Wikipedia talk:Requests for comment

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Dispute Resolution
WikiProject icon This page is within the scope of WikiProject Dispute Resolution, a collaborative effort to improve dispute resolution practices on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
          A Wikipedia ad has been created for this project page. Click [show] to view it.

RFC: Can originators dictate the scope of RFCs?[edit]

The following discussion is an archived record of a request for comment. Please do not modify it. No further edits should be made to this discussion. A summary of the conclusions reached follows.
There is no clear consensus for a yes/no answer to this question. The question itself seems on the face of it to be an attempt to legislate WP:CLUE and forcibly prevent thread drift (good luck with that). There are examples where it is appropriate, others where it is not appropriate, approach matters with good judgment and if necessary close any RfC and start again with a new, better question. Guy (Help!) 14:06, 28 May 2015 (UTC)

How much control may an RFC originator exert over the discussion? Is it appropriate (or should it be so) to actively prohibit alternatives from being proposed by other users and debated in the same discussion? The RFC guidance is silent on the subject of ownership. — (talk) 12:56, 25 April 2015 (UTC)

  • The best answer is a reasonable amount, subject to analysis on a case by case basis. On particularly complex or contentious issues, the only way to succeed in forming a working consensus to move forward may be to exercise some control, and narrowly focus the RFC on a specific proposal. So for example, if the choice is proposal A, or nothing, we may be able to achieve consensus on proposal A. But if there had been 3 more choices, support may have split between all of them, and there may have been no consensus for anything, despite there being consensus that keeping the status-quo, that the result of lack of consensus will cause, is worse than any of the proposals. So basically, you should respect reasonable amounts of RFC ownership by the creator, try to work with the creator to improve it, but the community can override that ownership if enough people feel its necessary. If the RFC is biased in its wording, or something similar, that can be fixed, and is going to be subject to much less deference to the creator than structural choices. Monty845 13:07, 25 April 2015 (UTC)
By way of example, Wikipedia:Pending changes/Request for Comment 2012 is a good example of how this can work in practice. It was an RFC on one question, out of many contentious issues that pending changes raised. Should we have it turned on at all? When there was a consensus on that one question, 3 followup RFCs, Wikipedia:PC2012/RfC 1, Wikipedia:PC2012/RfC 2, and Wikipedia:PC2012/RfC 3 where launched to implement that initial consensus. If all the issues, from all the RFCs, had been !voted on in the very first one, there is no way consensus on anything would have been reached. But with careful structuring, the community was able to work out a functioning compromise on a very controversial issue. And it worked pretty well given how little drama has happened since. (At least relative to the drama that has historically surrounded Pending Changes). Monty845 13:29, 25 April 2015 (UTC)
In your opinion, do you believe the OP appropriately exercised his powers in this RfC? ―Nøkkenbuer (talkcontribs) 15:20, 25 April 2015 (UTC)
The OP merely clarified the heading, before little discussion and no !voting, which is not the question. ―Mandruss  17:42, 25 April 2015 (UTC)
I think Nøkkenbuer was referring to a different RFC where a participant made a new proposal in the discussion section, and the originator shut it down by hidden archive. That was what prompted me to start this RFC, in fact. — (talk) 23:32, 25 April 2015 (UTC)
Ah, I failed to see the link. ―Mandruss  00:03, 26 April 2015 (UTC)

Writing up a new RfC and posting it in the middle of an ongoing RfC against the wishes of the author of that RfC is, in my considered opinion, disruptive. Anyone is free to post a new RfC that asks any question they choose. Hijacking an existing RfC -- especially after 50 or so people have responded and it becomes clear that the outcome is not going to go your way -- borders on gaming the system. --Guy Macon (talk) 00:01, 26 April 2015 (UTC)

I agree with Guy Macon. If the question changes, then previous respondents are responding to a question other than what was asked. In my opinion, any significant change to the lede of an RFC after it is published is disruptive. An RFC isn't an article. The question isn't one of ownership, but of consistency in what is being asked. If the original question was wrong, a new RFC can be published, and, if necessary, the previous one killed, but not rewritten. It isn't a matter of ownership, but of consistency. Robert McClenon (talk) 00:11, 26 April 2015 (UTC)
@Robert McClenon: What about posing a question after the lead, in the discussion section? Not replacing or changing anything, but adding alternatives. Such is the case that Guy is referring to. — (talk) 00:22, 26 April 2015 (UTC)
I don't think this can be answered in the abstract. A lot depends on the specifics of the RFC - what the original question was and what it was attempting to find out. Sometimes raising alternatives will be helpful and move the discussion forward, while at other times raising alternatives will be disruptful and sidetrack the RFC to the point where we never reach consensus on the original question. Blueboar (talk) 11:44, 26 April 2015 (UTC)
  • Comment I would like to add a bit of extra context to this discussion. This debate recently came up between myself and the IP that started this RFC regarding a sequence of edits at Wikipedia talk:Manual of Style/Film on the April 14-15 (I can't be bothered to post the diffs but you can see from the history where I 'spar' with the IP). The IP tried to alter the the RFC introduction and I reverted them, namely because the RFC was at an advanced stage and all of the replies had only considered one particular proposal. In short, I believed that altering the lead-in gave the impression that both proposals had been given due consideration, which was inappropriate. This led to the discussion at User_talk:Betty_Logan#MOS:FILM_RFC_reverts. A third opinion felt I had breached policy by altering another editor's talk page comments, so in the end I tried to restructure the RFC to allow for an alternative proposal and make it clear that which proposal the comments in the discussion releated to. The IP seemed to accept this solution (or at least that was the impression I got at the my talk page). I don't really have much interest in debating the merits of RFC ownership, or indeed whether I acted inappropriately by altering the IP's comments (I accept I breached policy and tried to rectify this the best I could without completely derailing the RFC), but hopefully the links I have provided will enable you to understand the IP's position better. Betty Logan (talk) 21:06, 26 April 2015 (UTC)
    • In the case Betty’s referring to (diffs: [1][2][3][4][5] and the discussion on her Talk), I had added a second option beneath her timestamped proposal (but above the survey section), making sure that my addition was itself clearly timestamped and distinct from the original post. I mention this because I feel it’s a critical detail that was left out here, making it sound like I’d altered the original proposal instead. Everything else recalled here is completely accurate, and is another example of what I’m seeking to clarify with this RFC. — (talk) 22:22, 26 April 2015 (UTC)
  • Some modifications are clearly helpful and do not interfere with the process of determining consensus. Examples from my RfC:[6][7] -- both welcome and helpful. Adding a new RfC in the middle of an existing one or making a change in the description of what it being !voted on after multiple people have supported/opposed the old description (not saying that's what happened here) clearly does interfere with the process of determining consensus. Likewise, my "ground rule" saying not to reply in the support sections. That is clearly helpful, and nobody would have a problem if, upon discovering such a reply, I moved it to the threaded discussion section. --Guy Macon (talk) 22:56, 26 April 2015 (UTC)

At DRN and MEDCOM the person controlling the discussion is a neutral party trying to keep the discussion on track and who has an obligation to insure that no one's position is slighted. They're also invested by policy with authority to do more at one of those venues than an editor has the right to do on an article talk page under the talk page guidelines. Having such rights in a RFC, however, where there's often no neutral party, simply invites someone to manipulate the discussion in favor of their preferred point of view. Some limited control allowing refactoring might be okay, but even that could be easily abused, I'm afraid. I think that it's important to remember that, at the end of the day, RFC is really nothing more than another garden-variety talk page discussion with two exceptions: first, the ability of the requesting editor to frame (hopefully narrowly) the initial point to be discussed and, second, a mechanism to invite the broader community to weigh in. (Just to avoid any internal-copyvio issues, let me note that this is is a slightly reworked and expanded version of a post I made at Betty's talk page].) Regards, TransporterMan (TALK) 14:49, 27 April 2015 (UTC)

  • As Monty said: The best answer is a reasonable amount, subject to analysis on a case by case basis. RFC initiators have as much (or as little) power and respect as participants give them. If an RFC initiator applies a hidden archive, people could either respect that as constructive, or revert it if it's abusive. If a significant number of participants object to the phrasing or handling of an RFC, that can invalidate the RFC. Sufficient participants can generate valid consensus which totally disregards the original question and disregards attempts at control. If sufficient-participants-to-establish-a-consensus voluntarily abide by RFC-initiator-guidance, everything's probably fine. Alsee (talk) 04:29, 1 May 2015 (UTC)

The above discussion is preserved as an archive of the debate. Please do not modify it. No further edits should be made to this discussion.

Updating the RfC guidelines to better reflect present usage[edit]

While I've not done a detailed statistical count, my impression is that at present RfCs are frequently (most often?) used as vehicles to host community '!votes'. If this is indeed the case, it seems to me it would be wise/pragmatic/kind-to-newbs to adapt the guidelines to better reflect common practice. i.e. To bring de jure more closely in line with de facto. 'Principal-of-least-surprise', etc.

Could be a matter of simply incorporating a few links and/or comments regarding '!votes', initiating a new section to cover the topic more thoroughly, or if feeling adventurous/ambitious revamp the whole page. Thoughts? --Kevjonesin (talk) 15:01, 7 May 2015 (UTC)
p.s.— Should we hold a RfC on adapting RfC? It'd be delightfully recursive. <wink> :  } --Kevjonesin (talk)

Kevjonesin, just to make sure: You do realize, don't you, that when there's an RFC that's set up to get in "opposes" or "supports" or other short-form responses that it's still not a vote? A proper consensus evaluation closing that RFC must not only take into account the numbers, but first and more importantly, the quality of the arguments made for and against the proposal. Straw polls, on the other hand, can take place via RFC or in the process of an RFC but can never create consensus (though they can help analyze or focus it). It may be worth mentioning here that RFC's can do those things in regard to straw polls, but it needs to be done pretty specifically so as not to suggest that RFC's can be used to determine things by voting and, frankly, I have to wonder if that's just not too fine-pointish to be worth the effort. Best regards (and I like your humor), TransporterMan (TALK) 15:44, 7 May 2015 (UTC)
Hi TransporterMan. Yes, I have some awareness of the official dogma/'wikidealism' surrounding such; hence my use of (and link to) the term "!vote". I'm not sure though how well (whether?) present common practice really coincides with such.
As a 'side note', I came across the origin story for the term "!vote" yesterday while wandering about—and found it somewhat amusing.[8][9] It would appear that there is some precedence for feelings of Orwellian dissonance regarding these things. :  }

"Remember, folks, voting is evil, but !voting is perfectly acceptable. --Deathphoenix ʕ 14:15, 9 June 2006 (UTC)"[10]

I'm not proposing anything so radical as officially overturning the underlying dogma that 'we do not vote'. However, I would like to more clearly and explicitly express to those coming to this page for guidance that there are in fact in common practice a variety of procedures involving editors explicitly expressing opinions 'for' and 'against' various proposals—which often carry not-insignificant weight in the process of determining and implementing 'consensus'—which frequently take place under the auspices of a 'request for comments'.
If an archetypical 'reasonable person', previously unfamiliar with Wikipedia lore, were to thoroughly peruse WP:RFC/A for a time and then start making inferences, what might they infer? I'm proposing that we explore ways in which to address such inferences a bit more forthrightly in our guidelines. --Kevjonesin (talk) 01:02, 8 May 2015 (UTC)
One of the problems with officially enshrining this approach is that it would make it happen even more often (which is not necessarily desirable) and that it would fail to highlight the differences between the English Wikipedia and most of the others, where actual votes and usually strict majority rule are typical. (Your comment in an RFC elsewhere is mostly an effort to persuade subsequent people to vote along with you, not to share your views.)
I'm leaning mostly "no" to this. WhatamIdoing (talk) 21:03, 13 May 2015 (UTC)
  • I think a lot depends on what the purpose of the specific RFC is... In some cases, the purpose is simply to assess whether a consensus exists on a disputed issue (and if so, to determine what that consensus is). This type of RFC usually takes the form of a quick support/oppose !vote... with comments explaining why participants hold the view that they do. This format makes sense if all we are trying to do is get a quick assessment of opinions.
In other cases, however, the purpose is to actually achieve a consensus... this type of RFC usually takes the form of a (often rambling) discussion, with different editors suggesting different ways to resolve the dispute.
Of course, if the editors who file the RFC are not clear what the purpose of the RFC actually is, then the resulting format will be equally confused... with some editors !voting, and others discussing.
Perhaps we could do a better job of explaining how RFCs with different purposes can be formatted in different ways... and which format works best for different purposes.
I find it is often helpful to intentionally create a series of "phased" RFCs. Set things up so everyone knows that there will be two back-to-back, related, RFCs... the phase one RFC (in !vote format) designed to assess whether there is a consensus (or not); and the phase two RFC (in discussion format) designed to figure out what to do next (if anything) - given the results of the assessment from phase one. The key is to clearly state the purpose of each "phase" in the RFC postings... If you do it right, you often get a productive RFC, with few objections. Blueboar (talk) 16:26, 28 May 2015 (UTC)
Blueboar, I think that one of the causes of the increased tendency to "vote" is that the "good examples" on the page are focused on binary questions. You either change that picture or you don't; there's no middle ground or compromise position. I wonder if you could come up with an example that would show a good (=likely to result in productive conversation, not too vague) question that doesn't have a 'pick one option' result. WhatamIdoing (talk) 08:55, 1 July 2015 (UTC)
I'll give it a try. Blueboar (talk) 10:45, 1 July 2015 (UTC)
You might also want to look at the newly created Wikipedia:Writing requests for comment. WhatamIdoing (talk) 16:16, 2 July 2015 (UTC)

Amending a proposal in an ongoing RfC[edit]

Question. I posted an RfC here asking if a statement is supported by its sources. In the course of the RfC, it became clear that the an amendment to the statement to clarify it, would be useful. So in this dif I amended it using underline redaction, and added a note below it to state that i did that. I then went to the Talk page of each editor who had !voted or commented to give them notice of the change. Some editors are objecting that this is "illegitimate". I don't believe it is and have seen it done before. The RFC instructions are silent on this. So... am I right or wrong? If i am correct, would it be useful to add something on this to the instructions? thx Jytdog (talk) 11:48, 2 June 2015 (UTC)

In the case that you mention, you notified the editors who had previously !voted, so that gave them a chance to change their comments. There is a common practice, unfortunately, for a tendentious editor, especially one who thinks that the RFC will go against him, but who knows beyond knowledge that he is "right", either to change the wording of the RFC, or to deface it, such as by inserting a statement above the first question criticizing the RFC. This is very disruptive (and is done precisely as a way to achieve one's will disruptively). I think that they were criticizing what you did because when an RFC is changed, it is usually done disruptively. Robert McClenon (talk) 02:34, 19 June 2015 (UTC)
In the broader case, people who complain about "non-neutral" or "illegitimate" procedures are usually worried that their side is "losing".
Jytdog, so long as the person who started the RFC doesn't object (because it's possible to 'clarify' a question from "is this a picture of a cat?" into "shall we delete pictures of dogs?"), and people can still make sense of the responses (e.g., not reversing support/oppose !votes), then your approach is a model of transparency. WhatamIdoing (talk) 08:59, 1 July 2015 (UTC)
Thank you both for replying! Jytdog (talk) 10:54, 1 July 2015 (UTC)

Suggestion on RfC process[edit]

I think RfC would work better if we banned threaded discussion and asked each user to comment in their own region.MissPiggysBoyfriend (talk) 05:43, 11 June 2015 (UTC)

I strongly disagree. I find that contributors to the discussion sometimes make good points, which make me change my view. Indeed, I am starting to wonder if my irony detector just failed. 07:10, 11 June 2015 (UTC)Maproom (talk)
Heh, you and I must read different RfC's. Try this one -- all the threaded discussion makes it difficult to get a sense of how the RfC is going.[11]. I have no objection to people answering each other in their own sections; I find that such discussion makes it a lot easier to skip over certain folks who are throwing more heat than light.MissPiggysBoyfriend (talk) 12:48, 12 June 2015 (UTC)
Yes, that's a bit difficult. But RFCs are just normal talk-page discussions, perhaps on a somewhat grander scale than average. You can no more ban threaded discussion from an RFC than you can from a normal talk-page discussion. Also, threaded discussion can be remarkably productive, if you can get people to stop voting and start trying to find things that work. WhatamIdoing (talk) 09:02, 1 July 2015 (UTC)

RFC Notice Template Taskforce[edit]

Question: Is there a taskforce established to create and promote new RFC notice templates?

Habatchii (talk) 07:05, 11 June 2015 (UTC)

Habatchii, I'm not aware of any. When do you think a new template would be used? WhatamIdoing (talk) 09:03, 1 July 2015 (UTC)

What RFC Tag to Use for Military[edit]

What RFC tag should be used for military questions? pol and sci seem the "least wrong". Are there other opinions? Robert McClenon (talk) 23:56, 17 June 2015 (UTC)

hist, if it's about military questions referring to past years; econ if it's about the military-industrial complex; bio for military people; and sci (as you said) for technology and military medicine. It could really be just about anything. WhatamIdoing (talk) 09:05, 1 July 2015 (UTC)

Changing the Wording of an RFC Other Than by Proponent[edit]

I don't see a statement that editors should not change the question at the top of the RFC without consensus or consultation with the proponent. I think that there should be a statement to this effect. This is an unfortunately common form of disruption. In addition to having any previous replies no longer be replying to what was originally posted, it causes work for the bot and may confuse the bot and complicate the result of the RFC. I have seen this done often enough that I don't think that prohibiting changing an RFC non-collaboratively is a case of advising someone not to stick beans in their nose. Robert McClenon (talk) 02:38, 19 June 2015 (UTC)

Actually, I think that there are very good grounds for suggesting that nobody should change the wording of an RfC without the express consent of those who have already responded. Even an apparently minor change can prove contentious - or give whoever 'loses' the RfC an excuse for Wikilawyering even if it makes no substantive difference. AndyTheGrump (talk) 02:55, 19 June 2015 (UTC)
I agree that no one should change the wording of the RFC without express consent. As Jytdog noted above, in the case he mentioned, he did at least provide notice. In the case in point today, an editor actually reversed the wording of the RFC, which had been whether to include a paragraph, and the proponent of including the paragraph reversed its wording to ask whether there was a reason to exclude the paragraph. The obvious problem with that is that it reversed the meanings of the !votes and (until corrected) made them seem ridiculous. Robert McClenon (talk) 03:00, 19 June 2015 (UTC)
To my way of thinking, anyone other than the creator of the RfC editing the wording without prior consent is doing so contrary to WP:TPG anyway. Editing other peoples' posts in a manner which changes their intended meaning is the sort of behaviour that frequently leads to immediate blocks. AndyTheGrump (talk) 04:12, 19 June 2015 (UTC)
Yes. However, an argument can be made that the question of an RFC is a special case, which is why I think that language concerning it in this guideline would be appropriate. Robert McClenon (talk) 04:49, 19 June 2015 (UTC)
Unfortunately, this idea is not practical. "Don't change it" sometimes means "let the RFC pages remain broken if the OP doesn't agree to change the question". Also, very long statements sometimes need a very brief 'question' added to the top (for the sake of the bot/central listing pages). And sometimes, the OP isn't handy, but it's clear that you can improve the question without violating its integrity.
One should be slow to change questions without consent, but I don't believe that we should prohibit it. However, it's often desirable for all but very minor or technical changes to not be posted under the OP's name. For example, if a six-paragraph rant gets summarized at the top as "Is X a good source for Y statement?", then that should be posted to the RFC with only the date, not with the OP's name (unless the OP wants to go back and add his/her name later). WhatamIdoing (talk) 09:10, 1 July 2015 (UTC)

RfC limit?[edit]

I've noticed that certain users bring RfCs with great frequency when they do not get their way on Talk pages. I suggest that each user be given a limit of, say, two RfCs per month, or, perhaps better: you have to stop bringing RfC's for a month if you bring two *unsuccessful* ones (that is, one where the requester's position on the relevant Talk page is not successful) in a row. Or some other kind of limit that would discourage abuse of the RfC process. -- Ssilvers (talk) 15:57, 20 June 2015 (UTC)

There are no limits, but if it becomes disruptive, and gentle education does not seem effective, then blocks or WP:TBANs for WP:Disruptive editing, talk-page abuse, WP:IDHT, and even WP:CIR problems are possible.
Also, it's more complicated than that. Imagine that someone creates 31 RFCs, at a rate of one RFC per day. Assume that all close on the bot's default date rather than a sensible point in the discussion (as chosen by the participants). Assume that all can be classified as "unsuccessful". When the second one finally closes as unsuccessful, there will still be 29 other RFCs in flight... and the ban will expire almost exactly when the last RFC (the one created the day before the ban kicked in) expires.
Rather than creating rules, you're better off seeking individual action on the basis of the real problem (e.g., disruption). (Also, it doesn't happen that often. I think I can remember two such instances over the last half-dozen years, although I know that I haven't seen all cases. But compared to, say, someone trying to use table formatting for the question, it's much rarer.) WhatamIdoing (talk) 09:20, 1 July 2015 (UTC)