This page is part of the Wikipedia:WikiProject User warnings. This means that the WikiProject has identified it as part of the user warning system. The WikiProject itself is an attempt to standardise and improve user warnings, and conform them to technical guidelines. Your help is welcome, so feel free to join in.
When I use Twinkle to apply {{uw-editsummary}}, I get a nice box asking for a "Linked page". When I apply this template, uw-editsummary2, I do not get that box, but it would be helpful. I always have to go back to the editor's talk page and add a link to the page in question, so that they are not getting a blanket "you did something undesirable" message, which seems unhelpful to me, especially in light of Wikipedia:Don't template the regulars. How can that "Linked page" box be added to uw-editsummary2? Thanks. – Jonesey95 (talk) 21:42, 28 February 2025 (UTC)[reply]
@Jonesey95, the first step would be adding a page parameter here. You'd then need to make a request with the Twinkle maintainers.
I'm not sure how helpful that would be, though. Unlike {{uw-editsummary}}, which can reasonably be given out after noticing a single instance of a new editor not leaving a summary, this warning should really only be used if there is a pattern of behavior across many edits. Yes, I'll initially notice it in a given edit, but I'll check their recent contribution history or XTools summary before leaving it. And this template already currently automatically links to their XTools Edit Summary usage, so it shows them that there's evidence of a pattern of undesirable behavior. The particular place you first came across it doesn't really add anything to that. Sdkbtalk17:22, 1 March 2025 (UTC)[reply]
"Please do your best to always fill in the summary field." Srsly? That is weak, and hardly worth adding to a user talk page. Filling in edit summaries is not a "Do-your-best" thing, it is a policy requirement. This needs to be reworded more strongly, and include a policy link. Mathglot (talk) 08:36, 4 May 2025 (UTC)[reply]
If we put harsher language in this template, is there consensus in the community to enforce that language with sanctions? I can't remember seeing anyone blocked or formally censured for not providing edit summaries, even though it is in the policy language. It doesn't make sense to have a template that barks if the dog has no teeth, unfortunately. – Jonesey95 (talk) 13:23, 4 May 2025 (UTC)[reply]
Yeah, my understanding has been that the should in the policy is should in the "ideally ought to" sense rather than the "required to" sense. Whether that ought to be stronger is a discussion that should happen at the policy page first (the argument against is that we don't want it to become a sledgehammer to bring down on newcomers still finding their footing) before introducing any "must"-type language here.
Nobody is talking about bringing down a sledgehammer on newcomers for lack of edit summaries. We don't even do that to newcomers who ignore WP:Verifiability, one of our WP:core content policies, we just revert or draftify for quite some time before the now aging, former newcomer is blocked after umpteen talk page warnings and a visit to AN/I. There is no reason to think that a newcomer violating a non-core policy would be blocked more urgently than that.
Imho, part of the reason we haven't seen anyone blocked for lack of edit summaries is circular. That is, because we don't have multiple policy-linked warning templates that express the requirement with increasing severity, nobody is going to block an editor who has received two warnings ending with the milquetoast message found in this one; it would be wildly out of proportion. And there is no *-3 and *-4, currently, and composing a hand-written message takes time. And because of that, we think there is no will to block, but it is all circular: no messaging, can't bring to AN/I; no AN/I discussion, can't block; no blocks, don't need more severe messages/nobody cares. We should break the circle by adding *-3 and *-4 and including policy links in all of them. (I have started by adding a policy link to level 1.)
Frankly, I don't think most editors who place the two existing templates are even aware it is a policy requirement, and therefore do not go to AN/I to complain, believing it is pointless. I was squarely in that camp for a very long time, though I have used template and hand-written warnings of increasing severity on occasion, sometimes linking WP:DISRUPTION because of my ignorance of policy on summaries, and then eventually giving up in the face of editors clearly flouting the policy and even saying they had no inclination to change.
I think we need to decide if we do or don't want edit summaries to be policy, which as you say, should be discussed at the policy page, and either downgrade it to a guideline, remove it entirely, or enforce it. I am in the last group, and I will now be open to taking editors to AN/I who have a pattern of egregiously flouting the policy and ignoring warnings, and I encourage everyone else to do the same. Lack of summaries have a distinct negative effect on the productivity of other volunteer editors, and it is right and proper that the policy be enforced with blocks after sufficient warnings have been issued. Mathglot (talk) 17:44, 4 May 2025 (UTC)[reply]
I'm fully with you in agreeing how disruptive it is to others for experienced editors not to use summaries, and I'm open to policy reform. That said, I think you're underselling the potential risk a bit. So that you're prepared to address it, I'll spell it out a little more. The concern isn't that admins are going to start blocking newcomers because of a minor violation. Rather, it's that we already have issues with editors applying the letter rather than the spirit of the rules, and spelling out the letter of the expectation to use edit summaries more firmly might create an additional issue there. I could see overzealous patrollers taking a spelled-out rule as license to spam {{uw-editsummary}} to very new editors, who are already overwhelmed/at risk of getting bitten, and for whom a reminder to use summaries probably isn't the single most crucial thing they need to hear. Again, that's not to say I'm against a move to strengthen our policy in this area, but just that you should make a proposal that is prepared to address opposition in that vein.
Two additional ideas in this realm, one for new editors and one for experienced refuseniks. For newcomers, I think if they go to publish an edit without a summary, the interface should provide them with a nudge to write one. For experienced refuseniks, this template already recommends the "prompt me when entering a blank edit summary" setting; how about we try to create a sanction that'd force the option on, or (more stringently) force a summary for every edit? This would require some collaboration with the WMF to create the technical capability, but it'd be quite useful. If the checkbox sanction didn't work, we could escalate to the forced summary. And if they abused that by e.g. writing nonsense summaries, we could then escalate to a block. Sdkbtalk18:08, 4 May 2025 (UTC)[reply]
Thank you for this, these are really good points worth thinking about before moving to a central venue (likely WT:WARN) to propose something. As far as the over-zealous editors, I am a huge believer in quality template doc being able to solve or avoid a lot of problems, and a When to use section explaining this, with maybe a bolded key point in it aimed squarely at the over-z group could nip that in the bud, imho. (Quite a few templates even have a When-not-to-use section that can sometimes be effective, but is probably overkill in this case.) I love the idea of the newbie nudge (adorably alliterative, n'est-ce pas? Galloping glossary!) Would that be something we should ping the Growth Team about, or do you see that as more of a Preferences/Phabricator task for the m-w folks? (WhatamIdoing probably will know.) For the experienced editors, as you say we already have the "remind me if I forget to include an edit summary" opt-in preference, and I would worry that forcing it would increase frustration, and ultimately plant a seed for contempt and just result in gaming the system. (I have the box checked, and I sometimes add just a period at the end of an pre-loaded undo summary in order not to get the reminder when I hit Publish, when I judge more summary detail is not needed.) However, I wouldn't object to a "remind me twice" option, though not sure it would help much. But I would remain open as to what is the best way to deal with experienced refuseniks. (Would actually be good to hear from some of them, maybe via email or other private comm [Discord?] without prejudice to their current behavior, just to get some insight into what we are dealing with.)
To clarify one point: I would not propose any change to current policy; I think it is fine as it stands. I am mostly interested in promoting awareness that the current policy exists, and a set of tools in the form of multi-level warnings, so that good-faith editors who wish to issue such warnings will not give up for the lack of such tools being available ("Oh, they already got a {{uw-edit-summary}} and there is no level 2; nothing more to be done, darn it") or ignorance that it is policy ("Oh, they were templated before and flat-out refused; guess that's it then, darn it"). Awareness + multiple levels gives us the tools we need, without being a sledgehammer.
One other thing I would like to do before going to a central location to propose conversion of uw-editsummary to a multiple-level series, is to create drafts of the three "missing templates". Imho, it is always better to accompany a proposal with mockups to give editors willing to offer feedback something concrete to look at, rather than have each editor try to imagine what you meant. As we currently don't have a uw-editsummary-3 or *-4 , I think it would be okay just to create them directly in Template space (unless there is a good reason not to), as they wouldn't be linked from WP:WARN and only discussants would see them. (*-2 exists as an alt instead of a multi series level-2, so would have to be named differently temporarily; ultimately the current *-2 would move to *-experienced or similar, if the proposal were adopted). I would add a big red message on the /doc page of *-3 and *-4 so no one would be tempted to use it in User talk space, and could even add temp code disabling it in User talk space. So the proposed multi-levels would be in: {{uw-editsummary2-alt}}, {{uw-editsummary3}}, and {{uw-editsummary4}}. I see no occasion for needing an *-im version. These new mockups would be linked from the discussion, and open to adjustments in response to feedback in the conversation. Their status would remain as mockups unless the proposal is accepted, and would be deleted if it is not. Mathglot (talk) 19:13, 4 May 2025 (UTC)[reply]
Another thought (and I hope I didn't just burden you with an edit conflict!): for newbies, I agree we don't want to bite or overwhelm. I think it could be appropriate to add {{uw-summary-subtle}} to deal with that case. Again, tempolate wording can help: something about this added to the When-to-use section would be a good idea, along with a link to *-subtle if we decide to create it. Mathglot (talk) 19:28, 4 May 2025 (UTC)[reply]
I don't think forcing summaries would work either, because as I said above, it would just incite frustration, resentment, or gaming the system, so I don't disagree with you, but the pinging was not about that. Pinging flouters would be to solicit (preferably anonymized) data on why editors who are aware of policy on edit summaries choose not to provide them. That could be very valuable in designing template wording that might help avoid making the problem worse, by guiding new editors towards a path that would minimize the issues expressed by the flouters. I still think that could be valuable data to have. Flouters gonna flout; but newbies are malleable to an extent (we hope). I am not suggesting forcing summaries because I think it would be counterproductive. Mathglot (talk) 22:36, 4 May 2025 (UTC)[reply]
Mathglot, always using an edit summary is not a "policy requirement". The policy says All edits should be explained (unless the reason for them is obvious)—either by clear edit summaries, or by discussion on the associated talk page. That means that you "should" (not "must"):
Sure, I don't disagree. Points 1 and 3 are rarely the worry in cases of an editor being templated for not using edit summaries, and again, that can and should be covered in the template doc (likely in the When-to-use section). The issue is when they use them not at all or hardly. It's pretty clear when an editor is avoiding using summaries at all, you get runs of hundreds of mainspace edits with many significant changes among them and no edit summary. If 1 and 3 do not apply, then 2 does, and the template, per its title and content, addresses only case #2. If an editor always uses the Talk page for non-obvious edits, then they do not merit being templated, and the doc should make that clear. When case 2 applies and they constantly avoid it, then they have a clear pattern of violation of policy. Mathglot (talk) 22:26, 4 May 2025 (UTC)[reply]
I agree that upper-level warnings should be only for editors who are repeatedly failing to use edit summaries when they clearly should have been provided. "Should" does not mean "must", but we have guidelines and policies of that nature that are actually enforced. I welcome a centralized discussion about whether this policy, especially point 2 above, should be enforced, in what general circumstances, and how progressive warnings should be delivered. – Jonesey95 (talk) 02:16, 5 May 2025 (UTC)[reply]
When it's "an edit whose purpose is obvious", then the policy says that no edit summary is required. Looking at Special:RecentChanges in the mainspace for experienced editors only, I checked the most recent 10 edits. Most of them had an edit summary. Mobile editors were less likely to add an edit summary (two missing on mobile, one missing on desktop). In every case, it was obvious to me from the diff what they were doing, but why they were doing it in one case required reading the first sentence of the article, and in two other cases, an edit summary could have been helpful (though I suspect that one of those would have just said "Update"). In no case did I actually need to read the edit summary to figure out the purpose of the edit. However, in those two instances I think I would have figured it out faster (i.e., in three second instead of in ten seconds) if there had been an informative edit summary.
@Mathglot, could you, offhand, name five experienced editors who have made "runs of hundreds of mainspace edits with many significant changes among them and no edit summary"? Don't post their names; just tell me whether you know that many people. WhatamIdoing (talk) 01:21, 16 May 2025 (UTC)[reply]
I understand your question as a yes-no, and the answer is no, I couldn't. But then, I also couldn't name five that do. My memory isn't that good to begin with, or maybe—just a guess—it's selective about what kind of things might be useful to me in the future if I remember it, and this doesn't qualify. And there's even the issue of whether I am aware of whether someone really is experienced, or just hits my watchlist a lot because they have similar interests. But now that you have asked, I will try to take note as I do other things, and when I see them, I will try to take note and hopefully give you a better answer at some point. Mathglot (talk) 02:04, 16 May 2025 (UTC)[reply]
If we could identify a handful of experienced editors who routinely do this, then we might be able to figure out why they do this. If the pattern isn't obvious (e.g., they're all on mobile), then it'd be best if it was someone you know/who would recognize your name. (From a complete stranger, it might sound like an accusation instead of curiosity.) WhatamIdoing (talk) 02:31, 16 May 2025 (UTC)[reply]
I actually have been thinking about this and heading off in a slightly different direction ever since your previous message, with some parallels to your most recent one. It is slightly o/t for this page, so I will add something to my Talk page and ping you to it, because I can't think of a good venue just yet. Stay tuned.... Mathglot (talk) 02:49, 16 May 2025 (UTC)[reply]
Haven't gotten to developing my other topic yet, but was thinking about how to find such users and developed this query which finds 182 user Talk page with three (or more) edit summary templates. That is probably way over-restrictive, and doesn't select for experienced editors (or active ones) but is a start. Users near the top of the list include two still active users with many edit summary notices (templated or manual) and still few or no summaries: Special:Contributions/Aldis90, Special:Contributions/Marxsafe, both with many notices, still active, still no ES. These (and others in the search results) may give us some insight into why they do or don't respond to requests about edit summaries. MS (but not Aldis) uses the mobile web interface, which always makes me wonder if WP:THEYCANTHEARYOU is an issue.
