Wikipedia talk:Edit warring/Archives/2023/April

Proposed further exception in the definition of a "revert".

WP:3RRNO contains a list of eight "exemptions" -- undoes that are not considered reverts. I propose adding a ninth. My suggestion is

9. Removal of material that has been in a page for a long time is not considered a revert. What counts as 'a long time' varies depending on the situation, but anything that has been present for years, with no interruption at all, is safe.

My intent here is not to change the policy. It is to bring the written policy more in line with how it is actually enforced. Adoring nanny (talk) 20:44, 19 March 2023 (UTC)

I'm wondering whether this issue comes up enough in the 3RRNO situation to warrant adding it to the policy (see wp:KUDZU). Even if it does, what are the variables that play into determining what is a "long time"? Good luck getting consensus on that. - Butwhatdoiknow (talk) 22:42, 19 March 2023 (UTC)

The eight exemptions are examples of actions that are reverts but are exempt from the edit-warring policy. An item about something that is not a revert would be off-topic for the list. As far as I'm aware, the only place that gets into the nitty-gritty of what counts as a revert is the essay Wikipedia:Reverting. At one point, it comes close to addressing this issue, saying "Any edit to existing text could be said to reverse some of a previous edit. However, this is not the way the community defines reversion, because it is not consistent with either the principle of collaborative editing or with the editing policy."

I'd support efforts to polish off the essay, or split off the most important parts, and bump it up to a guideline. Firefangledfeathers (talk / contribs) 23:00, 19 March 2023 (UTC)

Is there any particular problem that this thread is trying to solve? I've not really noticed anyone trying to claim that removal of longstanding text is a "revert", and not sure it's a thing we néed to fix in particular. I'd be wary of putting too much of a formal definition on the thing per WP:CREEP. It's mostly obvious what a revert is, and when it isn't obvious, discussion and administrator/editor judgement can be used.  — Amakuru (talk) 23:31, 19 March 2023 (UTC)

I don't disagree that it's mostly obvious what a revert is. The edge cases come up often enough that I'd love to see clearer guidance. The "removal of longstanding content" question came up in a recent ANEW report and in a user talk page discussion, tied to a consensus required restriction, in which the murkiness around what a revert is led an admin to tell an experienced user they'd breached the restriction.

Given how much reverts are a part of our conduct policies and guidelines, I don't think that a clearer articulation of what they are would be harmful instruction creep. Newer editors frequently end up relying on WP:RV, and there are parts that are definitely more essay-like (WP:REVONLY). There will probably always be some fuzziness and a need for editor/admin judgment. Firefangledfeathers (talk / contribs) 04:57, 20 March 2023 (UTC)

I do not object to (what I take to be) the underlying idea behind this proposal. However, I do object to including a mention of this kind in 3RRNO, which is a list of policy-based reasons to revert. The place for some such clarification (that the initial removal of longstanding text is not considered a revert) would be elsewhere.

I also specifically object to the proposed language, as it could easily be interpteted as meaning that the removal of longstanding text is never a revert, even if editors remove the same text more than once within a 24-hour period. In fact, in the context of 3RRNO, this seems the most plausible interptetation of the proposed language even if the proposer intends otherwise. The second removal of longstanding text should count as a revert IMO, and this seems to be the default interpretation at present at 3RRN. Newimpartial (talk) 23:52, 19 March 2023 (UTC)

If you try to come up with a very specific definition I think that the effort will die under it's own weight. Instead, maybe list "factors to consider" something like the 4 I offered in the previous section. North8000 (talk) 15:11, 5 April 2023 (UTC)

Regarding "what problem does it solve?" I think that in reality common sense regarding whether it was actual edit warring is applied when any blocks are made. But rules that are prima facie impossible to comply with are not a good thing both for the credibility of the rules, and the stress of people trying to follow them. The definition of a revert if taken literally says that nearly every edit is a revert. So three non-controversial edits of old material on a non-controversial article in one day is technically a 3RR violation. I have seen issues with warriors trying to weaponize this policy, usually in conjunction with a 1RR restriction. In that case, the editor did one non-controversial edit to old material (which is usually technically a revert) and one revert that the warrior opposes. Then they say that the "revert" and the revert add up to two reverts and a 1RR violation. And taking that further, 0RR is basically saying to not edit. North8000 (talk) 20:51, 24 April 2023 (UTC)