, admin intervention would be consistent since there is no ambiguity in whether the guidelines were violated.


There are two problems with this definition, however:

  1. Usage of the undo and rollback tools is recognised by the distinctive edit summary. However, when you use the undo tool you can change the edit summary. Since there is no tag for usage of the undo/rollback tool, doing so would make it ambiguous whether you used the undo tool or manually undid the changes.
    • This problem could be tackled by adding tags for these edits, though I have no idea how we could do that. (Contact Fandom?)
  2. This definition excludes complex scenarios where users perform partial reverts or where one user reverts and the other manually adds the contents back in.
    • I am not sure how we could work around this problem.

Loose definitionEdit

Now I will present my own definition, which is actually Wikipedia's definition: "Reverting means undoing or otherwise negating the effects of one or more edits, which results in the page (or a part of it) being restored to a previous version."

This interpretation includes complex edit wars regardless of what tools are used. One user could undo an edit, the other user could add it back in slightly differently, and the first user could undo that edit and it would still count as an edit war. I think this definition is more appropriate considering that edit wars rarely see exclusive usage of the undo/rollback tools, and because I think the community interpretation is closer to this definition to begin with. Consider the following real-world examples:


However, the weak point of this definition is that it would consider some scenarios edit wars even though they aren't. Consider this example:

Even though DirtyBlue929 "negated the effects" of Great Mara's edits, s/he also reworded the edit to respond to Great Mara's reason for undoing the initial edit. I would consider this a valid, non-edit warring response that actively tries to work towards a consensus by identifying the exact concerns of the other user through boldness.

The best way to prevent the problem with my definition is by applying good faith: If someone reverts an undo (regardless of whether they used the undo/rollback tool) and seems to be working towards a consensus, it is not an edit war. (And, conversely, if someone undoes an undo without working towards a consensus, it is an edit war.)

The third definitionEdit

Of course, it might be that there is a third interpretation of what constitutes a revert. I could not think of one, but if you can you can add it in the comments.


The current guidelines on edit warring do not define what a revert is, which has led to different interpretations, which may lead to inconsistent admin intervention. I have presented two interpretations and have shown their strengths, their drawbacks, and possible solutions to their drawbacks. Other interpretations may exist as well. Regardless of what interpretation is most suitable, the community should decide on one in order to remove ambiguity from the guidelines.

- FDekker talk 15:47, July 23, 2018 (UTC)


Recent versions of Mediawiki label reverts, both from rollback and the undo button. 2604:6000:B48F:85F0:807A:1EEF:A05D:EB5D 18:26, July 24, 2018 (UTC)

That is interesting. I doubt Fandom will update their MediaWiki version any time soon, but it does indicate that it's possible to label usage of those tools.
- FDekker talk 20:04, July 24, 2018 (UTC)
I think this one we can simple. Firstly, I'd like point an issue out with this thread and it has been a long term bugbear of mine: Discussion in Discord cannot be seen by everyone who may have an interest in the subject. If there is to be discussion using point from Discord, they should be brought over here (via quoting or screenshot) so all parties are equally informed.
Anyway, to keep this simple let's go with the textbook definition. To revert is to go back to a prior state. Tools used are irrelevant in how you reach that state, the revert is your destination.
Editwarring in definition is a disagreement between users who are flipping the contested point back and forth, instead of discussing. My view is as follows:
  • Contestable edit is made.
  • Contestable edit is reverted.
  • A user (doesn't have to be OP) restores the edit without consensus.
  • Contestable edit is reverted again.
  • Cycle continues.
Imo at the point the contested edit is restored, we have entered editwarring territory (even it hasn't reached "full scale" at this stage). If the contested edit doesn't break policy, discretion to uphold status quo should be considered, or left as is until resolved (as we have seen restoring quo can escalate things). If it breaks policy it has to go to quo, with additional protections if needed.
If one side then doesn't engage with the othet, a consensus of silence is considered made and the engaging side is considered to have the consensus in their favour. If neither side engages then the current state of the article wins. Sakaratte - Talk to the catmin 09:29, July 26, 2018 (UTC)
The image to the right is a screenshot of the Discord discussion that emerged after I posted a link to this forum topic. Click the image for a larger version, or click here for a transcript in the image's description.
Forum Ambiguous guidelines What is a revert Discord discussion

@Sakaratte I almost agree with your definitions of revert and edit warring: I think we should also consider the following situations:
  • Contestable edit is made
  • Contestable edit is reverted
  • Variation of contestable edit is made
  • Variation of contestible edit is reverted
In this case the contested edit is never restored, but it does look like a non-constructive edit war.
Additionally, I think we should scrap the whole status quo thing because it implies that there is a favourable state, even though edit wars occur exactly because no one can agree what the state should be. In my opinion, status quo is applicable only when there is vandalism going on.
- FDekker talk 12:05, July 26, 2018 (UTC)

( I would say that scenario should be handled in the exact same manner as the one I gave. After all, unless the changes resolve the original reversion reason, it is just a semantic argument.

As for status quo, I do feel we should keep some flexibility there, reverting a revert is only ok when it is vandalism is too strict; that is why I said policy based. As an admin/content moderator, when page lock consideration comes in to play, the merits of the arguments should be considered. If the edit has multiple flaws, (not just the information change, but other elements such as spelling, grammar, position in article, information lost) then restoration to quo should be applied. If the issue at hand is something that can live for a while, without any significant detriment, then I see no reason to enforce quo. Sakaratte - Talk to the catmin 13:15, July 26, 2018 (UTC)

Community content is available under CC-BY-SA unless otherwise noted.

Fandom may earn an affiliate commission on sales made from links on this page.

Stream the best stories.

Fandom may earn an affiliate commission on sales made from links on this page.

Get Disney+