Wikipedia talk:Closure requests

Point 1 at the top of the page currently states:

Do not list discussions where consensus is clear. If you feel the need to close them, do it yourself.

However, for someone who is WP:INVOLVED (especially the author of the RfC), closing the discussion is highly discouraged even if the consensus is clear. Closure requests by involved editors have been nonetheless been denied on this basis (e.g. this request by the RfC's author). should point 1 by amended to clarify this?

Do not list discussions where consensus is clear, except if you are involved. If you feel the need to close them, do it yourself.

Chaotic Enby (talk · contribs) 16:06, 19 December 2024 (UTC)

That would mean that every single closure would need to come here, even if the result is unanimous. Is that good? That said, refusing to do a requested close when the consensus is obvious seems profoundly unhelpful, and demoralizing to people who are trying to avoid even the appearance of impropriety. -- Beland (talk) 01:36, 18 September 2025 (UTC)

I would say that point 1 should not be amended. Only RfCs that actually need an external closure should be posted here. There was a time a few years ago where an editor used to mass-post unclosed discussions here, and CR was flooded as a result, delaying the closure of discussions that actually needed closing.

If consensus is clear and nobody is likely to dispute it, there is often no need for a formal closure at all, or the closure can be done by an involved party such as the editor who opened the discussion: see the guidance at WP:Closing discussions#Closure procedure, including the footnote at the end of the first sentence. Dionysodorus (talk) 10:01, 18 September 2025 (UTC)

The issue is that the consensus level that would justify an involved close is much higher than a non-involved close (the note you mention talks about in uncontentious circumstances and takes an example where the discussion is nearly unanimous). I'm also not talking about mass-posting any discussions, but specifically discussions where the poster is involved and the consensus isn't clear-cut enough for them to close it themselves, even though it would be acceptable for someone uninvolved to close it. Chaotic Enby (talk · contribs) 16:41, 18 September 2025 (UTC)

I think point 1 is only intended to exclude discussions where the outcome is so obvious that a close is not needed or that the participants involved in the discussion can legitimately close it for themselves. That is to say, if consensus is clear, the participants should either just proceed accordingly or should close the discussion for themselves; if consensus isn't sufficiently clear to do that, the participants should wait for an uninvolved closer, usually by asking for one here. In any given case, it is up to the participants whether they think the outcome of the discussion is clear or whether to post it here, and I don't think any closer would object to being asked to close any discussion that any participant in good faith considers even slightly too unclear for an involved close to take place.

If we rewrote point 1 as you are suggesting, that would imply that involved participants in a discussion should post listings here even where the outcome is absolutely clear (and where they could therefore close the discussion for themselves). That would be a bad thing in my view, because it could lead to discussions being listed here unnecessarily, which could contribute to long backlogs forming at CR (as in the case of the mass listings, but for a different reason), which would make it harder for discussions that really do need closing to be dealt with promptly.

Also, the current phrasing is perhaps helpful in discouraging the minority in an RfC from obstructively insisting on a formal close when the discussion has clearly gone against them. I think this sometimes happens even as things are, and in such cases a formal close is usually necessary (if only to prevent the losing side from further lawyering or edit-warring against the outcome of the RfC), but it is a waste of everyone's time, and editors who are on the losing side in an RfC ought to be discouraged from doing this as far as possible. There's no way of preventing one or two opponents of an otherwise agreed decision from insisting on a formal uninvolved close, but if we amended point 1 as you are suggesting it could potentially have the unintended effect of actually encouraging people to do that. Dionysodorus (talk) 17:12, 18 September 2025 (UTC)

Thanks! I had a broader reading of point 1, where consensus is clear would also cover non-WP:SNOW cases where a consensus is visible but an outside closure would still be ideal. I agree with your point of including that any participant in good faith considers even slightly too unclear for an involved close to take place, and wonder if that could be clarified in the text in any other way?

Regarding your last paragraph, wikilawyering is something that can always happen, and, if there is an involved close, disruptive editors on the minority side could also very much insist on undoing the close for that reason. Maybe that could also be clarified in the text here? Chaotic Enby (talk · contribs) 17:37, 18 September 2025 (UTC)

@Chaotic Enby: What if point 1 said something like this?

Do not list discussions where the consensus is obvious.

In discussions where consensus is entirely clear to everyone involved, there is no need for a formal close: just go ahead and implement the decision! Discussions should only be posted here when an uninvolved closer is actually needed to resolve the matter.

That might be clearer, since it would avoid suggesting to involved participants to "do it yourself", which might in some cases be misleading. Dionysodorus (talk) 18:05, 18 September 2025 (UTC)

That would work great for me! Chaotic Enby (talk · contribs) 19:03, 18 September 2025 (UTC)

FWIW, sounds good to me too. -- Beland (talk) 19:35, 18 September 2025 (UTC)

Hmm - perhaps ideally it would be good to have a slightly wider consensus for such a rewording; but maybe we could make the change in a few weeks' time if no-one has objected by then? Dionysodorus (talk) 11:28, 20 September 2025 (UTC)

I have boldly updated the text without prejudice to yet-unheard objections. -- Beland (talk) 22:29, 27 September 2025 (UTC)

Would there be support for changing the mechanism for indicating completion? It seems flipping the template to "done=yes" without putting {{done}} happens a lot, and vice-versa. We could have the bot archive in response to seeing "|done=yes}}" and have {{initiated}} display the word "done" in a celebratory color or something? -- Beland (talk) 01:34, 18 September 2025 (UTC)

@Beland: They have different purposes. The |done=yes parameter in {{initiated}} removes its colouring, so that it gets the default colour - black text on white, unless you are in dark mode. The yes value also deactivates Category:Administrative backlog. Absence of this parameter displays the text in blue, green or red, depending upon (a) how long ago the timestamp is and (b) the value of the |type= parameter. This coloured text is the attention-getter: blue means "somebody would like this to be acted upon but a few more days won't matter"; green means "this is now due for action"; red means "this is long overdue". Setting |done=yes essentially means "no further action required". The archiving bot is ClueBot III (talk · contribs) which ignores {{initiated}} entirely - it looks for the presence of specific templates, these are listed in the |archivenow= parameter of {{User:ClueBot III/ArchiveThis}}. If none of these templates is present, the thread will not be archived for 4368 hours (6 months). If you want ClueBot III to be amended, you need to ask its bot-ops. Past experience suggests that there is reluctance. --Redrose64 🌹 (talk) 18:01, 18 September 2025 (UTC)

Yes, my understanding aligns with your explanation. It looks like there's no need to modify the bot; we control its behavior with the "archivenow" parameter in {{User:ClueBot III/ArchiveThis}} on Wikipedia:Closure requests itself. I do find the color change useful. It sounds like not adding "done=yes" would be a problem for catgegorization if the snippet gets archived, so it makes sense to me that the bot should refuse to archive it until that parameter is set properly? -- Beland (talk) 19:33, 18 September 2025 (UTC)

Even having been reminded about setting done=yes, I still occasionally forget. Part of the problem is that you can't see {{initiated}} if you hit "Reply" instead of editing the whole section's wikitext. Since no one has objected, I have attempted to tweak the ClueBot III directive to look for "done=yes". I'm not entirely confident I've done it correctly, due to escaping, but will wait a few days to see if it's still working. -- Beland (talk) 22:25, 27 September 2025 (UTC)

Redrose64 It appears you were referring to this comment of yours, and similar comments:

we need the extra dummy subsections because of a bug in ClueBot III (talk · contribs), so instead, each new request is added as a last-but-one subsection. The dummy subsections are formatted as level 4 headings, the same level as the real requests, so that ClueBot III (talk · contribs) doesn't trash the following level 3 heading when archiving. See for example this archiving edit and this repair, dating from before we introduced those dummy subsections. There was a discussion on the bot talk page, which amounted to "this is how the bot works, get used to it". We tried hidden comments: they didn't work, as ClueBot III trashed the hidden comment too. After we made some more adjustments in May 2017, finishing with this edit, the archiving has worked smoothly.

I think you'll excuse me if I missed that; that discussion was 7 years old.

Do you think the bot would still work properly if the instructions weren't displayed in a hidden comment and we kept all the dummy headings, as you called them? I really think the instructions should be made more accessible to users of this page. FaviFake (talk) 14:58, 15 November 2025 (UTC)

@Redrose64 I have an idea; since the a header like ===Place new administrative discussions above this line using a level 3 heading=== is needed, we could use just a blank Unicode character instead of a header. This way, it'll mostly look like a blank line between h2 sections.

This would also have other benefits, as it would be less misleading. New discussions shouldn't be placed directly "above this line", they should actually be placed based on when they were initiated. I've seen this has also been brought up in past discussions as an issue. FaviFake (talk) 15:31, 17 November 2025 (UTC)

You're missing the point. Each request must be followed by a level 3 subheading because ClueBot III takes a block of text to archive as being everything from a heading of a given level to a point immediately before the next heading of the same level, regardless of the level of any intervening headings. So if a level 3 subsection is eligible for archiving, and happens to be followed by a level 2 heading, ClueBot III will suck up not just the level 3 subheading, but also that level 2 heading and any text that occurs between that and the next level 3 subheading. This has been covered several times on this page.

Chronological ordering is a nice-to-have, it's not essential. We order them chronologically so that people picking something to close will look at the older ones first. We ask people to place new requests above the line simply so that they don't put them above the next level 2 heading.

All these rules and instructions may seem petty, but they've been built up from years of experience observing problems as they occur, working out why something failed, and trying to devise some method that will prevent the issue from re-occurring. The present system has been both stable and reliable for some time now, so I don't see any reason to alter it. --Redrose64 🌹 (talk) 23:16, 17 November 2025 (UTC)

Yes! Sorry. I meant to suggest to still place the invisible character inside a h3 header like we do now, but I've not noticed my request was worded terribly.

So it sounds like these headers will remain as-is and the bot would also break if the instructions weren't displayed in a hidden comment. This situation sucks!

I had made a couple of other minor changes, like moving the archiving notice inside the archive box, and rewording the instructions for clarity. I suppose you're not against those if the instructions remain inside a hidden comment and the headers remain unchanged? FaviFake (talk) 05:19, 18 November 2025 (UTC)

Is there something wrong with ClueBot or the archiving config? It's been a week and still no arching. Noting some tweaks yesterday from TarnishedPath and Aidan9382 but still nothing archived before then nor after (and there's plenty to go). For reference, the last time ClueBot archived requests, the code updatenow param was {{User:ClueBot III/ArchiveNow}},done=yes, which is what it is now at least. CNC (talk) 11:35, 8 December 2025 (UTC)

No idea. I thought the issue was it being ,done=yes and so I adjusted it to |done=yes. Nothing happened and Aidan adjusted it back pretty much. TarnishedPath^talk 11:38, 8 December 2025 (UTC)

Being discussed on the talk page, didn't notice toptic already opened there. CNC (talk) 11:53, 8 December 2025 (UTC)

@Paine Ellsworth: Greetings! Regarding the edit summary note "gentle reminder to editor Beland to please sign your work": I am intentionally not doing so. I tweaked {{initiated}} so that it prints Done before items that are marked with "done=yes". I was tired of typing "done" three times every time I closed a conversation, once to mark done=yes, a second time to write a comment saying "Done!" and a third time in the edit summary. Now in addition to not being green or red, all the conversations that are done=yes are clearly marked with the word "Done" so it shouldn't be necessary to write a pro forma comment saying the same thing. -- Beland (talk) 08:33, 22 January 2026 (UTC)

To editor Beland: we agree that there is no need to be too "done redundant". What I ask is that you please remember to sign your work. Your sig includes a link to your talk page in case an editor has a question about your work. And although not as important as that, there are also editors like me who would like to see at a glance who closed a particular item without having to actually go to the discussion to see whodunnit. Not just here on this page, it would be good if you signed actual discussions you close, for example here, where I added your sig. So please, if it's not too much to ask, please let me and other editors know when you've closed a discussion, especially on this page (WP:CR), by including your signature to show your work. Thank you for your much appreciated help in closing these discussions! P.I. Ellsworth , ed. – welcome! – 09:42, 22 January 2026 (UTC)

PS. btw, I do like the "done" right at the beginning. It's always good to be right up front with a decision. P.I. Ellsworth , ed. – welcome! – 09:48, 22 January 2026 (UTC)

I already sign the close itself; anyone who has a question about the close should be looking at that page and not WP:CR. The person noting that the WP:CR request is satisfied is not necessarily the person who actually performed the close; sometimes I find that the discussion is already closed when I get there. If we are being paranoid that untrustworthy parties are closing discussions, we would actually want to check the pages where the close happens. -- Beland (talk) 10:27, 22 January 2026 (UTC)

As I said, we'd like to see at a glance who closed an entry rather than having to click a link and frequently have to scroll to find out who closed it. Life is short and there are a lot of other things we like to do, so why do you want to make it take longer for us to check these items? If you don't leave a signature, it's as if you're trying to hide something. or perhaps you're worried that someone might disagree with you? Why are you not proud enough of your work that you wouldn't leave your sig??? It's just 4 little tildes, after all. What's the big deal? P.I. Ellsworth , ed. – welcome! – 11:44, 22 January 2026 (UTC)

For the one-and-done cases (as opposed to where we're having a conversation about closing) I don't think it's useful to track on this page who did which closes. If there are people who can't be trusted to do closes without getting them reviewed but are in the habit of doing them anyway, then we should be just as interested in their closes that don't make it to this page.

No, I'm not boastful or prideful about my work. After reading through these heated discussions, I don't really have the emotional energy for that, and as uninvolved adjudicators, I actually think it's more appropriate for us to be humble. I'm not hiding anything, and resent the implication. My signature is plain as day on the page where I write the words that people disagree with. Given the disproportionate number of complaints and unwarranted appeals arising from closing these conversations (compared to the everyday ones that don't make it to this page), it seems people have few reservations about doing so. I haven't been doing closes because I'm looking for social approval or building karma (which isn't real, anyway). I've been doing them because 1.) I have made requests for closure, and I hope that doing the other tasks focuses attention and encourages cooperation on the discussions I can't close myself, and 2.) having unclosed discussions linger for weeks and months makes me anxious that we're wasting editor time and leaving some articles in a messy limbo.

Now I'm getting a complaint that I'm not double-signing and triple-done-ing every close. If we're going to build a bureaucracy like this on top of the already arduous close process, given everything else that's going on right now, I'd rather just not do them at all. So I'm going to take a break. Feel free to ping me if we get a streamlined process. -- Beland (talk) 18:36, 22 January 2026 (UTC)