Wikipedia talk:Requests for comment

From Wikipedia, the free encyclopedia

Ways to handle Bad RfC

An RfC was closed as bad on a page where I’m a contributor. I agree that the RfC needed to be reformatted. I’m wondering if there’s a more welcoming way to handle it. Here’s the close: Talk:2025 Potomac River mid-air collision#c-Rosbif73-20260218095600-RfC: Should it be changed from American Airlines Flight 5342 to American Eagle F Dw31415 (talk) 13:38, 18 February 2026 (UTC)

@Rosbif73, I agree with your assessment of the above RfC. There was a previous discussion about BadRfC’s here. Maybe this example helps find a consensus on how to handle. Dw31415 (talk) 13:42, 18 February 2026 (UTC)
I’ll look for the recent BadRfC discussion but I appreciate it if someone else links to it first. Dw31415 (talk) 13:56, 18 February 2026 (UTC)
  • I'm leaning towards reverting that close, which ought to have been a !vote. I'm almost, but not quite, certain enough to revert it summarily without discussion.—S Marshall T/C 14:21, 18 February 2026 (UTC)
    I say leave it closed because I coached the nominator on… wait a minute, the project RfC is still open. I’ll link to it here and on the talk page: Wikipedia talk:WikiProject Aviation#c-Zaptain United-20251101023400-RfC: Should the article title be styled as the IATA name, Branded name, or the I Dw31415 (talk) 15:01, 18 February 2026 (UTC)
    Then again, reverting would restore the FRS link. I didn’t revert because that felt like one aggressive response to another. I’d really appreciate some consensus on handling BadRfC’s, specifically, altering the question in the early days. Dw31415 (talk) 15:09, 18 February 2026 (UTC)
    I fixed the FRS link by adding ann anchor tag. Dw31415 (talk) 15:38, 18 February 2026 (UTC)
    My main rationale for the quick close was the rambling, non-neutral statement coupled with the apparent absence of any RFCBEFORE, and thus no apparent reason for holding an RfC rather than just a normal talk page discussion. It has now been pointed out that the issue had been raised previously. I didn't mean this to come across as an aggressive response, and I'd understand if the close is reverted. But I agree that consensus on how to handle "bad" RfCs would be helpful. It seems common that in cases where a poorly worded nom statement is amended after discussion has started, the whole process is often derailed, with meta-discussions about the statement mixed in with actual discussions about the issue at hand, and participants being unsure what they're !voting on. Rosbif73 (talk) 15:25, 18 February 2026 (UTC)
    @Rosbif73, please to go WP:RFCBEFORE and look for the little box in the sidebar. Find the words that say:
    "Are you trying to shut down someone else's RFC? "RFCBEFORE" is good advice, but it's not required."
    Having read that, do you now agree with me that "most importantly no WP:RFCBEFORE" isn't (ever) a valid reason to unilaterally shut down someone else's RFC? And what could we do to make that easier for experienced editors like you to discover this?
    The correct responses to an RFC such as that one are (in no particular order):
    • Add a brief, neutral RFC question to the top. For example, add "This article says 'American Airlines' throughout. Should we change that to 'American Eagle' instead? ~~~~~" (five tildes, so that nobody's name is in it).
    • Provide a sensible response (a "vote"). A clear reply can do a lot to reduce confusion.
    • Ask the OP if they'd like help improving or clarifying their RFC question (or if they'd like to use a different process).
    • Ask for help on this page.
    WhatamIdoing (talk) 18:01, 18 February 2026 (UTC)
    Having read WP:RFC more thoroughly, I entirely agree that my close was out of order, and apologise to @Zaptain United (the nom of the discussion in question) in particular.
    How could things be improved to make this easier to discover? Two quick comments: Firstly, the "are you trying to shut down..." sidebar isn't particularly visible, and the centralised discussion box pushes it down so that it isn't even opposite RFCBEFORE. Secondly, perhaps some advice about how not to close could be added to RFCEND, which seems a more logical place than RFCBEFORE to start looking for advice for someone considering closing an RFC. Rosbif73 (talk) 08:24, 19 February 2026 (UTC)
    Your gracious and reflective approach here is admirable. Respect.—S Marshall T/C 09:26, 19 February 2026 (UTC)
    +1 Dw31415 (talk) 12:43, 19 February 2026 (UTC)
    These are both great ideas. Maybe the box could get a little light blue background? Or an image that might catch someone's eye?
    I love the idea of putting something in RFCEND. I agree with you that this is actually the sensible place for people to look for that information. Do you want to add something yourself to WP:RFCEND? WhatamIdoing (talk) 22:57, 19 February 2026 (UTC)
    I've boldly added a {{tick}} and a {{n.b.}} to the side box. But I'm not yet confident that I understand the consensus surrounding early closes well enough to add something to WP:RFCEND. Would it be true to say that an early close of a bad RFC is always frowned upon, or are there cases where it would be recommended? For example, if an RFC proposed something that was a blatant violation of policy (say a suggestion to add information that would be a clear WP:BLPPRIVACY concern), would it be better to close the RFC preemptively, or just remove any concerns from the statement itself and then wait for it to SNOW? Rosbif73 (talk) 10:38, 20 February 2026 (UTC)
    That sidebox is against consensus (and worse yet, a bad idea), so drawing yet more attention to it is not an improvement. It should be removed.
    A minority thinks the RfC should be an unstoppable weapon, and there is no such thing as a bad RfC. This is incorrect. Polygnotus (talk) 10:58, 20 February 2026 (UTC)
    You disagree with the sidebox. However, you are not "consensus". WhatamIdoing (talk) 18:53, 20 February 2026 (UTC)
    Agreed. Neither of us is. That is my point. Polygnotus (talk) 19:00, 20 February 2026 (UTC)
    Here’s the last discussion: Wikipedia talk:Requests for comment#Bad RFC votes Dw31415 (talk) 15:51, 18 February 2026 (UTC)
    On my screen, that link doesn't work. On my screen it's at Wikipedia talk:Requests for comment/Archive 23#Bad RFC votes.—S Marshall T/C 20:02, 18 February 2026 (UTC)
  • OK. "Bad RfC" is an increasingly common response to a RfC question that the !voter doesn't want to proceed for various reasons, and we're increasingly seeing it appear in bold face. "Bad RfC -- question isn't neutral" or "Bad RfC -- see previous discussions X, Y, and Z" or "Bad RfC -- wrong place" all come into this category and they need different responses and get different weights.
    What we have for guidance on this is Wikipedia talk:Requests for comment/FAQ. I think it covers this situation.—S Marshall T/C 15:50, 18 February 2026 (UTC)
    Good point, Wikipedia:RFCRESPOND does explicitly say not to close a poorly worded RfC Dw31415 (talk) 15:57, 18 February 2026 (UTC)
    @S Marshall, the FAQ link above doesn’t work and I can’t find it. Would you please check the FAQ link? Dw31415 (talk) 16:09, 18 February 2026 (UTC)
    Well that's confusing. It's a valid link and it works fine for me. It's the template that appears near the top of this page.—S Marshall T/C 16:17, 18 February 2026 (UTC)
    Very strange! I get an empty talk page (Chrome on iOS) Dw31415 (talk) 16:20, 18 February 2026 (UTC)
    Oh, the FAQ is there hidden under “Learn more about this page”. Dw31415 (talk) 16:26, 18 February 2026 (UTC)
    All 'top matter' is hidden on the mobile site. WhatamIdoing (talk) 17:53, 18 February 2026 (UTC)
    Maybe some of the FAQ needs to find its way into the main page, especially since it's not so visible for editors using the mobile site. WhatamIdoing (talk) 18:02, 18 February 2026 (UTC)
  • Is there any way to deprecate the words “Bad RFC”? Sure, there are flawed RFCs (Non-neutral questions, overly rambling statements, etc) but it seems bitey to label any attempt to find consensus as “BAD” (naughty, naughty! Shame on you!). Blueboar (talk) 21:07, 18 February 2026 (UTC)
    This is the naming convention used everywhere else. Polygnotus (talk) 22:40, 18 February 2026 (UTC)
    We could make something like Template:Single-purpose account that can be used in opposition, e.g.:
    —See WP:RFCRESPOND and WP:RFCFAQ for advice on responding appropriately to RFCs that you believe are unnecessary, poorly framed, or otherwise improper.
    If there was relevant information at those two locations and the template got used fifty or a hundred times, I think people would get the message over time. WhatamIdoing (talk) 21:24, 18 February 2026 (UTC)
    The fundamental problem of BADRFCs cannot be solved by yet another obscure template, the problem can be solved by allowing people to close bad RfCs. Polygnotus (talk) 22:36, 18 February 2026 (UTC)
    Many reasonably contentious RfCs get !votes saying "Bad RfC", and they have different values and validities. For example:
    VoteActual objectionRecommended outcome
    Bad RfCQuestion isn't neutralUninvolved person to evaluate: if true, fix the question
    Bad RfCRepeat of recent, relevant previous discussionsUninvolved person to evaluate: if true, summarily close the RfC linking previous consensus
    Bad RfCWrong processUninvolved person to evaluate: if appropriate, move to RM, MR, DRV, PM, etc.
    Bad RfCPrematureUninvolved person to evaluate: if appropriate, convert to regular talk page discussion
    Bad RfCI want this to failUninvolved person to evaluate and offer support and guidance
    I wish we had more people clerking RfC.—S Marshall T/C 22:42, 18 February 2026 (UTC)
    @S Marshall I wish the tiny group of people who set the rules around here would stop trying to make the RfC the unstoppable weapon of mass destruction.
    Speculating that people who want to close an RfC are doing so in bad faith is a terrible look.
    Assuming that there is a huge number of "uninvolved" editors going around fixing bad RfC questions, closing and moving discussions is just incorrect. Polygnotus (talk) 22:46, 18 February 2026 (UTC)
    And you wrote if appropriate, convert to regular talk page discussion but RfCs are supposed to be regular talk page discussions with no special status. It is completely unworkable for me to go around closing lets say 60% of RfCs for being premature, but I wouldn't be surprised if 60% of RfCs are premature. Polygnotus (talk) 22:47, 18 February 2026 (UTC)
    I join issue with you on all that you say. I feel that summarily closing a RfC should be rare---although I have recently done it. Where the RfC is "bad", as it were, but we're dealing with a good faith user who thinks there's a problem to solve, the discussion should typically be delisted as a RfC with an explanatory note, but allowed to continue.
    I absolutely believe that there are people who use "Bad RfC" !votes as a spoiling tactic to prevent changes they oppose, and I think it's naive to pretend otherwise.
    RfC is Wikipedia's highest content dispute resolution process. It's an attempt to find consensus, and consensus is God Almighty.—S Marshall T/C 22:53, 18 February 2026 (UTC)
    @S Marshall Where the RfC is "bad", as it were, but we're dealing with a good faith user who thinks there's a problem to solve, the discussion should typically be delisted as a RfC with an explanatory note, but allowed to continue. I agree with that. In exceptional cases (e.g. trolls, IDHT) you can remove the template and close the discussion as well, but those are relatively rare compared to cases in which the template should just be removed and the discussion should continue.
    I absolutely believe that there are people who use "Bad RfC" !votes as a spoiling tactic to prevent changes they oppose, and I think it's naive to pretend otherwise. Maybe, but my point is not that they don't exist, my point is that its a bad look to assume bad faith in such cases. People disagree, you know. People have a different background, some people get the wrong end of the stick once in a while, some people consistently do and always say things that I believe are incorrect. But often they do it with the best of faith, which is very frustrating.
    RfC is Wikipedia's highest content dispute resolution process. No it isn't. A discussion is Wikipedia's only content dispute resolution process. The RfC tag is just a way to try to attract more people to a specific discussion.
    But since everyone thinks that the discussion they started is super important and needs immediate attention from a large group of people, we need to have a way to stop people wasting a fuckton of time. Wikipedia:Most ideas are bad and all that. Wasting everyone's time is a form of disruptive editing. Polygnotus (talk) 23:04, 18 February 2026 (UTC)
    We still get plenty of people - not all of them newbies - who launch an RfC as the first step in discussing a somewhat minor matter. It feels as if they believe that RfC is the only way of discussing a proposal. --Redrose64 🌹 (talk) 23:03, 18 February 2026 (UTC)
    For a low-traffic page, that's not a completely unreasonable belief. WhatamIdoing (talk) 01:10, 19 February 2026 (UTC)
    False, see WP:SEEKINPUT. Elevating RfCs over all other methods of getting more input leads to a lot of wasted time and effort. Polygnotus (talk) 01:11, 19 February 2026 (UTC)
    Another common failure mode for RFCs worth considering is when the question doesn't actually address the real problem on the page (I touched on one form of this in an essay at WP:MOTTE, but it's important to understand that it can happen very easily by accident just because the people on the page are talking past each other.) If the question for the RFC isn't well-formulated, all the time spent answering it could be wasted; and it sometimes won't be clear that there's a problem unless you're knee-deep in the dispute. Sometimes an ill-formatted question can be fixed by responses to the RFC along the lines of "this RFC asks X, but I think the real underlying question is Y, so I'm gonna answer that" (common for eg. questions where "is this a WP:RS" and "is this WP:DUE" are confused), but it can be a bit rough and make for a pretty confusing or hard-to-close RFC, especially if you end up with a bunch of people answering different questions. --Aquillion (talk) 06:29, 3 March 2026 (UTC)
    +1 for a template. I think it adds a little heft to reinforce the consensus view of how to respond to either the preemptive close or a more gentle nudge that the RFC question needs some attention. I'd be happy to join on working on it. Dw31415 (talk) 23:44, 18 February 2026 (UTC)
    Generally speaking, if the proposed solution to an alleged problem is concentrating yet more power in the hands of a tiny few, it is a terrible idea.
    Not just on Wikipedia, also in real life.
    Wikipedia talk:Requests for comment/FAQ is not supported by consensus and is mainly the work of 1 editor. Polygnotus (talk) 01:43, 19 February 2026 (UTC)
    I think a template would be better than one editor preemptively closing or another reverting the preemptive close, but open to ideas. I’d just like to see something change. Dw31415 (talk) 02:08, 19 February 2026 (UTC)
    Making a template available to everyone is the opposite of "concentrating yet more power in the hands of a tiny few".
    I have seen no evidence that the WP:RFCFAQ doesn't have consensus. I'm the person most likely to take the trouble to document the answers to common questions, but the answers are not just something I made up personally. WhatamIdoing (talk) 02:12, 19 February 2026 (UTC)
    Then you won't mind if I put my personal opinion somewhere and then post an official-looking template everywhere to link to it which heavily implies my way is correct and supported by consensus while others are not...
    But of course you would object to that, because that would feel unfair to you. So you doing the same feels unfair to me.
    I'm the person most likely to take the trouble to document the answers to common questions And I think that it is really important that someone does that job, and a lot of your edits are improvements, but we do end up in a situation where less than a handful of people decide important stuff for everyone else.
    And that wouldn't be a problem if those people never got the wrong end of the stick. But in this particular case, these attempts to make RfCs unstoppable weapons are misguided and they will exacerbate the problems we already have with RfCs (low quality, misleading, unfair advantage for the starter, presenting limited options leads to inferior decisions compared to a discussion et cetera et cetera).
    And since your position is in the minority, even with some of these advantages, it is probably not a good idea to make even more changes against consensus.
    Again, if you want to handle this fairly, lets combine forces and make an RfC that presents both our views side by side and then we can let the community decide. Polygnotus (talk) 12:04, 20 February 2026 (UTC)
    I think an RfC on preemptive closes would be a good thing and suggest we workshop it in a new section here. Dw31415 (talk) 13:22, 20 February 2026 (UTC)
    Sounds like a good idea. There are a few things it would be good to get wider input on. Polygnotus (talk) 13:35, 20 February 2026 (UTC)
    That's a tricky kind of discussion to have, especially with semi-random editors. Maybe just start with a regular discussion? It might help us understand how to frame it. WhatamIdoing (talk) 18:28, 20 February 2026 (UTC)
    Yeah we started in the section below with a regular discussion. Please do not start an RfC. Polygnotus (talk) 18:30, 20 February 2026 (UTC)
I'm against a "Bad RfC" template. If you don't happen to like me putting my opinion on an RfC, you're free to object rather than trying to limit my freedom to object. Peter Gulutzan (talk) 20:37, 23 February 2026 (UTC)
Are you for or against preemptive closes (shutting down an RfC)? Like Talk:2025 Potomac River mid-air collision#c-Rosbif73-20260218095600-RfC: Should it be changed from American Airlines Flight 5342 to American Eagle F Dw31415 (talk) 01:47, 24 February 2026 (UTC)
That doesn't contain "Bad RfC". I thought from the thread title and the suggestion about finding a way to "deprecate" saying those words and the "template" proposal that this was about shutting down people like me who use those words. Peter Gulutzan (talk) 16:43, 2 March 2026 (UTC)
I started this conversation in hopes of reducing the occurrence of editors “shutting down” or “preemptively” closing RfC’s. I’m personally less concerned with “BadRfC !votes”. I’ve been waiting to see if a challenge arises to the current project page text, before making a concrete template suggestion. Your message is a good prompt that maybe it’s time to stop waiting. Dw31415 (talk) 19:24, 2 March 2026 (UTC)
"Free to object"...so long as the editors who are freely objecting don't use a template to do so? It's not like the template has any special consequences or power over the person posting a "bad RFC" !vote. WhatamIdoing (talk) 06:03, 3 March 2026 (UTC)
  • After reading above, broke out a new level 2 section on this page called "Changing the RfC question". I immediately reverted because I think there should be discussion about it first. Please see . Please let me know if I should have just left the edit (i.e. if BOLD applies here too). Dw31415 (talk) 13:23, 19 February 2026 (UTC)
    Bold edits to this page are allowed. WhatamIdoing (talk) 23:16, 19 February 2026 (UTC)
In the RFCBEFORE section, would it be helpful to elaborate on "If you are not sure if an RfC is necessary, or about how best to frame it, ask on the talk page of this project"? Two possibilities:
  • "Consider discussing your planned RfC question on the talk page or on this page's talk page, to see whether other editors have ideas for making it clearer, more concise, or more neutral, or whether they think an RfC is necessary or the right process." Then for #1 in RFCOPEN, add "and that you've considered workshopping the question" to the existing text. I think it could help to bring this possibility up sooner.
  • Add some questions to ask oneself, drawing on some of the issues that S Marshall raised in his table above, perhaps including the next step as well. For example, "Is this issue a repeat of recent, relevant previous discussions? If so, another discussion may not be needed at this time." "Is an RfC the right process, or would it be more appropriate to use RM, MR, DRV, PM, etc.?" "Is it premature? Consider using a regular talk page discussion." Etc.
FactOrOpinion (talk) 22:02, 20 February 2026 (UTC)
The first point is currently made under WP:RFCNEUTRAL. One question is whether we want to scatter it all around the page (that might sound bad, but it's sometimes the right choice) or concentrate it in one spot.
The second point sounds a little WP:CREEPY for this page but sounds like a valuable expansion of Wikipedia:Writing requests for comment. WhatamIdoing (talk) 05:28, 21 February 2026 (UTC)
In this case, I think it's best to scatter a bit. My guess is that editors who read this page are most often inexperienced creating an RfC (or perhaps they want to check something, such as whether modifying someone else's RfC question is allowed). I think it would help to say more than once that if you're having difficulty writing a brief, clear, neutral RfC question, consider workshopping the question before opening the RfC, and you can ask for help or feedback on this talk page or on the article's talk page (or policy talk page, etc., if the RfC is about a policy, ...). That way, if it doesn't sink in the first time, or they miss the suggestion in one location, hopefully they'll see it in another location. FactOrOpinion (talk) 17:54, 23 February 2026 (UTC)
I wonder how many RFC questions are explicitly workshopped in advance. WhatamIdoing (talk) 01:49, 24 February 2026 (UTC)
  • Part of the issue is that once an RFC gets going, it is very hard to fix issues with it or to close it early (unless it's a WP:SNOW situation.) For this reason I do think we should extend the benefit of the doubt to people who immediately lunge onto an RFC and close it right after it was opened, provided they articulate clear problems they have with it and are clearly working towards having a better RFC opened in a reasonable timeframe rather than just stonewalling. Yes, sure, that RFC could in theory have continued, but I don't think that that was a particularly great opening statement (to put it mildly) and I think taking a day or so to hammer out something better is not unreasonable. If a fatally-flawed RFC continues, we risk wasting the entire time that it's open and all the time and energy editors put into it. And it won't always be obvious to people at a glance what the problem is (one common form of fatally-flawed RFC asks a question that seems reasonable at a glance but which doesn't actually cover or resolve the underlying dispute on the page, which means that it can go for a month, get a bunch of people weighing in, then solve nothing regardless of how it's closed.) I think it's fair to give everyone involved in a dispute a chance to weigh in, which includes broad latitude to say "wait hold up" for at least a brief period of time provided they're not stonewalling and they articulate at least somewhat reasonable objections that point towards a potentially improved RFC. --Aquillion (talk) 06:29, 3 March 2026 (UTC)
    That does sometimes work, and I've even seen it work when the person stopping the RFC advertisements is a declared opponent of the RFC's proposal. But it often fails, sometimes very badly, so I'd rather that the 'loyal opposition' normally took a hint from WP:INVOLVED and found someone else to shut down a poorly conceived RFC. WhatamIdoing (talk) 06:10, 4 March 2026 (UTC)

Potential hypothetical theoretical disagreements

There are a bunch of things that it would be good to get input on.

This is a work in progress/brainstorm. Please do not yet ask for more input. My brain is old and slow and it needs a couple of days to think.

I also need to go through the history of this page to see the various ways in which RfCs were elevated to unstoppable weapons (e.g. WP:BADRFC was deleted, WP:RFCBEFORE was neutered (for example here), a sidebox was added and there was even an proposal for a new template to truly make RfCs unstoppable).

  • Should one person just be allowed to change the rules for everyone without consensus?
  • Should RfCs be like !votes or should the RfC template just be another way to draw attention to a discussion
  • Should the person who starts the RfC get to choose the !voting options, if any, which makes it easy to present a false dichotomy
  • Should the RfC starter get to post a long non-neutral introduction
  • Should others be allowed to remove the RfC template when appropriate
  • Should people be forced to follow RFCBEFORE (before it got neutered)
  • Why not have an RfC where both sides present their view side by side

What else am I missing?

I actually had to make a tool that uses Claude to show how Wikipedia:Requests for comment evolved over time. Polygnotus (talk) 13:42, 20 February 2026 (UTC)

Far too many RfCs end up a mess, and at least some of it is down to the issues you raise here. As the situation currently stands, it is way too easy to construct a 'loaded' RfC, one that asks an ambiguous question, or one that clearly isn't going to solve an underlying issue. As to how to how best fix this, I'm unsure. If it wasn't yet another layer of bureaucracy, some sort of formal system for approving RfC questions in advance might possible help, though no doubt people would find ways to game that too. At minimum, I think we need to make it clearer to people starting RfCs that if they are going to ask for the broader community's time and attention, they need to put some serious thought into what they are doing first. And make it clear that failing to do so is likely to waste their time as well as everyone else's, since a messed-up RfC is unlikely to resolve anything. AndyTheGrump (talk) 14:19, 20 February 2026 (UTC)
@AndyTheGrump, I agree with you that these problems will always be with us. I think that part of our defense system has to rely on responding editors being smart and sensible.
You mention the importance of the underlying problems, and I think that's important to keep in mind when we talk about "bad" RFC questions. Sometimes the problem that needs solving isn't "What's the consensus?" but "I don't feel heard about ____". Accepting a loaded or biased RFC question may be the only way we can solve the underlying problem. Sometimes we have an editor who needs to be able to ask the community "Since WP:DAILYMAIL is obviously the most reliable source in the history of the universe, I propose that it be declared WP:GREL on Wikipedia:Reliable sources/Perennial sources", and then see what the community says, because a facially neutral question wouldn't make them feel like the community understood their view of The Daily Mail. Humans tend to believe that their view is the neutral, middle-of-the-road view, even if they're quite on the extreme end. WhatamIdoing (talk) 20:23, 20 February 2026 (UTC)
It is hard to believe that 50 people need to get messaged to get an answer to that question.
And the couterpoint is obviously that a truly dedicated hardhead baddie will just keep starting more and more RfCs and waste more and more of the only precious resource we have, the willingness of volunteers to help out. Polygnotus (talk) 20:29, 20 February 2026 (UTC)
50 people won't get messaged, so you can stop worrying about that. FRS never notifies more than 15, and rarely does even 10.
The problem of the occasional "truly dedicated hardhead baddie" is addressed in Wikipedia:Requests for comment#Multiple simultaneous RfCs on one page. We created this "rule" (it's not even a rule, but just a factual statement of what's statistically normal) years ago because of two editors, neither of whom are with us any longer, and it's been gently pointed out to a couple of editors since then, all of whom have immediately adjusted their behavior to conform. While the scenario you describe isn't impossible, based on existing experience, it is extremely unlikely, and even if it did happen, experience shows that we already have the tools we need to address it. WhatamIdoing (talk) 20:40, 20 February 2026 (UTC)
Well I have certainly wished for a way to stop someone with a Mohs 9-10 head from starting RfCs and there wasn't one (other than going to ANI, but their behaviour wasn't bad enough for that). They would laugh at a mere link to an information page; you'd need God himself to have a chance to convince them.
I'll read the Yapperbot code, thanks. Polygnotus (talk) 21:04, 20 February 2026 (UTC)
If they have more than one RFC open at a time, then put the links on this page, and let someone else have a chat with them. WhatamIdoing (talk) 21:08, 20 February 2026 (UTC)
In that case it was more a series of time-wasting RfCs IIRC. A huge timesink. Polygnotus (talk) 21:10, 20 February 2026 (UTC)
I suggest we stay away from:

Should one person just be allowed to change the rules for everyone without consensus?

Let’s assume good faith on those doing the hard work of trying to document the understanding of consensus and if the documentation has shifted from consensus, we, as a community, should take ownership for not having more discussion about it. I generally get the sense that the few editors being referred to are just out here doing the hard work and we owe them gratitude for that work. Dw31415 (talk) 16:13, 20 February 2026 (UTC)
Agreed, and I have also expressed my gratitude for that work. Especially because it is the kinda stuff I would find horribly boring.
Note also that I disagree with (or can find ways to improve) at least half my edits, so I'll happily admit my shit stinks.
I try to keep the "Main" slice of the XTools piechart as large as possible.
But we shouldn't end up in the situation that the rules for everyone are decided by less than a handful, and if that happens the way to counteract that is to point out that n=1 is worthless. And I think this topic is important enough to request far wider input (but not yet, still working on that, may take a few days). Polygnotus (talk) 16:31, 20 February 2026 (UTC)
One idea: Add (more) delay to when FRS sends it’s notification. On the above topic/RfC, I could have helped the nominator format the question because I follow the page. Instead I learned about the RfC through FRS and found another editor (from FRS) had preemptively closed because of lack of RFCBEFORE. That happened to be inaccurate because of 5 previous discussions on the RfC question. That said, the question had a major RFCNEUTRAL issue which could have easily been fixed. Adding a delay to FRS would give involved editors a chance to tidy up the question before it gets broadcast. Dw31415 (talk) 17:14, 20 February 2026 (UTC)
Or even have a special section on RFS where people can sign up who are willing to give feedback on future RfCs so that the bot can give those people a day or 3 to improve the RfC before it goes live. I think there are quite a few people who would be willing to help. Polygnotus (talk) 17:16, 20 February 2026 (UTC)
We've said for years that editors starting RFCs are invited to ask for help in writing the question on this talk page, and while we get more such requests than we used to, it's still on the order of 1% of RFCs. Most people who are starting RFCs don't believe that they need help. WhatamIdoing (talk) 20:06, 20 February 2026 (UTC)
Most people who are starting RFCs don't believe that they need help. I fully agree. So in this comment the workflow is: RfC created -> smaller group of people makes improvements -> RfC advertised to a larger group of people. This would sidestep the issue that people assume they know it all.
And sure, some people have a very complete understanding of the topic they want to start an RfC about. But many do not. Polygnotus (talk) 20:12, 20 February 2026 (UTC)
About an RfC where both sides present their view side by side: This is listed as an option in Wikipedia:Requests for comment/Example formatting#Pro and con. Very few RFCs have ever used that option. RFC questions are usually simpler than that (more "Shall we do this or not?" than "Shall we do this, with these three arguments in favor, or that, with those three arguments in favor?"). WhatamIdoing (talk) 20:05, 20 February 2026 (UTC)
Often the person who starts the RfC does not have a complete understanding of the topic (or, if you prefer a bad faith explanation, want to push people in a direction) which means they often do not ask the right question, do not provide the right context and do not offer the right choices.
The worst on that page is clearly "Separate votes from discussion" which sidesteps Wikipedia's consensus forming mechanism in favour of voting. Do you mind if I remove that one?
The second-worst is Separate support and oppose opinions because people will just form an opinion based on the original post and then post in their preferred section without reading the other comments. Do you mind if I remove that one? Polygnotus (talk) 20:27, 20 February 2026 (UTC)
Pushing in a particular direction is only a bad-faith behavior if the pushy person believes that direction will harm Wikipedia.
I suggest that you not remove any sections that describe styles that get used regularly and/or in highly publicized RFCs. You give the Zak Smith RFCs as an example below, and Talk:Zak Smith#RfC: Sexual abuse allegations (the second one) uses the "Separate votes from discussion" format. Large RFCs (also WP:RFA) such as Wikipedia:Requests for comment/Deployment of Vector (2022)#Support used the "Separate support and oppose opinions" format.
I think these should be used only for a minority of RFCs, but they should not be hidden or banned. WhatamIdoing (talk) 21:05, 20 February 2026 (UTC)
Pushing in a particular direction is only a bad-faith behavior if the pushy person believes that direction will harm Wikipedia. Intentionally (trying to) feed people misinformation in an attempt to influence them is pretty badfaithed in my book.
I can't imagine a scenario in which these formats would be superior to a freeform discussion (and I think that neither of the examples above benefit from these formats, compared to a normal discussion). Polygnotus (talk) 21:13, 20 February 2026 (UTC)
Pushing in a particular direction ≠ feeding people misinformation. I push people in the direction of using WP:MEDRS-compliant sources for Wikipedia:Biomedical information. That's "pushing people in a particular direction" but not "feeding people misinformation". Even if the people in question have Pathological demand avoidance and hate feeling "pushed", it's still not a bad-faith action or misinformation. WhatamIdoing (talk) 00:51, 21 February 2026 (UTC)
Agreed. But you can probably imagine a hypothetical situation in which an RfC opening statement attempts to misinform potential commenters/voters in order to push them in a particular direction. Polygnotus (talk) 00:58, 21 February 2026 (UTC)
In my experience, we have very few deliberate "attempts to misinform". We have many people who are attempting to properly inform – to the best of their understanding and ability, which often turns out to be poor understanding and limited ability. WhatamIdoing (talk) 01:14, 21 February 2026 (UTC)
Well, I gotta admit that it is impossible to read the minds of the various weirdos with internet connections (thank God!). But sometimes my spider-sense just tingles, you know? Polygnotus (talk) 01:16, 21 February 2026 (UTC)

Unstoppable weapons

Here's why RfCs are unstoppable weapons: they attract contributions from a broader range of editors. If you don't bring in people from outside, then you get the local WikiProject making shit decisions. We've found this out the hard way, several times, but for a really clear example, Wikipedia used to have a special notability guideline for porn performers at WP:PORNBIO. It was so ghastly inclusionist that for several years, Deletion Review refused to enforce it, so when porn performers' biographies got deleted at AfD, they stayed deleted regardless of the rules. We finally got rid of it here.

NSPORTS is also crazy inclusionist. From 2010 to 2022, it had a rule that if you'd ever in your life played professional sports for five minutes, you were notable. We binned that in WP:NSPORTS2022. But the 2010 version of the rule only passed because the pre-2010 version was even more crazy inclusionist. These are biographies of living people.

In other words RfCs have become unstoppable weapons because if we don't want self-selected groups dominating our decision-making processes, then we have to bring in people from outside the bubble.

If you don't want RfCs to be unstoppable weapons, then your proposal needs an alternative way to bring in the wider community to stop localconsensus from making shit decisions.—S Marshall T/C 17:42, 20 February 2026 (UTC)

I think we should delete a massive amount of BLPs that meet NSPORTS but not GNG. If we do not have enough reliable sources to write a somewhat decent article about that particular human being then why have an article? It doesn't make sense to me.
WP:LOCALCONSENSUS says: Consensus among a limited group of editors, at one place and time, cannot override community consensus on a wider scale. For instance, unless they can convince the broader community that such action is right, participants in a WikiProject cannot decide that some generally accepted policy or guideline does not apply to articles within its scope.
SmokeyJoe said it best: The SNGs are predictors of whether the topic will pass the GNG. The GNG is a predictor of whether the topic will pass AfD.
Luckily RfCs are not the only way to WP:SEEKINPUT but I agree that it is important that people can solicit external input, and I also agree that this fear is probably the underlying motivation for making RfCs unstoppable weapons of mass destruction.
But we probably shouldn't try to fix one problem by creating an even worse one. Polygnotus (talk) 17:57, 20 February 2026 (UTC)
We're not "making" RfCs unstoppable in the present tense. They were made, past tense, unstoppable over the course of years because we learned why they have to be.—S Marshall T/C 18:49, 20 February 2026 (UTC)
Nah, the push to make em unstoppable was pretty recent.
And they don't have to be. Polygnotus (talk) 18:51, 20 February 2026 (UTC)
I join issue with you on both of those claims.—S Marshall T/C 22:39, 20 February 2026 (UTC)
@S Marshall I actually have a tool that gets all revisions of the page, removes unnecessary fluff, filters out minor edits, diffs the revisions, figures out how many edits it can combine to stay under a preset token limit and then calls Claude to determine which were the most impactful changes.
It is pretty interesting to see how policy pages evolve over time. I may have to make a Wiki github. Polygnotus (talk) 22:50, 20 February 2026 (UTC)
Policy documents practice. By the time I was aware of policy pages, around 2009-ish, if you wanted to make a decision of any importance that was going to last and that wasn't shit, you held a RfC. And once you'd held a RfC, that decision was crystallised for a while and much harder for POV pushers to get rid of. In fact at that time RfC was broader in scope than it is now, because we could use it to discuss users as well as content decisions. RfC got cut back to broadly its current size when we deprecated RFC/U in 2014. The edits WAID made, which you seem to find so deeply objectionable, are actually just writing up how things work. You can certainly have a RfC about the RfC process if you want, and I'm starting to think that's easiest, because the community will set you straight.—S Marshall T/C 23:41, 20 February 2026 (UTC)
actually just writing up how things work Au contraire, mon frère. Knowing the Wikipedia community it seems rather unlikely that they are fine with these changes being made without being consulted. Even if they think they are a good idea then they still want to be able to voice their opinions.
Policy documents practice. In theory. No theory survives contact with reality. Just ask LTCM. Polygnotus (talk) 00:19, 21 February 2026 (UTC)
Ok, I genuinely tried to explain to you. These things you say are "changes" and you think the community would object to---you should go ahead and draw their attention to it. The community will set you straight.—S Marshall T/C 00:35, 21 February 2026 (UTC)
Thanks! Polygnotus (talk) 00:38, 21 February 2026 (UTC)
The "change" is that there was never a written rule allowing Poly to unilaterally shut down someone else's RFC over a perceived (but not actual) violation of RFCBEFORE (example), so they changed this page six minutes later to justify their action, and I eventually reverted their undiscussed change. If we're going to talk about "Should one person just be allowed to change the rules for everyone without consensus?", then I think that would make a reasonable example of someone changing the rules for everyone without consensus. WhatamIdoing (talk) 01:10, 21 February 2026 (UTC)
We already discussed that. And I explained how I could do the exact same if I wanted to, because it is easy to do such things. But is also counterproductive and not a nice thing to do. So its best if we don't. Polygnotus (talk) 01:15, 21 February 2026 (UTC)
About If we do not have enough reliable sources to write a somewhat decent article about that particular human being then why have an article?: What is the definition of "a somewhat decent article"? I believe that for the reader who wants to know what sport Alice Athlete played, then a single sentence saying something like "Alice Athlete was a long-distance runner in the 1976 Olympics" constitutes "a somewhat decent article". I believe that for the reader who wants to know a lot of information about Alice Athlete, this is not "a somewhat decent article". The result is that the community has to decide: Should the first reader not get an article that meets their limited needs, just because that limited article would not meet the second reader's more extensive needs? WhatamIdoing (talk) 18:53, 20 February 2026 (UTC)
What is the definition of "a somewhat decent article" My personal definition that is valid for only me? Or what the community accepts as a somewhat decent article? I am not sure I can answer either of these questions. Saying "it depends" feels a bit weak.
But it can be a bit worrying to see a BLP for which there is basically no source, except for perhaps a row in some database. For an astronomical object I find that less problematic. Polygnotus (talk) 18:58, 20 February 2026 (UTC)
Maybe an example is useful, because we all have a different experience. This is the kind of RfC I come across. Just one example out of many. Polygnotus (talk) 20:42, 20 February 2026 (UTC)
I'm not sure that I agree with @S Marshall that RFCs are an unstoppable force. This feels true in a collective sense, but the complaint seems to be about individual RFCs that an individual editor disapproves of.
I've shut down or re-written the question for individual RFCs that aren't helpful or functional, and I'm not the only one who has done that. In fact, I'm not even the most prolific of the RFC regulars to do that. So we have proof that this can be done, because people do that.
Rosbif73 asks above about problematic questions, so I'll give you some examples, and others can chime in with others (numbered to make it easier to disagree with me, of course). For clarity, I'm making a distinction between shutting down and closing: Shutting down happens earlier and is like getting your request rejected. Closing is what can happen if you've gotten useful responses and don't want/need any additional comments.
  1. The test edit. There's no question; there's no nothing. You should wait a bit to make sure that they're not in the middle of drafting the question, but if it's been more than half an hour, then you can assume it was a mistake and revert it.
  2. The l-o-n-g question. This is moderately common. First, the definition of an overly long question is one that looks a lot longer than (all/almost all) the others in Wikipedia:Requests for comment/All. A paragraph or two is okay, but 500 words is not. This RFC usually doesn't need to be shut down. Instead, figure out a short way to ask the question or describe the problem, and add that to the top with a ~~~~~ (date only) timestamp. The original version becomes the first "comment" rather than the RFC question itself.
  3. The formatting problem. The RFC system can't handle certain kinds of templates or formatting (e.g., tables) in the RFC question. Treat this like an overly long question and shorten the 'question' so that the RFC bot isn't trying to transclude the complex formatting.
  4. The nonsense question. We're talking word salad here, not just a question that's open to interpretation. Ordinary people have no idea what, if anything, is being asked, so none (or almost none) have responded. Pull the RFC tag and leave a note at the end of the section that invites the OP to discuss a clearer way to phrase their question here at WT:RFC.
  5. The hopelessly vague question. There's one of these open right now: What do you think of the article, and should we maybe try to improve it somehow? Editors dislike these and avoid answering them. It's not worth shutting down because the drama over shutting it down could exceed the cost of editors looking at the question and silently closing the tab. Sometimes these editors can be redirected to Wikipedia:Peer review. Otherwise, we mostly let these languish. If the OP complains about the lack of response, we can suggest starting over with a specific question.
  6. The lost cause question. This is the one where someone tries to argue that WP:DAILYMAIL is reliable for medical content. Editors generally see straight through this no matter how it's phrased. The most effective response is to leave the question alone and WP:PILEON with a clear rejection. If you want to be kind to the OP, or if there are already a dozen editors joining the fray, then you can point out to the OP that WP:RFCEND allows them to withdraw the question at any time.
    • NB that this kind of question isn't necessarily a waste of editors' time. Sure, there's zero chance of The Daily Mail being declared reliable for medical content. But there are also social benefits to giving editors an opportunity to publicly and collectively affirm their allegiance to the community's values and principles, and there are practical benefits to having some people learn that their POV is shared by a very small minority. Consequently, Wikipedia gets value out of this, even though it's not the face value of answering the nominal question.
  7. The slightly biased question. Maybe the question says 'many sources' but you think 'some' would be more accurate. Maybe it asks you to choose between picture A and B, and you think there should be no picture. The first line of response should be providing a sensible answer, which may include addressing your concern about the question in your response ("The question refers to 'the victim', which implies that he's actually a victim, which is exactly what's in dispute, but we all know what the OP is asking anyway"). Otherwise, this should be ignored because the cost of perfecting the question is higher than the cost of having a slightly imperfect RFC question.
  8. The obviously biased question. This is the one in which there's a legitimate dispute that could be solved by an RFC, and the question is written to support one answer. This is "Shall we include this WP:DUE material that is supported by the WP:BESTSOURCES?" or "Wikipedia is WP:NOTCENSORED, so we should put this contentious material in the lead." In an ideal world, it wouldn't be possible to look at the RFC question and tell what the OP's view is. However, as a practical matter, discovering the OP's view in the 'question' versus discovering the OP's view two centimeters lower on the page in the OP's 'response' does not seem to make much difference. I'm participating in one of these at the moment; last I checked, the discussion is five to one against the editor who cited a WP:GUNREL source in an WP:ARBPIA dispute. My conclusion is that the community is capable of handling this kind of biased question. I decided to participate this time; other times, I might have handled it like an overly long question (especially since that question is also long), or talked to the OP about the advantages of re-writing the question.
    • One thing to look out for here is whether there are voting-style responses that could get screwed up if you change the question. You need to avoid changing "Let's include this important material!" to a "Shall we exclude this?" question, because then the early votes are backwards.
  9. The subtlely biased question. This is the difficult one. This is usually in the 'question' and sometimes in pre-supplied voting options (yes/no, when it should be an open-ended discussion, or 1/2/3 when it should be 0 to 10). My preferred approach to this is discussion of the question in the RFC itself. My preferred approach is probably more effective if it's the first/early response. If these run their course, they may not settle the dispute. A subsequent RFC, with the question crafted based on what we learned from how this one failed, may be able to resolve it. (Of course, some disputes will never really be settled, no matter what we do.)
  10. The background section or the first response. This is officially Not a Problem, as only the RFC question itself (the bit that appears on Wikipedia:Requests for comment/All) "should" be brief and neutral (NB "should", not "must"), and responses are actually meant to not be neutral. That said, it generates some complaints, usually from people whose POV is not the most popular one ("How dare he explain why his view is correct before I could explain why my view is correct?!").
    • The reality behind this, which is that WP:TLDR is one of the iron laws of the internet, so people will read a few things but not everything in a long discussion, is why I think that when subsections are wanted, the ===Discussion=== section should precede the ===Survey=== one.
  11. The redundant question. Didn't we just do this last month? Sometimes these should just be shut down, especially if you think that the OP didn't know about the previous one. Give the OP a link to the prior one and pull the rfc tag. But if last month's result was weak, has been overtaken by events, or otherwise might not represent the current consensus, it's best to link to the prior discussion(s) and let this run again. If the problem is, as Andy describes above, that the RFCs aren't solving the underlying issue, then it can be helpful to facilitate a discussion with a goal of pointing people towards that underlying problem ("That's a good point, and how do you think it relates to the broader question of...?"). This category also overlaps with the lost cause: Yes, the community said 'no' last month, but WP:I didn't hear that and I need to be told again. If the result of the prior RFC is still valid, then reminding participants about the prior RFC will usually result in people voting the same way as the prior RFC, except louder (all the people who voted X last time, plus all the people who are voting X now just because they think the result of the prior RFC should be respected).
I also add some myths that are sometimes given by people who disapprove of a particular RFC question:
  • There was no prior discussion on this subject.
  • There was no prior discussion of the wording of the RFC question.
  • There was no prior discussion of whether to have an RFC.
What else would we add? WhatamIdoing (talk) 00:22, 21 February 2026 (UTC)
The wrong question. e.g. the dispute is not about W vs X but about underlying factors Y vs Z. Polygnotus (talk) 00:26, 21 February 2026 (UTC)
we should avoid providing pre-set answers That would be a great addition to the information page. Please add it. Polygnotus (talk) 01:23, 21 February 2026 (UTC)
I like preset answer options where appropriate. As in the Flight 5342 question, there’s three options raised in previous discussion. (American Airlines, American Eagle, PSA Airlines). What’s the problem with predefined options? Dw31415 (talk) 02:25, 21 February 2026 (UTC)
Two common problems:
  • When they're not necessary, they encourage and normalize voting behavior. As an example of "not necessary", if the RFC question is a simple yes/no question, you don't need to tell editors that the options are Yes and No.
  • When they're badly written, they can bias the discussion. As an example, if the RFC question should offer "0 to 10", and the preset answer options are "1 to 3", then it implies that neither "0" nor "4 to 10" are appropriate responses.
However, there are also benefits. Sometimes the preset options are the only way to figure out what the OP is actually asking. Sometimes the way the question is worded, you end up with "Yes, omit" or "No, support", and if people make up their own summaries, we end up with confusion. WhatamIdoing (talk) 04:33, 21 February 2026 (UTC)
Also the Anchoring effect, False dilemma/false exhaustiveness, framing, the Compromise effect and all that good stuff.
I have a book around here somewhere that lists like 10 more. Polygnotus (talk) 05:10, 21 February 2026 (UTC)
Another: The wrong process. Some are explicitly discouraged; like where the RfC statement basically asks "Should the article be deleted/merged/split/renamed". Others are not explicitly forbidden, but blatantly obvious; like this one. --Redrose64 🌹 (talk) 12:34, 21 February 2026 (UTC)
+1 for discussing that category. I just added a Template:Side box to the example you gave. I’d generally favor a template like this for a “softer, kinder, gentler” shut down of a “BadRfC”. Maybe a new template could include the RfC id so it renders the anchor tag and doesn’t break all the FRS links. I’m holding off proposing template solutions to see if @Polygnotus is going to test consensus of the RfC shutdown (and I like the term shut down for the preemptive close, especially because shutdown can refer to removing the RfC tag more than adding Template:Closed rfc top). Dw31415 (talk) 09:53, 22 February 2026 (UTC)
@Dw31415 and WhatamIdoing: Well I think the problems are broader than just the inability to shut down and change an RfC.
We want people to be able to attract the attention of uninvolved people. I think we all agree on that.
But from there, thing diverge.
We now have a few very oldskool ways that evolved over time and use terrible workarounds for Wikipedia's limitations.
But when you think about it, why do we do it this way? Is the way things evolved ideal? Then why do men have nipples, and not heated cupholders?
What are the differences between the various ways to WP:SEEKINPUT?
What if we had an editfilter that tagged an edit "RfC started" or "RfC closed"?
What if we had a bit of JS that follows the EventStream like ExternalLinkMonitor and tells people (who opted in) "yo there is a new RfC".
Why are we promoting using RfCs instead of telling people something like: "In most cases you can just use 3O. If that doesn't work, and you did RFCBEFORE and it still doesn't work, then maybe start an RfC"? Why use an unstoppable superweapon to kill a mosquito (Confucius 551-478 BCE)?
If we look at what the RfC starter wants, sometimes a broad consensus is indeed required (policy changes, whatever) but often just 1 or 2 people who show up quickly and agree with them is fine. And RfC starters don't even read this entire page full of instructions.
So why don't we flip it on its head, make a system where you can quickly attract 1 or 2 people to a potential problem, and then if you need more maybe you can start an RfC. Maybe. Polygnotus (talk) 10:41, 22 February 2026 (UTC)
Interesting. So maybe like a two-step process. Or an escalating one. Or maybe like in Bob’s rules where an RfC (a motion) requires a second before proceeding to debate.
Overall I’m willing to accept that the current system is the worst except for all the others that have been tried, but I have a lot of fun thinking about different governance ideas. Dw31415 (talk) 11:04, 22 February 2026 (UTC)
Another thing is that currently the RfC starter has to provide a topic area which could be replaced by a bit of code (and you can make it as granular as you want).
E.g. you start an RfC on a page about a writer, it shouldn't be too hard to automatically classify that as something that WikiProject Literature may care about.
For example see Category:Pages within the scope of WikiProject Astronomical objects (WP Astronomy Banner). Polygnotus (talk) 11:11, 22 February 2026 (UTC)
The difficulty with requiring a Second (parliamentary procedure) is that RFCs also function as signal flares for problems that we need to hear about. Imagine being stuck in an intractable dispute with two other people on a low-traffic page. None of your edits stick, and all of theirs do. There are three of you, so Third opinion won't work. Both refuse to 'second' your RFC. Everyone's being polite, so ANI won't help. Now what?
Well, you could go ask a friend, assuming you have any. You could go ask a WikiProject, assuming you know about them and assuming they'll respond (unfortunately unlikely). You could go to a noticeboard, which works best if you have a single point of disagreement. You could go to Wikipedia:Dispute resolution noticeboard, but they can refuse. The answer is: Let people start RFCs when they think they need them. WhatamIdoing (talk) 02:47, 23 February 2026 (UTC)
There are three of you, so Third opinion won't work.
So why must it be:
  • 1 person (who presumably agrees with themselves, unlike me)
  • 2 persons -> 3O
  • 3 persons -> RfC
Instead of Third Opinion why not have a similar "We'd like some more input please" option that is not limited to 3? I don't believe that the solution is not to make 3O basically unused and to tell everyone to start RfCs all the time for every minor thing. If the message is just "I'd like some more input please" then why have 2 options at all? And if we do have two ways of getting extra attention, why are they so functionally similar? Polygnotus (talk) 03:01, 23 February 2026 (UTC)
People may leave a neutrally-worded notice at the talk page of a WikiProject, directing them to the discussion (not necessarily an RfC). Templates such as {{fyi}} and {{subst:WikiProject please see}} are available for this. --Redrose64 🌹 (talk) 07:05, 23 February 2026 (UTC)
Problem is, most WikiProjects are dead. Polygnotus (talk) 07:06, 23 February 2026 (UTC)
People don't start RFCs all the time, much less for every minor thing. The median number of RFCs started by experienced editors is zero per lifetime. WhatamIdoing (talk) 07:10, 23 February 2026 (UTC)
Maybe some numbers would help. These days, people post about 9,000 comments on talk pages per day. They're starting about three (3) RFCs per day this month. If people were starting RFCs "all the time" or "for every minor thing", we'd get a lot more than one RFC per 3,000 (three thousand!) talk page comments. WhatamIdoing (talk) 07:18, 23 February 2026 (UTC)
@WhatamIdoing
Sure, but if you look at RfCs, do you always think "oh this is a good use of an RfC" or do you often think "hmm, I wonder why this person chose to use an RfC when that was unnecessary"?
I want RfCs to be useful for things like policy changes, that actually require a decently attended discussion.
Anyway I am answering you over here and that is actually a more interesting topic. Polygnotus (talk) 07:21, 23 February 2026 (UTC)
When I look at RFCs, I am more concerned with whether the discussion is likely to lead to a durable consensus than whether it feels "necessary" to me. RFCs do not have to be carefully rationed. It doesn't matter if we have 22 RFCs open this week vs 23 or even 25. WhatamIdoing (talk) 01:55, 24 February 2026 (UTC)
But the fact that RfCs lead to what you describe as a "durable consensus" is actually one of the major problems with RfCs.
Lets say X starts an RfC saying "Should political party Y be described as radical left, left, center, right, radical right".
See how they sneakily turned what should've been a discussion about reliable sources into a discussion about the personal feelings of the participants in the RfC? There are at least 2 such RfCs going on right now.
And then of course people who don't notice such things will voice their opinions.
And then person Z comes along and they have the perfect descriptor and 3 high quality academic sources to back that descriptor up.
But the second Z edits the article he will be reverted, because a headcount of uninformed opinions is more "durable" than high quality scholarly work.
Wikipedia is one of the very few places in which a tiny group (usually 1, sometimes 2 or 3) of good Wikipedians armed with reliable sources and common sense can defend against a horde of POV pushers.
Making RfCs unstoppable weapons that get used all the time makes life impossible for anyone defending against a flood of CPUSHers. Wikipedians are herd animals, and if the defender is offline for 2 weeks here are 3 new consensuses they will just have to accept (because restarting a new RfC after its been closed is bad form, and so is starting one about the same topic as a recently closed one).
What you call a "durable consensus" is antithetical to the "wiki" philosophy in which nothing is static and people are allowed to fix problems. What we need is iteration and improvement, not problems that cannot be fixed because some person managed to find or create Wikipedia accounts that agree with them. !Voting is the enemy, except in incredibly rare circumstances, otherwise those with the most sock- and meatpuppets will "win". Polygnotus (talk) 05:48, 24 February 2026 (UTC)
This comes back to the idea that RfC’s are really bad but better than all the governance methods available to a project like this. Are you closer to proposing any specific change? I think the change most worth discussing is some kind of draft stage before FRS does its magic. If you feel strongly that experienced editors should be able to shut down an RfC than I suggest you test that first. Any ideas on where we go from here? Dw31415 (talk) 13:50, 24 February 2026 (UTC)
@Dw31415 Well I would like to hear from @WhatamIdoing:. I am specifically interested in whatever common ground we may have.
In my view the status quo ante is that good-faith people are allowed to close bad RfCs, that good-faith people are allowed to edit RfC questions, that people can not unilaterally change the rules and that people should follow RFCBEFORE. So what we should do is edit the information page to make it reflect reality and then if people want to change how things work they can start a discussion here or elsewhere.
Using the appropriate tool for the job is easy: you start small and scale up when it doesn't work.
It would be interesting to have a wide(r) discussion about the direction the alleged community would like to go in, and if needed then something like an RfC to nail down the details and make it official. Polygnotus (talk) 14:22, 24 February 2026 (UTC)
This 2010 conversion Wikipedia talk:Requests for comment/Biographies of living people/Archive 1#Compromising on WP:BEFORE? is the only one I could find on mandatory prerequisites. I’d be interested to read others. Dw31415 (talk) 16:06, 24 February 2026 (UTC)
@Dw31415 I learned from WAID that 'should' as in people should follow RFCBEFORE should must be interpreted as described in RFC 2119.
But when I use the word 'should' its basically guesswork what I mean.
So according to 2119 it means:
3. SHOULD This word, or the adjective "RECOMMENDED", mean that there may exist valid reasons in particular circumstances to ignore a particular item, but the full implications must be understood and carefully weighed before choosing a different course.
So under that definition the fact that people should follow RFCBEFORE does not mean that people, you know, should follow RFCBEFORE. Polygnotus (talk) 16:24, 24 February 2026 (UTC)
When we say people should follow RFCBEFORE, I'm not sure that we are talking about the same thing.
When I say people should follow RFCBEFORE, I mean something like "You should follow RFCBEFORE, which says not to use RFCs to propose a WP:MERGE".
When you say people should follow RFCBEFORE, I have the impression that you mean something like "You shouldn't start an RFC at all unless it's utterly unavoidable or mandated by another policy like WP:PROPOSAL, because posting a comment on the talk page is almost always going to be good enough, and if it isn't, then it would be better for you to try any form of Wikipedia:Dispute resolution except an RFC, because we need to minimize RFCs as much as possible". WhatamIdoing (talk) 18:31, 24 February 2026 (UTC)
I think that most of what you've said is wrong, and based on unreasonable opinions.
  • a "durable consensus" is actually one of the major problems with RfCs – The alternatives are unresolved disputes (the antithesis of WP:Dispute resolution, and so failing at the whole purpose of RFCs).  NB that "durable consensus" does not mean "permanent decision that can never change".  It means something much closer to "at least they'll stop edit-warring over this for a while".  
  • an RfC saying "Should political party Y be described as radical left, left, center, right, radical right". See how they sneakily turned what should've been a discussion about reliable sources into a discussion about the personal feelings of the participants in the RfC? – Nope.  This RFC is not asking for personal feelings.  The bit about "In the context of all our policies, especially verifiability and neutrality, instead of your personal feelings" is unstated because it's so obvious.  
  • Making RfCs unstoppable weapons that get used all the time – They aren't unstoppable, they aren't weapons, and they don't get used all the time. As an example of an unreasonable, logic-defying opinion, consider this claim that they get used all the time, right after you've been told that as a strictly factual matter, only 0.03% of comments start an RFC, and 99.97% don't.
  • makes life impossible for anyone defending against a flood of CPUSHers – RFCs are a favorite tool of those defenders, so it's not reasonable to say that they make life impossible for them.
  • restarting a new RfC after its been closed is bad form, and so is starting one about the same topic as a recently closed one – Sometimes, but not always, and the regulars on this page support repeats when the previous result does not seem to have reached a durable consensus.
  • !Voting is the enemy – I agree.  Fortunately, RFCs are not votes.
WhatamIdoing (talk) 18:26, 24 February 2026 (UTC)
Fortunately, RFCs are not votes. Agreed, but the problem is that people act as if they are. Polygnotus (talk) 02:03, 25 February 2026 (UTC)
I mean, RFCs usually have to be closed (and an RFC that lacked a formal closure doesn't have much force behind it.) If someone started that political-party RFC you mentioned, most people didn't cite any sources or policy-based arguments, and one person dropped a huge wall of high-quality sources that unambiguously supported one position, any reasonable closer would give more weight to the position with all the sources backing it up. But the flipside of your objections is that sources themselves aren't magically self-enforcing or self-evaluating; it's rare for RFCs to be that lopsided. More commonly we have people with competing lists of sources of differing degrees of quality or comprehensiveness or which word things in various different ways, and while it's great to talk over these things and dig for more sources, we do eventually have to put our pens down at least temporarily, because the article does have to say something (and even declining to say something about the topic is its own decision.) RFCs are necessary as a way to allow us to reach that point; and as someone who has edited a lot in lots of controversial topic areas I'd say they usually serve that purpose. They don't always reach the outcome I want specifically, but it is very, very rare for an RFC to reach an outcome that I'd consider totally and completely unsupportable, so to speak. Like, you talk about how Wikipedia is a place where a tiny group (usually 1, sometimes 2 or 3) of good Wikipedians armed with reliable sources and common sense can defend against a horde of POV pushers - but in my experience the way they do that, at least on more obscure articles that are facing a flood of new inexperienced users, is usually to fire off an RFC like the Bat-Signal, which usually attracts more experienced outside opinions and an even more experienced closer who can assess all their sources and common sense and policy-based arguments and reach a conclusion that is, if not always perfect, at least reasonable. --Aquillion (talk) 06:45, 3 March 2026 (UTC)
This is mostly wrong: RFCs usually have to be closed (and an RFC that lacked a formal closure doesn't have much force behind it.)
A recent spot-check showed that two-thirds of RFCs got closing summaries. I'd say that roughly one-third needed a summary because the result was not obvious, another third got a summary (despite the result being obvious) because it was a CTOP or other drama-prone situation, and the remaining third needed no summary, received none, and didn't seem to have any problems as a result. (If everyone cheerfully agrees to ____, then you don't really need someone to officially summarize it; folks just implement it and go their merry way.) If you mostly work in drama-prone areas, then you probably don't see the ones in that last category, but it does exist. WhatamIdoing (talk) 06:06, 4 March 2026 (UTC)
@Dw31415 We now have an editfilter and an RfC tag. See User:Polygnotus/Scripts/RfCs.js. Polygnotus (talk) 00:33, 5 March 2026 (UTC)
That’s interesting. I’ll take a look. Dw31415 (talk) 03:11, 5 March 2026 (UTC)
@Dw31415 Newer version at User:Polygnotus/Scripts/RfCMonitor.js. Polygnotus (talk) 03:13, 5 March 2026 (UTC)
You can also see the tags in RecentChanges now. WhatamIdoing (talk) 03:30, 5 March 2026 (UTC)
@WhatamIdoing I assume the intended link target was something like this, where you can see me polluting logs by being an idiot Polygnotus (talk) 04:19, 5 March 2026 (UTC)

"Invalid RfC/Biased question"

An editor who is involved with the Magan content dispute is claiming that an RfC I started (Talk:Feeding Our Future § RFC: Kayseh Magan and Hamse Warfa's commentary) is "invalid" and "biased". Asking here to get opinions from uninvolved editors to see whether that's really the case. Some1 (talk) 00:03, 21 February 2026 (UTC)

@Some1, overall I think the question is acceptable.
I also think this is a good example of why we should avoid providing pre-set answers, because if someone changes what "A" means, then the responses are all confused. If the other editor had to type out their actual view in plain old English, then there could be no such confusion. WhatamIdoing (talk) 00:37, 21 February 2026 (UTC)
Thanks WhatamIdoing, I'll go ahead and remove the A, B, C, D from the RfC question. Some1 (talk) 00:41, 21 February 2026 (UTC)

RFC with no categories

I think I know the answer to this question, but would like another opinion. An editor started an RFC, and did not include any categories in the {{RFC}} template. An error message shows up on the talk page along with the RFC. The RFC does have an RFC ID. If another editor adds categories to the template, will the bot notice that the RFC has been categorized, and add the RFC to the lists for those categories? Robert McClenon (talk) 20:47, 16 March 2026 (UTC)

I think Redrose told me it would. I’d add it (I have in the past). That doesn’t necessarily mean that FRS will notify the category though. Dw31415 (talk) 21:53, 16 March 2026 (UTC)
(edit conflict) Yes, add one or more category parameters, and Legobot will add it to the relevant lists next time it runs (it's often forgotten that once an hour Legobot looks at every page with a transclusion of {{rfc}}, if only to check the expiry). But the RfC will remain in Wikipedia:Requests for comment/Unsorted unless the |rfcid= is removed. --Redrose64 🌹 (talk) 21:58, 16 March 2026 (UTC)

Please close

Why is Absolutiva inserting themself into a twenty-year old featured article without discussion? I found no several discussions above, only edit summaries and an RfC. I believe the talk page is the correct place to discuss disagreements. Can this RfC please be closed? Thank you. -SusanLesch (talk) 14:26, 18 March 2026 (UTC)

The tag has been removed. WhatamIdoing (talk) 16:06, 18 March 2026 (UTC)
Thank you. -SusanLesch (talk) 23:13, 18 March 2026 (UTC)

Question about Question about RFC

I have received a note on my user talk page saying: I'm thinking of requesting an RfC, but it'd need to be based on Wiki policies, not simply the number of votes. I interpret that as meaning that the poster wants to request in advance that the closer pay little or no attention to the number of votes. I think that the poster is concerned that a majority of responders may make arguments that are not based on policies and guidelines. I am not sure that I understand the rest of the post. But my question is what should I reply to this poster? Are they making a reasonable request that I can help them with? Robert McClenon (talk) 19:27, 20 March 2026 (UTC)

I read the question on your talk page: User talk:Robert McClenon#c-Israell-20260320105500-RfC question.... I’d suggest they review RFCBEFORE and post a regular discussion topic first (or provide a link to one of it exists). Dw31415 (talk) 20:06, 20 March 2026 (UTC)
A little more digging seems to indicate that it’s the editor’s proposal is to include “songwriter” in the info box for Rihanna: Talk:Rihanna#c-Israell-20251125145800-Songwriter (in the infobox). Seems like an RfC is appropriate to me. This page is the appropriate forum to discuss potential RfC’s so maybe just tag them here. My take is that info box contents are one of the most super-majority-dependent areas of editing. So the editor should open the RfC knowing that it might not go their way, but it’s a good way to get editors less invested about this artist to comment. Dw31415 (talk) 20:17, 20 March 2026 (UTC)
Hi @Isreall, Robert inquired here about your question to him. Please see my answer above. If your question was about the linked discussion, I recommend you draft an RfC in the existing discussion and give existing editors a chance to review it. Also decide which RfC category you might select. Robert, my apologies if I’ve overstepped. Dw31415 (talk) 12:47, 21 March 2026 (UTC)

Related Articles

Wikiwand AI