Wikipedia talk:Edit warring/Archives/2017/February
 | This page is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
When should a series of reverts with intervening edits by other users count as edit-warring?
With the caveat that I am commenting from a theoretical point of view rather than seeing this in action, I would like to suggest a change to a sentence used in the "three revert rule", from:
A series of consecutive saved revert edits by one user with no intervening edits by another user counts as one revert.
to:
A series of consecutive saved revert edits by one user which do not revert any action performed in an intervening edit by another user counts as one revert.
I believe that this would avoid unfairly penalising somebody who reverts over a series of edits while others make intervening unrelated edits. Hopefully in this situation, administrators would apply some common sense in any case, but if the stated purpose is to have a "bright line" rule then it needs to be explicit. Admittedly such a change could introduce leniency in some cases of genuine edit warring, where intervening edits are not themselves reverted but are similar to the ones which are. But that sort of situation is hard to define objectively in any case, and the page is perfectly clear that the rule is clearly not intended to cover every conceivable situation.
--Money money tickle parsnip (talk) 18:04, 3 February 2017 (UTC)
- This would put an extra burden on admins without obvious benefit, in my opinion. The bright line is already bright, you are just proposing to draw the bright line in a different place that is a bit harder to calculate. Also, we hope that the edit warring rules are clear to the editor and anything that makes the rule more complicated creates a further burden on understanding. EdJohnston (talk) 18:16, 3 February 2017 (UTC)
- Yes but the position of the bright line matters. It has to be drawn on the correct side of the ideal boundary, and should not cross that boundary anywhere. Here's an example of where it could go wrong. Suppose that a British editor editing the Donald Trump article makes a number of helpful substantive additions, but in the same edits also changes a number of words to UK spelling in a good-faith but flawed attempt at proofreading. The edits cannot be undone without undoing the good work, so User:John Doe decides to go through and fix the spellings. He knows it's a busy article so he edits one section at a time per Help:Edit conflict#Prevention while other people make other edits. This all works fine from a technical perspective, but in so doing he violates the letter of the "three revert rule" because each edit is counted as a separate revert, even though all the UK spellings were inserted before John Doe started on any of it, and all the intervening edits were unrelated. It is not an edit war by any reasonable definition, so my point is, the line is currently drawn in a place where an administrator couldn't (reasonably) interpret the rule to the letter. Assuming that we are agreed that John Doe should not be blocked in this situation, in some sense the more nuanced rule I suggest is only codifying in writing the additional test that a reasonable administrator would have to apply anyway, but writing it down has the added benefit that it gives reassurance to John Doe that it is okay to do this. --Money money tickle parsnip (talk) 19:28, 3 February 2017 (UTC)
- If the changes to UK spellings were just due to misunderstanding, the reviewing admin would probably not count undoing those changes as reverts. What we usually count against the 3RR limit are 'changes that you know somebody else is opposed to'. EdJohnston (talk) 19:39, 3 February 2017 (UTC)
- Thanks. That helps, but not entirely. Tweak the example a bit to an original editor who inserts biased material mixed in with good additions, and I think the rest of my example still stands. --Money money tickle parsnip (talk) 19:45, 3 February 2017 (UTC)
- And indeed, according to the definition, it wouldn't even matter how long ago the original changes had been made. The unfortunate John Doe could in principle be held to be "edit warring" against an adversary who had long left the scene. --Money money tickle parsnip (talk) 19:54, 3 February 2017 (UTC)
- If the changes to UK spellings were just due to misunderstanding, the reviewing admin would probably not count undoing those changes as reverts. What we usually count against the 3RR limit are 'changes that you know somebody else is opposed to'. EdJohnston (talk) 19:39, 3 February 2017 (UTC)
- Yes but the position of the bright line matters. It has to be drawn on the correct side of the ideal boundary, and should not cross that boundary anywhere. Here's an example of where it could go wrong. Suppose that a British editor editing the Donald Trump article makes a number of helpful substantive additions, but in the same edits also changes a number of words to UK spelling in a good-faith but flawed attempt at proofreading. The edits cannot be undone without undoing the good work, so User:John Doe decides to go through and fix the spellings. He knows it's a busy article so he edits one section at a time per Help:Edit conflict#Prevention while other people make other edits. This all works fine from a technical perspective, but in so doing he violates the letter of the "three revert rule" because each edit is counted as a separate revert, even though all the UK spellings were inserted before John Doe started on any of it, and all the intervening edits were unrelated. It is not an edit war by any reasonable definition, so my point is, the line is currently drawn in a place where an administrator couldn't (reasonably) interpret the rule to the letter. Assuming that we are agreed that John Doe should not be blocked in this situation, in some sense the more nuanced rule I suggest is only codifying in writing the additional test that a reasonable administrator would have to apply anyway, but writing it down has the added benefit that it gives reassurance to John Doe that it is okay to do this. --Money money tickle parsnip (talk) 19:28, 3 February 2017 (UTC)