Wikipedia talk:Consensus/No consensus RfC 2022

Important points section

Are there any more "Important points" that should be added to that section? Possibilities will be discussed below.--John Cline (talk) 23:34, 20 January 2022 (UTC)

• @John Cline, I think you should remove the "policy is defined" bit, because Wikipedia:The difference between policies, guidelines and essays is subtle and blurry. Different policies have different expectations. WhatamIdoing (talk) 23:12, 22 January 2022 (UTC)

  Thank you for your suggestion WhatamIdoing. I am not averse to it's removal but let me first describe why it's there and how it came about. In anticipation of the rebuttal argument: "policy is descriptive, not prescriptive", which many esteemed Wikipedians hold closely as "the wiki-way", and to preempt its potential negative impact: I wanted to give policy a chance to self-define. And, while I do understand that different policies have different priorities and expectations, and that things become blurry. I also believe that successful governance depends on human nature to do the right thing which allows that as the letter (of rule) begins to blur, the spirit of rule will remain. This, of course, depends on the hope and belief that acceptance and clear understanding (of the rule) has occured at some point before the onset of fog. And this, of course, requires that a foundation of understandable rule must first exist (giving opportunity to accept and understand the things needed to survive a blizzard's whiteout if the blur is so great). From such a premise, I assumed the best place for policy to most clearly self-define would be the policy about policies: Wikipedia:Policies and guidelines. If I was correct, I expected to find what I was looking for. Now consider the fuller definition which is the continuation of that second "important point": "Policies are standards all users should normally follow, and guidelines are generally meant to be best practices for following those standards in specific contexts." If that doesn't set out which governing instrument describes, and which prescribes, from where would the prescription for standards of conduct come? And if standards can not be prescribed, only described, this entire exercise is folly, as is writing policy in the first place. That's why I thought it was an important point, that's why it's there, as I still believe it should be. And that's why I will wait to remove it until you reassert your belief that it is misplaced, or until another editor agrees with you that it is misplaced.--John Cline (talk) 01:14, 23 January 2022 (UTC)

  > where would the prescription for standards of conduct come?

  Mostly WMF Board policy. That's what forces us to have Wikipedia:Non-free content criteria and Wikipedia:Copyright violations and similar "prescribed" standards. For the rest, e.g., WP:V and NPOV and IAR and NOT, we looked at what good editors were already doing, and we tried to write down a description of what they were already doing, so that other people who wanted to be good editors would be able to figure out the best things to do faster, with less trial and error.

  I don't think you're wrong about descriptive policies being folly. I do think that this sentence is pointless verbiage. It doesn't matter what a policy is; it matters that there is (allegedly) a significant gap between what we do and what we write down that good editors (or admins, in this case) usually ought to do. WhatamIdoing (talk) 03:42, 26 January 2022 (UTC)

  I've removed the policy definition from the important points section.--John Cline (talk) 01:03, 31 January 2022 (UTC)

• About the bit that says This illustrates the difference between policy and practice. It is the conflict this RfC endeavors to resolve.:
  I don't feel like this illustrates the difference between policy and practice at all. It illustrates the difference between two policy options, but there isn't a word in there about whether and how the current policy differs from the current practice. WhatamIdoing (talk) 23:35, 22 January 2022 (UTC)

  I understand your point in this regard, and agree. I will copyedit that section to clarify matters. Thank you for pointing this out to me.--John Cline (talk) 01:14, 23 January 2022 (UTC)

How the RfC will be closed

Should this RfC be closed by one, two, or three closers? Should the closer(s) be administrators?--John Cline (talk) 00:10, 21 January 2022 (UTC)

I don't think there is any reason why the closure either should or should not be an administrator. If the outcome is clear then any uninvolved editor in good standing will be fine (although this should probably not be anybody's very first close). If the outcome is not clear then a single uninvolved experienced closer (with or without the mop) will be fine. A panel closure will only be needed if the discussion is extremely voluminous with lots of different arguments presented very few of which have an obvious consensus. Thryduulf (talk) 01:47, 21 January 2022 (UTC)

Thank you for your reply Thryduulf, I agree. I suppose the obvious follow on question would be how the decision will be made and by whom? My opinion is that the closer should be given the additional discretion to request a panel closure if they believe matters of complexity warrant the additional eyes and allow them to stipulate the panel size (up to 2 additional closers). This and the related matters of possibly relisting the discussion would probably indicate the prudent inclusion of a dedicated section titled "Instructions to closer" or some such. Do you agree, or do you have another idea for how and whom shall make the decisions?--John Cline (talk) 08:43, 21 January 2022 (UTC)

I don't think there is a need for as much formality as that. When the time comes either someone will step up to the plate themselves or we can request closure at WP:ANRFC, where someone suitable will see it. Any closer can request a panel to close any discussion if they think it warrants, no special authority is required. Thryduulf (talk) 10:04, 21 January 2022 (UTC)

Fair enough, I'm on board with your rationale.--John Cline (talk) 10:07, 21 January 2022 (UTC)

The RfC's duration

Should this RfC be closed after 30 days or are there provisions which you think would indicate it should remain open?--John Cline (talk) 00:10, 21 January 2022 (UTC)

If new arguments are being presented or discussion is both ongoing and constructive then I say leave it open, otherwise close it at approximately 30 days. Thryduulf (talk) 01:50, 21 January 2022 (UTC)

Here again, how and whom must be stipulated in some form at the RfC's outset. And here again instructions to closer seems possible.--John Cline (talk) 08:43, 21 January 2022 (UTC)

how and whom must be stipulated in some form at the RfC's outset Why? Thryduulf (talk) 10:05, 21 January 2022 (UTC)

I misspoke saying "must", let me ask this: Is there anything now stated in the RfC draft that you think needn't be said?--John Cline (talk) 10:16, 21 January 2022 (UTC)

I'm not certain whether "discard" in Consensus to discard the policy provision... is a neutral wording. I'd like other opinions on that. Thryduulf (talk) 10:28, 21 January 2022 (UTC)

Your feeling that it's not neutral, or even may not be neutral is enough to change it. I am going to mirror the question, "should it be retained or not?" to be consensus to retain or consensus not to retain. I got to go to work so I'll talk to you later.--John Cline (talk) 10:54, 21 January 2022 (UTC)

How are you going to interpret "Remove, or at least say my favorite process works the other way" and "Keep, but add that my favorite process works the other way"? WhatamIdoing (talk) 23:40, 22 January 2022 (UTC)

Future planned RfCs

If consensus is for retaining the policy provision, will this RfC stand on its own or should a contingent RfC be anticipated/planned?--John Cline (talk) 05:35, 22 January 2022 (UTC)

What would any contingent RFC be about? "Editors voted to keep this in an RFC, but did you all really, really, really mean that you want to keep this sentence?" WhatamIdoing (talk) 23:39, 22 January 2022 (UTC)

In this case, I mean to imply if it's kept while simultaneously believed to be unclear regarding "normally reverted" where perhaps example exceptions should be added, a clarification on what constitutes "contested administrator actions" for example: must tool usage be involved or is discretion included as in closing a discussion where tools are not used and the discussion did not specifically require an admin to close, Is an admin's self-requested review of a potentially controversial action exempt from contest by the self-request or is it in itself contested, and others that may arise. This RfC is certainly not poised to have these questions answered here and will hopefully maintain its narrow focus and scope. Only for the purpose of allaying these and similar concerns might I anticipate any need for a subsequent RfC provided consensus is for keeping the policy provision.--John Cline (talk) 01:36, 23 January 2022 (UTC)

General discussion section

Should the discussion in the "General discussion" section be one long discussion with bullet points and comments threaded underneath or should each new point be a separate level 4 header like this RfC?--John Cline (talk) 23:50, 20 January 2022 (UTC)

I'd say leave it as one long discussion initially, moving to separate sections (refactoring if needed) if it turns out there is a lot of different points being made. A hybrid with a general "general discussion" and section discussions of specific points is also a possibility that might naturaly evolve. Thryduulf (talk) 01:54, 21 January 2022 (UTC)

I like the hybrid model. Do you think an internal comment saying something like: <!-- COMMENTS GENERATING MORE THAN 10 REPLIES FROM MORE THAN 5 DIFFERENT EDITORS WILL BE CONVERTED TO A LEVEL 4 SUBSECTION --> should be added?--John Cline (talk) 05:35, 22 January 2022 (UTC)

Suggestions

I suggest making the initial question more direct, such as If an administrator action is reviewed and no consensus is reached, should the action be reverted? The background section can then point to Wikipedia:Consensus § No consensus; I suggest using the page and section names rather than a visible shortcut link. I also suggest making the background section more concise. I confess that I don't feel strongly enough about this issue to draft a proposed version, though.

In addition, I suggest not using "!vote" jargon. I realize some people think it's a part of English Wikipedia culture that everyone should learn. My personal preference, though, is to avoid use of jargon when it doesn't substantially simplify communications. In this context, it seems to mean "these look like votes but are actually expressions of reasoning to be considered by the closer when determining consensus". I think it would be simpler to label the section something like "Responses". isaacl (talk) 06:33, 22 January 2022 (UTC)

Thank you for those suggestions. I will begin copyediting the draft directly to implement most of them. I am reluctant to use the more direct question because it's already known that some exceptional scenarios exist that would confuse the answer between yes and no. Additionally, upon very sound advice, I was shown and came to agree that a necessary first step was to determine if the community supported the policy provision. I started to ask if the community"still supported the provision" but even that encroaches on non-neutral verbiage. Anyway, let me copyedit the draft and see how things shape up from there. Thanks again.--John Cline (talk) 07:27, 22 January 2022 (UTC)

My personal feeling is that it's more straightforward to directly ask for approval of the provision in question, rather than point to a section elsewhere and say, do you support what's written over there? I think this still qualifies as determining if the community supports the guidance in question, but I understand others may feel differently. isaacl (talk) 16:06, 22 January 2022 (UTC)

I understand what you mean and your rationale is sound, even convincing (at face value). In fact, I predominantly agree and in a perfect society, it's clearly the better way. Even in a "more perfect society" (far from perfect) my "real world", it's a good approach. But the laws of physics themselves wane in the global microcosm that is Wikipedia, and the shortest distance is rarely achieved in traveling straight. On Wikipedia, postulating a direct question that is faithful to a given source (so that nothing is lost or gained in its formulation) while ensuring that it isn't so convoluted that it effectively becomes unanswerable, is daunting at best. It's an expensive task (if time and energy have value) and probably a bad investment if done for expectant profit. I'll rest my forward on these things, while still on dry land, for the deep waters are not worth treading on matters so quaint as a Wikipedia RfC. Considering your example of a direct question that, in itself, also answers the larger question of the community's level of support for the underlying policy: "If an administrator action is reviewed and no consensus is reached, should the action be reverted?". Firstly, we have an "administrator action" v. the policy language: "contested administrator action"; aspersions for that unfaithful mismatch alone will range from non-neutral bias all the way to a deliberate attempt to self-serve some obvious nefariously motivated goal and the plethora of other wrongful attributions that will be affixed to call foul and allege a bad RfC. Then we have "should the action be reverted" v. the policy language: "normally reverted"; not only will this cement the first round of allegations, new aspersions will be born and the pairing will aggravate the degree of malice and you'll soon be worn out from defending any minor error that came out of your wholesome efforts in good faith that now have you painted as either incompetent, cunning, and/or contrived. Fixing the question to faithfully align with the policy provision further convolutes the question by opening the discussion to more debate: is an administrator action limited to tool usage, is discretion a contestable action, can an administrator ever function as a regular editor or is every move the action of an administrator by virtue of the flag, what constitutes a contest, is an administrator's self-request for review of a potentially contested action a backdoor to avoid a contest or is it in itself a contested action, and then there's "normally reverted", and the endless discussion for what that entails. For transparency, I'll admit that I'd like to see this RfC end in a well attended discussion that results in a strong and clear consensus to retain the policy provision. I also know that discussion ad nauseam, endless walls of text, and countless off-topic sub-discussion is the precise recipe for derailing that goal. And though I know the nay-sayers will come to ply the tools of their chosen trade (the strawman fallacies) that doesn't mean I am willing to serve up their fodder on even a paper plate let alone silver. And so, it was a decision, the exercise of choice after considering the things now said, to ask the question in a slightly less direct manner (but more easily answered). Because it is possible to answer yes that a policy should be retained without having to parse its actionability, or the clarity and quality of its prose. For example, I could respond at an RfC to retain wp:gng without having to agree that "sufficient coverage" was particularly well defined. That is also why I asked above whether or not we should anticipate/plan a contingent RfC for discussion after this RfC if the provision is retained (a preemptive effort to keep this discussion focused and within scope). I apologize for the long reply, but concision is not my best skill in writing. If it can be seen as honest and adequately thorough, I'm happy with that. Best regards.--John Cline (talk) 23:19, 22 January 2022 (UTC)

My sample question was just an example, and of course could have been worded to more closely match the page in question. (Concision is one of my skills, and so I naturally wrote a shorter version.) The background section can go into full detail about the intent to review the existing statement, with all of its connotations. In my view, the key is for the RfC intro to have a short, easily understood question that helps people decide if they want to participate, without having to go read another page. Nonetheless, I defer to your judgement in wording the question. isaacl (talk) 23:35, 22 January 2022 (UTC)

Is an RFC necessary?

@John Cline, have you considered mentioning your concern about this sentence on the policy's talk page? Or even trying a WP:PGBOLD edit to remove or fix it, with an edit summary like "This doesn't seem accurate"? You've put all this effort into creating a heavily structured page, but you haven't done the basic WP:RFCBEFORE work. WhatamIdoing (talk) 23:47, 22 January 2022 (UTC)

That's a fair criticism and I accept that some of it is deserved. Firstly, I should disclaim that I support the policy provision and that any bold changes I might have made to it would have been to strengthen its remit. My concern was not that it existed in policy, it was that it did exist, and had existed for over ten years yet I'd never seen it invoked in practice. While understanding that a result that "normally" occurs doesn't mean always, I also understood that it couldn't mean "never" occurs or even "almost never". To be sure it wasn't a phenomena where I just happened to always miss seeing this "normal" result, I did an extensive review of the AN/ANI archives and a 100% review of all challenged discussion closures. And while the AN/ANI archives have not been exhausted (there are thousands more to review) I did not find and still have not found a single example where a contested admin action was reverted on the basis of a no consensus outcome being reached in the discussion. I did find, and can easily find, several to many examples where Wikipedia:Consensus § No consensus (point 3) could have and arguably should have been used but was not. Notwithstanding the aforementioned, the bulk of my RfC-before work involved researching archives, editing histories, and policy for relevant information and initiating discussion queries with tenured Wikipedians whom I esteem to gain historical knowledge regarding this policy from its inception, and policy knowledge regarding the validity of a policy provision with a long history of non-use and being ignored. That research led to an open discussion with Arbcom that was attended by past and present members of the committee. Sub-topics discussed included: actionable policy enforcement, policy nullification for non-use, means of compelling admin observance and enforcement, as well as the enforcement arm of Arbcom from full case review to resolution by motion (all relating to the issues giving rise to this RfC). And, in that discussion it was more than reasonably shown to me that an RfC to establish whether the community still supported this policy provision or not was a necessary first step (because of its long standing non-use) before any effort to compel its usage in practice could even be possible. With that being the direct reason I began the work of drafting the RfC, I never concluded the RfC-before work which is ongoing even now. Therefore: yes, I think the RfC is necessary, and: yes, I have done and am doing RfC-before work.--John Cline (talk) 09:40, 23 January 2022 (UTC)

I think it is correct that we run this RfC...not least, it may nudge individuals down the path of complying with the policy - a "this is good in theory, we're just bad at accepting the somewhat abrasive feel of reverting actions with the consensus to cover us" Nosebagbear (talk) 11:19, 24 January 2022 (UTC)

I'm struggling to understand how removing incorrect content would strengthen the remit of the now-nonexistent content. WhatamIdoing (talk) 03:45, 26 January 2022 (UTC)

John, I've pinged you to WT:CON. Whenever editors can settle something this simple through a normal discussion, we should. WhatamIdoing (talk) 03:52, 26 January 2022 (UTC)

Request for help with statistics and search string formulation

If anyone can help with statistical determinations and/or the development of filtered search strings that will: demonstrate a comparison between contested administrator actions closed as no consensus that were reverted and those that were not (by year if possible) and/or deliver results showing the closure of contested administrator actions (with a no consensus outcome if possible) respectively, it would be of great help and tremendously appreciated. Thank you.--John Cline (talk) 11:04, 23 January 2022 (UTC)

The four scenarios

I've commented out the inclusion of the four scenarios said to demonstrate how one course of action can confer an advantage to the first admin to act and the other to the second as confusing and potentially unnecessary. If anyone thinks they should be reinstated, please explain how their inclusion is helpful. Thank you.--John Cline (talk) 00:37, 31 January 2022 (UTC)