Wikipedia talk:Consensus/Archive 17

Archive 10

←

Archive 15

Archive 16

Archive 17

Archive 18

Archive 19

Archive 20

Decisions by WMF software developers

PROPOSAL WITHDRAWN. While I still believe the wording of policy could be improved, clearly there is no consensus for my proposed changes. As always, I am happy to follow the consensus; my attitude is that if my proposal was as good as I thought it was, it would have gained some traction and attracted some supporters. --Guy Macon (talk) 14:55, 10 July 2013 (UTC)

I would strongly support a proposal in the spirit of yours. I only learned about this proposal right now. In fact, I only discovered the "exception" to Consensus in Wikipedia's policies just now, through the discussion on Wikipedia:VisualEditor/Default_State_RFC. That discussion is an example of developers forcing changes on the site as a whole, when there seems an overwhelming opposition among editors. Personally, knowing that the WMF can force changes on the site against the consensus of editors makes me (a) less likely to want to contribute my time and energy to editing the site, and (b) less likely to want to donate money or to recruit other, wealthier individuals to donate money. It seems to go against the spirit of the site as a whole to enable such easy overriding of consensus. I would strongly like to reword this too, although I'm not sure your proposal is the best one. It's a good starting point though to spark some discussion though. Cazort (talk) 17:20, 6 August 2013 (UTC)

I would like to propose the following change in order to clarify our policy. More comments after the two versions.

CURRENT:

---

• Some matters that may seem subject to the consensus of the community at the English-language Wikipedia (en.wikipedia.org) are, in fact, in a separate domain. In particular, the community of MediaWiki software developers, including both paid Wikimedia Foundation staff and other volunteers, and the activities of Wikimedia Commons, are largely separate entities, as are the many non-English Wikipedias. These independent, co-equal communities operate however they deem necessary or appropriate, such as adding, removing, or changing software features, or accepting or rejecting images, even if their actions are not endorsed by editors here. This does not constitute an exhaustive list as much as a reminder that the decisions taken under this project apply only to the workings of the self-governing community of English Wikipedia.

PROPOSED:

---

• Some matters that may seem subject to the consensus of the community at the English-language Wikipedia (en.wikipedia.org) are, in fact, matters that affect the many non-English Wikipedias or even non-Wikimedia Wikis that use MediaWiki software.

One example is the activities of Wikimedia Commons, a separate entity that serves images to Wikipedias in all languages. They have their own policies and procedures for accepting or rejecting images, even if their actions are not endorsed by editors here. Images that are stored on the Wikipedia website instead of Wikimedia Commons are subject to consensus.

Another example is the rollout of new software features. The same software is used by all of the different language Wikipedias, and we cannot force our consensus on other Wikipedias by mandating changes in how the software works. On the other hand, other than a few exceptional cases, turning a software feature on or off or deciding which access levels can use the feature is configured on a per-wiki bases, and thus we can decide by consensus to turn it on, turn it off, turn it on only for administrators, etc. Other configurable options include naming pages or features. On the English Wikipedia we can "edit" the "main page". On the Icelandic Wikipedia they would "breyta" the "forsíða", and on the German Wikipedia they would "bearbeiten" the "hauptseite". These names are configuration options, and thus we could, by consensus, decide to "update" the "home page" without that decision affecting other Wikipedias.

This does not constitute an exhaustive list as much as a reminder that the decisions taken under this project apply only to the workings of the self-governing community of English Wikipedia.

---

Background: There is an ongoing discussion at Wikipedia talk:Flow‎ about whether changing the term "talk page" to "boards" is subject to consensus. A couple of editors cited WP:CONEXCEPT, saying things like "The devs are not obliged to follow onwiki consensus (per WP:CONEXCEPT)". At that point I realized that the paragraph in question can be interpreted to mean that there can be no consensus-based restrictions on the actions of WMF software developers. This, of course, is not true in practice, as seen by our experience with Pending Changes.

Because of this, I have tried to create a more balanced statement that better reflects actual policy as it is typically applied. I am not married to the wording I chose above (it is longer and possibly instruction creep) but I do think we need something better than "These independent, co-equal communities operate however they deem necessary or appropriate, such as adding, removing, or changing software features". We need language that makes it clear that we cannot impose our preferences on other Wikis or create extra work for the WMF developers, but neither can the WMF developers have absolute power over the English Wikipedia, as the current language implies. --Guy Macon (talk) 04:30, 9 July 2013 (UTC)

In the case of a disagreement, what mechanism would prevent the developers from implementing a change that they believed was correct over the objection of Wikipedia? --Ring Cinema (talk) 13:57, 9 July 2013 (UTC)

There are two answers to that question; what could happen in theory and what does happen in practice.

In theory, the WMF developers can do anything they want including turning Wikipedia into a Facebook clone, and there is nothing the English Wikipedia could directly do to stop them. Jimmy Wales or Sue Gardner could stop them, and if the decision was egregious, they would. Then again, in theory whoever has the keys to the data center can override any decision by a software developer who logs in remotely, and in theory a court order can override the WMF.

Those are all things that could happen in theory, but in practice, the developers are reasonable people who really do want what is best for Wikipedia, and WMF management is extremely cooperative and helpful, balancing meeting our needs with making sure that the needs of all of the different language Wikipedias are met.

My main concern is that the present wording implies that we can't make consensus decisions like the decision we made to turn off pending changes a while back or the more recent decision to turn pending changes back on. Clearly we can and did make those decisions, and the wording should reflect that. --Guy Macon (talk) 15:32, 9 July 2013 (UTC)

Two points:

• If the consensus here requires any effort or cooperation from the devs, then you can't actually "can't make consensus decisions like the decision we made to turn off pending changes" unless they voluntarily agree to go along with it. NB that in the case you mention, the devs did nothing at all. PC was "turned off" only in the sense that we instructed admins not to click that particular button.
• You need to read this again: "...when we say "board", we're not discussing what the term is on enwiki - we're discussing what the software concept is called. That's not something that falls under the consensus policy, and if you think it is then I'd suggest you re-read it and take into account the section about the community not having control over technical decisions. What the software concept is called, what the code refers to the object or class as, is totally unrelated to what it's called on the English-language Wikipedia. We could call it "Happy-Fun-Talky-Place" on enwiki and still have the software refer to it as a board. Each project can call it a totally different thing, because it's just a string in a localisation file."
  Nobody has told you that the en.wikipedia name must be "board". They've told you that what feels good to you personally isn't the deciding factor (they're going to get data), and someday they'll probably get around to telling you that "talk" may be technically impossible because both "boards" and the existing talk pages will have to co-exist during the transition (unless "No, I meant my other talk page, not this talk page" is the sort of conversation you like having), but nobody has said that it will definitely be called "board" here. WhatamIdoing (talk) 20:14, 9 July 2013 (UTC)

@WhatamIdoing: unfortunately, the very reasonable diff you linked was subsequently contradicted – strongly! – by another WMF staffer, which is why I think Guy proposed this idea here. That said, I have low enthusiasm for this proposal, as much as I sympathize with its intent. The current wording is descriptive of the way things currently are, and a change here is not, in itself, going to change the facts on the ground. If the existing wording says that developers can sometimes act outside of WP consensus, and we take that out, that's not going to make them act within consensus if they don't want to: you can't enforce it. --Tryptofish (talk) 23:11, 9 July 2013 (UTC)

See also: Wikipedia:Petition to the WMF on handling of interface changes. --Tryptofish (talk) 23:52, 9 July 2013 (UTC)

I don't think that Jorm actually contradicted Oliver. NB that the person you're saying "contradicted" this reasonable statement is the exactly same person who wrote "Nomenclature in this document (e.g., "Flow board") is product design facing and not user facing. User facing terms will very likely be different than those included herein." in the official documentation at Mediawiki. Twice. We can (and should) quibble with Jorm's inappropriate omission of hyphens, but I don't believe that anyone could read that statement and come away with a rational belief that "Flow board" is the final, user-facing name of the feature on all projects. WhatamIdoing (talk) 01:42, 10 July 2013 (UTC)

Jorm and I are, hyphens aside, in full agreement. Let me reiterate; there is pretty much no way to make a MediaWiki extension that wouldn't allow for language- or project-specific user-facing terminology. Theoretically I guess you could do it, if you only wanted it to work in English, but.... Okeyes (WMF) (talk) 03:56, 10 July 2013 (UTC)

I would also suggest that the commentary around the community turning it on, turning it off, etc, does not align with how things actually work. We will listen to the community, certainly, and we have disabled/scaled things back in response to power user concerns in the past - pending changes, liquidthreads, AFT5. On the other hand we can point to just as many examples - Echo, the VisualEditor, ACTRIAL - where power user concerns have not prevented the software moving forward (although they have led to things like changes in deployment timetables or feature sets). To write into policy "the community can demand anything be turned off" is, as WhatamIdoing astutely points out, to pretty much guarantee dev-community conflict: and to guarantee that it is a conflict the developers win. You cannot convince people to do something differently merely by citing a policy they are not bound by. Okeyes (WMF) (talk) 04:00, 10 July 2013 (UTC)

"The term 'Board' was chosen because it's just that: a bulletin board about you. A place for people to pin messages. Now, to the idea about 'being too social network', I say 'nuts'. This is the year 2013. We have scads of tests and data that show new users simply do not understand what 'Talk' means. This is the reason why the 'talk' tab was labelled 'discussion' for so long (and then reverted back, because too many templates said 'leave a message on the talk page'). This is an artifact of history, and it's one we'll do well to get rid of and move on with our lives on."

I am having trouble interpreting the above as not being about what new users see. Yes, I realize that the phrase "we're not talking in 'user facing' terminology at this point, just software terminology" was used earlier in the same comment, but I am having trouble reconciling that with "We have scads of tests and data that show new users simply do not understand what 'Talk' means". How is that relevant if we are talking about something that new users will never see? --Guy Macon (talk) 06:16, 10 July 2013 (UTC)

I interpret it as Jorm basically saying "We're not talking about user-facing terminology and shouldn't be, but if you insist on doing so, here's why my user-facing terminology is better". Not the best structure to an argument, but a structure. Okeyes (WMF) (talk) 13:08, 10 July 2013 (UTC)

It had not occurred to me to interpret it that way, but now that you have written the above, I will start doing so. I find your interpretation to be quite reasonable. This resolves all of my concerns. --Guy Macon (talk) 14:47, 10 July 2013 (UTC)

At this point, I have no idea what to think. And hyphens are the furthest thing from my mind. I consider myself capable of reading the English language in the plain meaning of words, and I cannot see how what Jorm said as meaning anything other than what I thought it said. Okeyes (WMF), you could save everyone a lot of drama if you would go back to WT:Flow, and explain in detail how editors should understand the "board" thing. I say that knowing that you already did it once, but the discussion subsequent to when you did it turned in a very different direction. Thanks! --Tryptofish (talk) 16:37, 10 July 2013 (UTC)

Sure; I'll try to get something up later today, but the VE is sucking up a lot of my time at the moment. If I haven't posted something by the end of the day, ping me. Okeyes (WMF) (talk) 17:10, 10 July 2013 (UTC)

Thanks! I just made an edit there that could serve as a perfect place for you to reply to me. --Tryptofish (talk) 17:14, 10 July 2013 (UTC)

Jorm? Oliver? What's with the obscurity? I don't think this is a place for the thrill of insider conversations. --Ring Cinema (talk) 20:45, 12 July 2013 (UTC)

defining "local" etc.

I suggest that we use "local consensus" here to mean a consensus on a specific article talk page, where editors who are not already involved in the article are unlikely to appear. We then have "project consensus" applying to policies and guidelines where one expects a much larger number of people to be involved in any changes to a policy or guideline. Third, we have "noticeboard consensus" which, we hope, has a different group of editors than are present in the local article, but who are basically making their opinions on the basis of Wikipedia policy and (we trust) not on the basis of personal points of view about the topic.

Lastly, we have "RfC consensus" where basically we mix type 1 and 3 together -- hopefully still in accord with policies.

If we tentatively work with these definitions (arbitrary as they might seem) perhaps we can generally state that if a local consensus consistent with Wikipedia policies can be worked out, that such a consensus is good and proper. If concerns about the applicability of policies and guidelines arise, then the use of an RfC or noticeboard discussion to weigh the policies is proper, and no "local consensus" can actually override certain policies (inter alia BLP and NPOV for example).

Does this work for anyone as a basis to more specifically elucidate what goes on (or should go on) in a consensus discussion? If not - oh well. Collect (talk) 22:40, 30 July 2013 (UTC)

It's a good idea but perhaps you will accept a little fine tuning. 'Project consensus' as you describe it might be misleading because the causality, I understand, flows from the practices of Wikipedia editors -- or maybe just the good practices -- which the policy supposedly reflects.

Secondly, I would suggest we keep an open mind about how it really could come about that a certain policy could overrule editors who reach consensus after discussion. The implication that good faith editors would ignore an important policy seems to me a contradiction; if they think the policy should be ignored, it must not be as important as their reasons. While this objection does not speak to the accuracy of your definition, we could spend a lot of time figuring out how to express this idea accurately when in fact it is more than possible that it is an empty set. --Ring Cinema (talk) 22:59, 30 July 2013 (UTC)

As I have seen "consensus" try to override WP:BLP as a practical matter, I can assure you that such cases exist, and in abundance where strongly held political, religious or sexual beliefs are found. To wit, an editor who felt that (essentially) "conservative Christians are anti-gay, and that this must be exposed in their BLPs." And another editor who felt "the most important part of a BLP was to expose the hypocrisy of those whom he felt needed to be exposed". Yes - there are editors who would indeed try to override such policies as BLP and NPOV. Collect (talk) 01:00, 31 July 2013 (UTC)

Okay, thanks. That is the first genuine counterexample I've come across. So it's not the null set. The flip side of the problem, as I'm sure you're aware, is an outlier who would try to commandeer a discussion based on a questionable reading of a policy. But that's not a matter of definition. --Ring Cinema (talk) 01:35, 31 July 2013 (UTC)

Presumably those who discussed it on the basis of the policy relevant to that noticeboard and not having any content bias one way or another would not reach consensus contrary to policy, we hope. This does presuppose that we can aver that noticeboards exist primarily to deal with the policy relevant to that noticeboard. Collect (talk) 01:58, 31 July 2013 (UTC)

I don't follow you. As a rule, content bias or its lack is not connected to policy conformity, while there is no problem with editors reaching a consensus at odds with policy if that is what they want to do. It's only an issue if another editor wants to follow the policy, but then there's not a consensus to ignore the policy. Perhaps I don't understand you. --Ring Cinema (talk) 12:20, 31 July 2013 (UTC)

I know the theory - but in practice some editors do have different interpretations of policy depending on what they wish the result to be <g>. The purpose I suggest is to ensure that the policy discussions be grounded in policy, and for that folks who have no axes to grind on a topic are possibly (hopefully) better suited for determining exactly where that line is than are people with labile interpretations of policy. Is that a bit clearer? I really think this may set a groundwork for substantial improvements here. Thanks. Collect (talk) 12:44, 31 July 2013 (UTC)

Now you don't understand me. My point is exactly that tendentious interpretations are ubiquitous. There is no one without an ax to grind. Further, there is no avoiding interpretive bias; it is a fact of life. With respect, there is no authority outside of consensus and no consensus outside of discussion. Thus, there is no method that "guarantees discussions are grounded in policy." To imagine that someone outside a discussion can or should impose an interpretation of policy application on editors who know a subject best is not a prescription for excellence. Ignoring policy is fine if that is the consensus view but I believe you don't accept that. --Ring Cinema (talk) 16:09, 31 July 2013 (UTC)

As the Five Pillars note - NPOV is "non-negotiable" and WP:BLP is strong both for moral and legal reasons. Some other policies may be weaseled around, but not those particular two AFAICT, Can we agree on that part? <g>. Then once we have the language figured out - perhaps we can make this policy the strong "dispute resolution tool" it is intended to be. Cheers. Collect (talk) 22:11, 31 July 2013 (UTC)

There will always be disagreement about how to interpret policy, even when the relevant text is not in question. That's one point. Secondly, new policy arises from new practice. The way to change a policy is for a new practice to emerge. Thus, violating policy is sanctioned -- in both senses. Thirdly, I sense that you believe there is some authority somewhere that can force compliance. Not only is that not desirable (see my second point), it is not possible. The authority of consensus comes from consensus. Please let's not pretend there is a God's-eye point of view from which we can see all the truths. It's not there. --Ring Cinema (talk) 22:30, 31 July 2013 (UTC)

Well said. --SmokeyJoe (talk) 23:40, 31 July 2013 (UTC)

It is actually a two way street... policy does and should influence consensus discussions ... but consensus discussions should in turn influence policy.
In any "local" (or even "project level") consensus discussion, we obviously should discuss what our policies say on the issue ... however, if enough "local" consensus discussions "go against" the same policy point, those discussions are a good indication that community consensus about the policy point has changed... and that the policy needs to be re-written to better reflect the new consensus. Blueboar (talk) 15:28, 2 August 2013 (UTC)

In the interest of consistency I am trying to mentally recast those comments, Blueboar, without resorting to "community consensus". See above where we were thinking that community consensus always refers to policy or guidelines, so it's redundant terminology, therefore confusing. However, nascent policy changes might be the one area where something is community consensus but not policy. If that is so, perhaps this area requires some codification in writing. --Ring Cinema (talk) 17:10, 2 August 2013 (UTC)

Silent consensus is also a form of consensus - e.g. if there is something widespread, that nonetheless violates written guidelines, this can be considered a form of community consensus to overrule the written guidelines.--Obi-Wan Kenobi (talk) 16:02, 3 August 2013 (UTC)

If there is something widespread, that nontheless violates written guidelines... that is usually an indication that the guidelines don't actually reflect community consensus... and that the guidelines need to be re-written.

I'm not sure we can recast in the way Ring wants us to. I don't think it is accurate to say that "community consensus" always refers to policy or guideline pages. It certainly isn't limited to just policy/guidline pages. A broad based RFC can be a very good indication of "community consensus". Indeed, sometimes a broad based RFC can be a better indication of "community consensus" than a policy page - because more people were involved in the RFC than were involved in crafting the policy/guideline. Blueboar (talk) 16:39, 3 August 2013 (UTC)

Neither do I, based on the discussion above, but the meaning of 'community consensus' seems limited to policy, guidelines, and nascent policy changes. That is not much wider than just policy and guidelines. --Ring Cinema (talk) 17:30, 3 August 2013 (UTC)

I would say the definition of "community consensus" is: "A wishy-washy mixture of the stated opinions and unstated practice of the Wikipeida community at large".

Policy should not be confused with "community consensus"... Policy is supposed to reflect community consensus... but it can sometimes get out of sync with it. Blueboar (talk) 20:22, 3 August 2013 (UTC)

I see these abstract conversations are continuing ad infinitum - but unless you are prepared get down to the gritty detail of who decides what the consensus is and on what basis, and how that consensus can then be enforced if necessary, then the whole thing will have no practical effect whatever. People will always claim that their edits are supported by consensus or for some reason don't need to be, and that the edits they oppose lack consensus. With no way of adjudicating such claims, they become empty, and the matters will be decided by (if it comes to that) an edit war. In most cases it doesn't come to that, because consensus happens naturally (or one party can't be bothered to argue) - but you don't need a policy to say what's going to happen anyway. Victor Yus (talk) 09:36, 14 August 2013 (UTC)

Re: "People will always claim that their edits are supported by consensus..." Or (perhaps more commonly) people will claim that the other guy's edit is not supported by consensus. That puts the other person in the position of trying to "prove" that consensus actually does support his edit. In most consensus disputes, the reality is that neither side has a true consensus... and both sides need to back down from their "I'm right" attitude and and reach a compromise. Blueboar (talk) 12:43, 14 August 2013 (UTC)

Yes, and I suppose mostly they do (or else one side simply concedes that the other is right, or decides it's not worth arguing about). But what about when they can't, or won't, reach a compromise? Or when one side (arguably) does have a "true" consensus? Or when one side claims it doesn't need a consensus, because its position is the status quo or is dictated by policy, or something (and possibly then makes no effort to find a compromise)? Who decides these things? On what principles? By what process? The policy continues to lack any information on these essential matters, and hence remains woefully incomplete and of almost no practical use as a policy. Yet in spite of all our chat on the talk page, we don't seem to be making any effort to repair these deficiencies. Victor Yus (talk) 08:21, 15 August 2013 (UTC)

Involvement

The policy does not mention anything about necessity of not being involved/neutrality for editors/administrators who close the discussion. When there is no clear consensus between sides of discussion, involved editors should not attempt to close the discussion even if they believe their argument is much more stronger, in this situation it's better to leave this decisions to uninvolved editors who are familiar with rules of Wikipedia. -- Dalba 31 Mordad 1392/ 10:46, 22 August 2013 (UTC)

OK... first of all, very few consensus discussions are of the type that would require a formal "closer". Most consensus discussions are simply that... discussions. Anyone can end a discussion (just stop responding), and anyone can summarize what they believe the consensus to be. If they are wrong in their analysis, others will speak up and disagree. Blueboar (talk) 11:06, 22 August 2013 (UTC)

In situation I'm trying to picture, the other side has already spoken up and disagreed. For a discussion that needs formal closure and all sides insist on their positions, a third party's opinion would have more acceptance and less chance of being rejected by others (most probably based on WP:COI). This could be included as a reminder or suggestion to avoid such situation. Apart from this, I agree with what you are saying. Probably this topic is more related to Wikipedia:Closing discussions, but I thought maybe it was important enough to be included in this page, too. -- Dalba 31 Mordad 1392/ 12:31, 22 August 2013 (UTC)

Determining consensus section

On August 13, 2013 User:SmokeyJoe edited the section "Determining consensus" and removed the statement that says:

Consensus is determined by the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy.

This sentence has been the subject of extensive discussion on Wikipedia, particularly on this very talk page and I contend that it cannot be removed unilaterally unless Wikipedians as a whole reach consensus to do so.

Three minutes later that very same user changed the title of such section from "Determining consensus" to "Reading and interpreting an apparent consensus". I, once again, contend that such content cannot be unilaterally modified without Wikipedians reaching consensus beforehand to do so.

Per both WP:AGF, WP:BEBOLD, and WP:IAR, User:SmokeyJoe did not violate any policies and should not be subject to any disciplinary actions. However, per my contention, it is advised that this content is left as is until we reach a consensus. User:SmokeyJoe has been notified of this.

—Ahnoneemoos (talk) 01:38, 26 August 2013 (UTC)

I bring attention to the fact that "determine", like most words in English, has several meanings, one of which is "cause (something)", another of which is "ascertain". In fact, ascertain is a synonym of determine, and, I believe, the intended meaning of determine in this context. Would it be more clear if we changed "Determining consensus" to "Ascertaining consensus"? --B2C 01:42, 26 August 2013 (UTC)

We cannot use the word ascertain as it would imply that the outcome would be definitive, without doubt, certain or with assurance. Consensus is both volatile and flexible: one day we might reach consensus to do things a certain way and one month later we might reach a different and completely opposite consensus. I'd say that a better way would be "analyzing discussions for consensus" or something similar since per WP:IAR, WP:FIVEPILLARS, and WP:NOTBUREAUCRACY, our policies and consensual outcomes are not authoritative but descriptive. —Ahnoneemoos (talk) 02:20, 26 August 2013 (UTC)

Ascertain does not necessarily have a definitive or certain connotation - certainly no more than does determine. For example, "The appeals court ascertained that the law was unconstitutional; the supreme court reversed this determination." I don't see how this is a reasonable objection to the use of "ascertain" instead of "determine". --B2C 02:32, 26 August 2013 (UTC)

Ahnoneemoos Reverted my edits on the basis that my edits did not follow sufficient process, and does not seem to actually contend the edits.

Subsequently, B2C has significantly improved on the version reverted to. I agree with B2C, that "Determine" is easily read to say something not right, and that "ascertain" is a much better word.

However, I think my edits should be restored, and B2C's correct point should be referred to Wikipedia:Closing discussions. There is a big problem here of scope creep. Wikipedia:Consensus is concerned with the process of achieving consensus. Closing a discussion, such as a RfC or XfD, is NOT a consensus process. If it were, then every subsequent passer-by would be entitled to tweak the close, to "improve" it. No. The community decided in the early days that certain discussions needed to be time-limited, decided and closed, so that we can move on. This is why WP:Rough consensus is not WP:Consensus. Under WP:Consensus, arguments are weighted by the respect they are given by subsequent events, such as edits, or processes actually practiced. Instructions on how to close a discussion, including "How to ascertain a consensus", belong at Wikipedia:Closing discussions. --SmokeyJoe (talk) 02:22, 26 August 2013 (UTC)

Summarizing something that is explained in more detail elsewhere is not a WP:CREEP issue. WP:CREEP is concerned with creating new unnecessary rules. The issue of determining (or ascertaining) consensus is relevant enough to the topic of consensus to include a summary statement of what that means here. --B2C 02:32, 26 August 2013 (UTC)

I do not agree with your assessment. What we place on our policies is up to us. We have no rule that prohibits policies from including content on how to act in our internal processes. WP:CONSENSUS is a policy concerned with the subject of consensus and all its related matters, not merely on the process. Removing or moving such an integral part of our policy to an information page such as Wikipedia:Closing discussions which is NOT a policy, would be detrimental to the project as consensus must always be viewed through the lens of our policies. Removing such statement would give our editors the ability to apply essays, guidelines, information pages, and whatever the hell they want when we are attempting to reach consensus. Once again, nothing prohibits a policy from covering behavior or procedure. Finally, no, arguments are not weighted "by the respect they are given by subsequent events, such as edits, or processes actually practiced", no, arguments are weighted by their soundness and by how they are backed up by policy. Practices expand our vision but are not the determining factor on consensus, policies are. —Ahnoneemoos (talk) 02:36, 26 August 2013 (UTC)

I have been BOLD, and tried a different formulation entirely. Does this resolve the issue? Blueboar (talk) 02:31, 26 August 2013 (UTC)

It reflects community consensus as I understand it. It does not address SmokeyJoe's concern that this page should not address determining consensus at all, other than reference WP:Closing discussions. But I disagree with that view, as I explained above. --B2C 02:39, 26 August 2013 (UTC)

I too, was WP:BOLD, and tried to further expand your edits in good faith in order to make them more consonant with what consensus truly is. —Ahnoneemoos (talk) 03:04, 26 August 2013 (UTC)

I think some of the qualifications went too far, so I reverted that part, but retained much of it. And added that closers should be prepared to explain their reasoning futher [1]. --B2C 04:10, 26 August 2013 (UTC)

Thanks, I had a hard time putting in prose what I wanted to express but I think you did a good job. The only problem that I have is the phrasing, "what that consensus is." You see, in the real world consensus is something all participants agree to. In Wikipedia, however, consensus in a discussion is what the closer judges, not necessarily what the participants agreed to. We must express that in the policy. That what the closer judged is what he perceived to be the consensus but not necessarily what all the participants agreed to. —Ahnoneemoos (talk) 05:27, 26 August 2013 (UTC)

Isn't that implied. We already have that clarified at the top of the page: " Ideally, it arrives with an absence of objections, but often we must settle for as wide an agreement as can be reached." Do we need to restate this every time we reference "consensus"? --B2C 05:38, 26 August 2013 (UTC)

The text has substantially changed, but it remains full of problems, and it does not contribute to the purpose of this policy.

"When a consensus cannot be achieved ...". Who decides this? The paragrpah begins in dream space.

"It may be necessary to call". Weak passive tense. Who decides? When level does "necessary" rever to. Only rare extreme circumstances rise to necessity.

"An uninvolved editor to ascertain whether there is a consensus". A concensus is obvious for all to see, and is recognized and respected by consensus. Or is this referring to a rough consensus. Is this referring to closing process discussion? It doesn't say so. In general, this is attempting to describe a non-existent process.

"This editor will evaluate the comments made by others". Now it gets predictive. It is clearly trying to write paraphrasing of WP:Closing discussions into policy. Why, I don't know.

"Closers should give more weight". Loosely worded prescription. Closers giving weight has nothing to do with editing discussion and negotiation.

"quality" (bolded). Emphasizing this undefined word, for what purpose. It begs an explanation of how to measure "quality". Soon, someone will try to define debate quality into policy. No! This is nonsense.

"as viewed though the lens of Wikipedia policy" cloying or beautiful, this is poetry to people who understand it and of little help to anyone else, and flies in the face of normal definitions of consensus.

"Discussion closers should ..." This is an instruction to a closer. This falls squarely in the scope of WP:Closing discussions. It has nothing to do with finding consensus through editing.

This sloppy text does not belong. --SmokeyJoe (talk) 09:06, 26 August 2013 (UTC)

• OK... I need to repeat a point I raised a while ago... we are not really discussing "consensus" here... we are discussing dispute resolution. Most of the time consensus simply emerges organically...either through editing or through discussion. The only reason to call in a "closer" is when our normal methods of achieving consensus break down... when those trying to reach a consensus can not do so. At that point we turn to various dispute resolution procedures ... procedures that impose a consensus upon us (such as holding an RFC or RM discussion and having an official "closer" make a determination) whether we like it or not. Blueboar (talk) 15:35, 26 August 2013 (UTC)
  • Right, most of the time there is no need to "determine consensus". That's only needed when there is a dispute. For certain types of disputes (RM, AfD) we have a formal mechanism for setting up a discussion and retaining a closer. For ad hoc disputes, when the dispute cannot be resolved (and consensus achieved) through discussion, we use the RfC mechanism which widens the net and also eventually is closed by a closer. When there is a closer the goal is to determine a rough consensus out of the discussion. --B2C 16:46, 26 August 2013 (UTC)

On the filed RfC question

Consensus is determined by the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy.

This statement is true in respect to closing discussions, and poor advice with respect to editing. Moving it outside Section 1 is good. "Achieving consensus". It would be even better if it referred to closing discussions explicitly, and could not be selectively misquoted to refer to editing. --SmokeyJoe (talk) 02:11, 27 August 2013 (UTC)

I've just tried copy-editing this in the Finding Consensus preamble/summary, and given up. This quote falls decidedly on the side of instructing the closer to weigh by quality and interpret applicability of policy and to thus WP:Supervote. While it is true that arguments are weighed by their quality, and quality is convoluted/lensed with policy, it is not true that doing this produces "consensus", nor is it acceptable for a closer to do this a priori. --SmokeyJoe (talk) 05:00, 27 August 2013 (UTC)

Agreed. In my experience, Wikipedia consensus has always included the concept that, at the very least, all involved parties are able to live with a decision even if they may not entirely agree with it. If a dispute is occurring and a single party evaluates the arguments and comes down on one side or the other, we are unlikely to have consensus at all. That's more like arbitration. Omnedon (talk) 12:13, 27 August 2013 (UTC)

"Consensus is determined by the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy." Since this is the first sentence in the section "Determining consensus", it should be very precise. Instead, to me it is confused and clashes with lucid reasoning. I would say that the writers of this sentence wanted to impart that the participants to the consensus should pay attention to policy (don't violate it) and the arguments (accept the good ones). The metaphor of the lens is perfectly apt, but the rest gives me pause. First, we have the word 'determined'. Formally, if X is determined by Y, there would normally be a causal relation. (e.g., The thermostat determines the use of the furnace. Your test score is determined by your answers to the questions.) But there is not a causal relationship between the arguments and the consensus, except as mediated by the participants. Then there is the subject: "consensus". I think we are trying to talk about a consensus decision, not just "consensus", so this phrasing invites confusion. Consensus decisions are determined by the participants to the consensus. Their relationship to the quality of the arguments is that they should accept good arguments and reject bad ones. The arguments can't tell the participants what to think of them, as seems implied in the current text. The arguments from all participants should be considered. Proposal: Consensus decisions are determined by the participants' assessment of the quality of editors' arguments, viewed through the lens of Wikipedia policy. --Ring Cinema (talk) 20:53, 28 August 2013 (UTC)

Yes, the metaphor may be apt, but I'm undecided as to whether is it cloying or beautiful. Is it meaningful to people who don't already understand. I have doubt.

Another problem is that it is written in passive tense. Passive tense writing lends to imprecision. The sentence appears to prescribe how to determine consensus (a notion inherently flawed), but on careful reading it says very little. --SmokeyJoe (talk) 22:35, 28 August 2013 (UTC)

• Ahnoneemoos (talk) 03:18, 29 August 2013 "will disagree with any removal to the statement that "Consensus is determined by the quality of the arguments... as viewed through the lens of Wikipedia policy." Use the verb you want, but that sentence must remain intact."

This may be difficult.

"Consensus is determined by ...", in such bald simple terms, is wrong. It is a redefinition of "consensus" at odds with the preceding sections. It doesn't match reaching consensus through editing, where you don't persuade others by argument but instead by showing them a compromise that is jointly acceptable. And if it is really a perfect consensus result, they further edit the product in the next iteration towards consensus. Consensus, in general, is built by mutual learning, negotiation and concession. A pocket full of quality arguments against a rabble that doesn't believe them does not give a consensus.

Perhaps Ahnoneemoos doesn't mean in general, but some special cases?

Perhaps by consensus, Ahnoneemoos means rough consensus? In many situations, by policy, we respect a properly recognized rough consensus as a proper decision. Rough consensus is moderately well defined and used more often than just by administrators in deletion discussions.

How about:

"Where discussion participants who do not reach a common consensus, and a decision needs to be made, a discussion closer may declare a binding rough consensus on considering the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy"

--SmokeyJoe (talk) 04:01, 29 August 2013 (UTC)

Header for the section

OK... now we can't even agree on what to call the section. We've called it

• Determining consensus,
• Reading consensus
• Disputes over consensus
• Finding consensus

So let's discuss...
The rational for my suggestion (Disputes over consensus) is that this is really what the section is about... what to do when the normal consensus building process (a mix of editing and discussion) breaks down and we have a dispute. Why not call a spade a spade. Blueboar (talk) 00:28, 27 August 2013 (UTC)

I went with "Finding consensus" because I thought it was likely to address Smokey's concern with "determining" as well as still retain the original intended meaning of this section heading.

"Disputes over consensus" is inaccurate. The disputes are over what the title should be, whether to delete the article, whether to insert or remove certain material, what to change and how, what the section heading should be, etc. The disputes are not about whether there is consensus and what consensus is - which is what "Disputes over consensus" would mean.

What this section is about is how a closer is to find consensus when there is a dispute over what to do... hence Finding consensus. --B2C 00:46, 27 August 2013 (UTC)

Re: "The disputes are not about whether there is consensus and what consensus is"... I disagree. It think that is exactly what the dispute is about. When there is no dispute that a consensus exists, there is no need formal procedures such as RfC, AfD, RM... we just enact whatever that consensus may be. We only use formal procedures when we can't agree on whether there is a consensus, and if so what it is. Blueboar (talk) 01:16, 27 August 2013 (UTC)

I'll think about that. --B2C 01:41, 27 August 2013 (UTC)

In an RM, say a proposal to move A to B, you're suggesting we characterize the discussion to ultimately be about whether consensus supports the move or not. Those in support are arguing that consensus supports the move, while those opposing are arguing that consensus opposes the move.

But that's just not the way it is, is it? Some people point out how certain policies indicate A should be the title, while others argue that B should be the title for other reasons. The issue being debated is not whether there is a consensus supporting the move, but which title is better supported by policy and practice. Only after some time can it be determined if there is a consensus supporting either proposition. --B2C 02:28, 28 August 2013 (UTC)

"Finding consensus" is too much like "Deciding consensus", suggesting finality, which does not sit with (in) Wikipedia:Consensus#Achieving consensus.

Wikipedia:Consensus describes a wiki-editing process directed to improving though consensus (the real world meaning). In practice, consensus building means negotiation over the aims, the questions, the assumptions more than compromise on the answer. In closing a discussion, or determining whether, at a certain point, there is or is not a consensus, of some level, it is necessarily required that the question is locked in, meaning that the closing of discussions is not a subject to be considered within Wikipedia:Consensus#Achieving consensus.

The following sections, "Level of consensus" and "No consensus" are fine, but a section on closing discussions at subsection 1.4 is out of place.

If the section is on "Reading consensus" or "Disputes over consensus", and it is intended to help participants following the preceding sections of advice to recognize whether it is working, then I can see how it might be made to fit. While "Finding" may be better, it is still wrong because it implies completion of consensus building processes. If someone can find it, then it already exists, and the process stops.

While B2C and Blueboar are making positive contributions to the problematic section, and Ahnoneemoos writes sensible things on this talk page, the subsection is simply out of place. It currently is unclear as to what sort of discussions it is referring to (definitely not all), it can be read to say weird misleading things, and it feels to be motivated to include buzzy sentences and not motivated to provide advice to ordinary editors. To the extent that it refers to closing process discussions or dispute resolution, neither are within the scope of the page, both have other dedicated project pages. --SmokeyJoe (talk) 01:41, 27 August 2013 (UTC)

Why or how does "finding consensus" suggest finality? Finding what consensus is now has no bearing on what consensus might be 6 months or 6 years from now. --B2C 05:24, 27 August 2013 (UTC)

That was perhaps the least important of my points. I'd much rather hear an update on your 01:41, 27 August 2013 post.

Ahnoneemoos's solution left me feeling too wordy and unclever. Do people really want Wikipedia:Closing discussions to be anchored in WP:Consensus. I think Wikipedia:Closing discussions is, usually, an exception to WP:Consensus, but if people want it included I guess I can keep breathing. --SmokeyJoe (talk) 01:42, 28 August 2013 (UTC)

There are different strengths of consensus on WP, from unanimous to strong to weak to none. The strong to weak range is known as a rough consensus, but it's still consensus. That is, it's not "no consensus". Where consensus lies on that spectrum for a given issue is often a subjective matter itself - it is common for there not to be a definitive answer, but we need to have a mechanism to make a call, one way or other, in order to make decisions. When there is no unanimous or clearly strong consensus, and it's also not clearly "no consensus", we rely on an uninvolved party to evaluate the discussion and decide/determine/find whether there is consensus, and, if so, what it is. Isn't that aspect of consensus, and a summary of how that mechanism works, appropriate to describe on this page? --B2C 02:28, 28 August 2013 (UTC) change final statement to question. --B2C 03:09, 28 August 2013 (UTC)

Either there is consensus or there is none. If there is consensus, everyone involved can at least live with a decision. If you have parties on two sides that strongly disagree with each other, then there is no way this can be called consensus. Disregarding the views of some participants in favor of the views of other participants is not finding/identifying/determining consensus. Omnedon (talk) 12:28, 28 August 2013 (UTC)

If you mean "consensus" in the dictionary sense, you're right. But if you're referring to consensus on Wikipedia, this is utter nonsense. In many if not most RM and XfD decisions in which a closer finds consensus to support/delete or oppose/keep, there will be parties that strongly disagree.

Heck, in this very discussion we have one person who strongly disagrees with the inclusion of summary information about consensus finding in closing discussion on this policy page, but since all others involved favor its inclusion, we say there is consensus for its inclusion. --B2C 17:46, 28 August 2013 (UTC)

It's what consensus means. You are once again trying to use a term that has a meaning in the English language, and apply a different definition to it for use solely on Wikipedia, just as you have attempted with "concise". Do you honestly believe that everyone else agrees with you on what you're trying to do here? You calling the comments of others "utter nonsense" is certainly not conducive to consensus-building. And there you go again with "we" -- "we say" such-and-such. If there is disagreement, you can't validly make that claim. Omnedon (talk) 17:52, 28 August 2013 (UTC)

You're not talking about consensus here. You're talking about tallying points in an argument, and saying that the side with more perceived points "wins" the discussion. That's counter to the meaning of "consensus". Omnedon (talk) 17:57, 28 August 2013 (UTC)

Consensus on wikipedia does not mean everyone more or less agrees. wikipedia consensus is not the same as Quaker-style consensus, for example. But the word "consensus" is broad, and can be used to denote many things - but unanimity is not one of them. --Obi-Wan Kenobi (talk) 18:10, 28 August 2013 (UTC)

I have never equated "consensus" with "unanimity", nor do I believe that they are the same. I'll re-state once more: if you have a case where parties do not entirely agree, but where everyone can agree to live with a given decision, that might well be consensus. If you have strong disagreement, and no indication that all parties are willing or able to live with a given decision, then there is no consensus. It's these cases, where there is strong disagreement, that the term is mis-applied. Omnedon (talk) 19:08, 28 August 2013 (UTC)

Your definition of consensus may work for the Quakers, but it won't work here and has no meaning here. What does this mean: "All parties are willing or able to live with a given decision" - have you not seen any move reviews? Or other contentious decisions? We have to move forward, and sometimes that means making a call that some people disagree with, and will ALWAYS disagree with. The only way they "live" with it is people will block or ban them if they continue to fight against it. I don't think your definition can work here. Consensus is, especially for hard calls, a "rough" consensus, and something that most people can live with - but not ALL.--Obi-Wan Kenobi (talk) 21:30, 28 August 2013 (UTC)

I don't appreciate your tone here. Have I seen any move reviews? Indeed, such as the review of your incorrect closure on Hillary Rodham Clinton. And that's a good example -- where moves are concerned, we certainly do NOT have to move forward. If there is no consensus, we need NOT to move forward with the move. If there are editors that simply don't like it and cannot or will not express any reasons why, that's one thing; but if you are suggesting blocking or banning people because someone came in and made a supervote, and they don't like it, then that's a very dangerous approach. What you're describing is not consensus. Omnedon (talk) 21:47, 28 August 2013 (UTC)

The point is, you are trying to bring a definition of consensus that I've never seen applied to closes, on any topic. Especially titles, where it's X or Y (and no room for compromise between the two), someone is always unhappy with a move, but we have to make a decision anyway. Suppose 100 people !voted, and 99 !voted for X, and 1 !voted for Y, and the guy who !voted for Y will NEVER be satisfied with X as the title. But that is the consensus, that's what we call it here. I'm not suggesting blocking/banning ppl b/c someone made a supervote, I'm suggesting that when a decision has been made, if someone comes and tries to contest it and won't accept it, they will be blocked or banned (such as if they try to move war, or constantly re-open RMs, etc.). This happens, ALL THE TIME. We enforce consensus with violence, at times. Deal with it...--Obi-Wan Kenobi (talk) 22:07, 28 August 2013 (UTC)

The use of the word "consensus" is especially sloppy in RM closes and Move Review. Wikipedia:Move_review/Log/2013_August#Z.C3.BCrich is an example of an acceptable close that was definitely not consensus or WP:Consensus. That the use of "consensus" instead of rough consensus happens often doesn't make it right, or helpful. --SmokeyJoe (talk) 22:19, 28 August 2013 (UTC)

You are escalating here. Do not tell other editors to "deal with it". Please tone it down. In your 99 to 1 example, you are not describing a situation that would be likely to need outside closure. What you are describing is "voting". You even used the word "vote". Consensus is not about voting. Omnedon (talk) 22:16, 28 August 2013 (UTC)

I think you'll be fine. I know consensus is not about !voting, but numbers do count, and you'd be a fool to not consider them, especially if overwhelming. Again, you're speaking of some theoretical ideal world where every time we have a move discussion, we come to some sort of agreement everyone can get behind. I haven't seen that in reality, but articles MUST have titles, so decisions must be made. In article editing, you can always reword and reword and find a wording that everyone is happy with, but with article titles, it's usually a zero sum game. But I don't think it's useful to describe article title discussions as not being consensus - it's just a more rough form of consensus. Omnedon's assertion that everyone has to be willing to live with it is unworkable, for article titles, and for many other issues besides (e.g. how should the lede read, should there be an info box, etc). I guarantee you, no matter where Chelsea Manning ends up, there will be LOTS of people who aren't willing to live with it, but if they aren't, they will suffer the consequences... --Obi-Wan Kenobi (talk) 22:29, 28 August 2013 (UTC)

You are speaking of some sort of theoretical situation where 99 people vote yes and 1 person votes no and you call it consensus. If you are talking about simple majority rule, then call it that -- not consensus. But your militant attitude is not helpful to the process. "Suffer the consequences", indeed. Omnedon (talk) 02:13, 29 August 2013 (UTC)

Obi-Wan Kenobi: The cases I've seen where there's divided opinion on the subject of article titling tend not to be about how to apply a title to an article that's untitled, but rather about changing existing titles; that being the case, your assertion above that decisions (presumably other than those recognizing a lack of consensus) must be made because articles MUST have titles seems odd. (Obviously all articles have titles and one doesn't have to agree on any given change for that to be so.) The question is normally whether the current form or some alternative form is the more suitable... and in such cases if editors advancing reasonable arguments cannot agree that one form is clearly preferable to another, then policy favors the current form.

As for consensus, it's certainly different from unanimity (as I see Omnedon clearly notes above). A party who can't live with the result of a decision for no other reason than that he just doesn't like it has no recourse; however, it seems inappropriate for a decision other than "no consensus" to be imposed in situations where opposing parties disagree on reasonable and valid grounds. Speaking for myself, one of my concerns in such cases is with closers who go further and try to "pick a winner" where one has not clearly emerged. While I think it's sometimes well-intentioned, such an approach inappropriately magnifies the role of the closer into something closer to a super-voting arbiter, and diminishes the role of the participants in the discussion to mere counsels trying to convince a judge. ╠╣uw [talk] 02:39, 29 August 2013 (UTC)

• On What discussions get closed?

B2C wrote: "I'll leave "time-limited process discussions", but I'm not sure what exactly is meant by that" By Process discussions, see Category:Wikipedia processes, I am referring to discussions, like XfD, RM, etc. In general, discussions and disagreements do not get closed by uninvolved editors. We wouldn't want editors having mainspace editing disagreements to think that "per the WP:Consensus policy", they should ask an uninvolved editor to close their discussion. --SmokeyJoe (talk) 01:42, 28 August 2013 (UTC)

(ec)I have to agree with Smokey... the majority of our consensus discussions don't actually get "closed"... we simply edit and/or discuss until either a consensus is achieved or people realize that no consensus is likely (well... to be honest, discussions often go on for a while after that point... it can sometimes take a bit of time for everyone to accept the consensus, or lack there of.) Only a few, very specific types of discussions actually get "closed"... and I question whether it is really accurate to call them "consensus" discussions. They are more dispute resolution discussions that use consensus as part of their process. Blueboar (talk) 02:36, 28 August 2013 (UTC)

I also agree. I just think there has to be a better way of saying that than "time-limited process discussions". Now that I know the intended meaning, I'll try to come up with something. --B2C 03:06, 28 August 2013 (UTC)

Progress reverted

Over the last few days I felt a lot of progress was made on the section using consensus through editing as well as discussion here. Now, in one fell swoop all of that is reverted, despite almost no objections being stated here on the talk page. It is far less efficient and less productive to fully develop text on the talk page before adopting on the project page. In this case no changes to policy were at issue - only how to word the explanation best. This kind of effort benefits from an evolutionary approach, like we were all engaged in. I strongly object to the wholesale revert of all that work progress. --B2C 22:30, 28 August 2013 (UTC)

I submit that there is no consensus for the "Determining consensus" heading and preamble, and that it should not be in the policy. --SmokeyJoe (talk) 22:37, 28 August 2013 (UTC)

What is the basis for your submission? --B2C 22:48, 28 August 2013 (UTC)

That is seems to have crept in without much attention at the time.

That it has specific serious problems.

It does not enjoy consensus support, at this time. I compliment you on your productive input here, and think we will achieve something soon, but 'its been there for a while' type status quo arguments should be set aside. I think the inclusion of this text on closing discussions should be treated like a proposal to include new text. --SmokeyJoe (talk) 23:17, 28 August 2013 (UTC)

I gotta agree with Smokey here. I think the best course of action is to revert back to its previous state before Smokey edited it before we settle down for a specific text. Changes to this policy would have a substantial impact on the project as a whole since it's so intrinsic to everything that we do here. I can tell you however that I will disagree with any removal to the statement that "Consensus is determined by the quality of the arguments... as viewed through the lens of Wikipedia policy." Use the verb you want, but that sentence must remain intact. —Ahnoneemoos (talk) 03:18, 29 August 2013 (UTC)

Replied above at Wikipedia_talk:Consensus#On_the_filed_RfC_question. --SmokeyJoe (talk) 04:02, 29 August 2013 (UTC)

It seems to me that, as a policy, any changes to this page should be discussed extensively here before being moved into the mainspace. And I echo Smokey, that a revert does not signal a problem with the proposed change, but merely the necessary caution required to make such changes. Eusebeus (talk) 23:41, 28 August 2013 (UTC)

I agree that we have achieved some degree of progress on the section... but there is still a lot of disagreement (I am not even sure we agree on what the section is supposed to be about). I don't see a consensus... yet. This is why I agreed with Ring and reverted back to the "stable" version from a few weeks ago. Have patience. I think that with continued discussion we may yet reach a consensus. Then we can amend the section. Blueboar (talk) 00:02, 29 August 2013 (UTC)

My perception is that the changes proposed have not been considered with enough care. Introducing new terms is likely to have unpredictable consequences. Is there a problem with the current text? If so, it is not a major problem, since the policy seems to function adequately as it is. Are the changes proposed to the text supposed to leave the meaning of the policy unchanged? If so, I think they are not successful on that score. --Ring Cinema (talk) 03:45, 29 August 2013 (UTC)

This is what it is now, and where we started:

This is where we were before it was reverted:

There is nothing in the new/expanded version that is not already reflected in policy and/or practice. It's just more explanation. I would like to know a specific objection to this wording. --B2C 04:21, 29 August 2013 (UTC)

• It's befuddled. It does not indicate where applicable, and it is not generally applicable. It confuses WP:Consensus with WP:Rough consensus (just written). "Just more explanation" is called WP:CREEP, and this is definitely pushing the scope WP:Consensus. --SmokeyJoe (talk) 04:44, 29 August 2013 (UTC)
  • It's not perfect, but you really don't know what it means? It's applicable on RMs, XfDs, RfC, certain ANIs, etc. WP:Rough consensus is not separate or distinct from WP:Consensus... it is a valid degree of consensus on WP. WP:CREEP is adding more rules; this adds no new rules. --B2C 04:56, 29 August 2013 (UTC)

• I was unable to copy edit clarity into it. I think I know what was intended but am disturbed to see text that could be read as implying that there are consensus experts who go running around deciding closing discussions in mainspace.

Rough consensus is quite distinct from consensus. Consensus is what happened when participants work together, learn from each other, negotiate and concede, adopting a common position. Rough consenus is what you call an attempt to predict what the final consensus would be if the current discussion were allowed to continue. By consensus, it is agreed that certain types of discussion should not go on indefinitely, wasting editor resources, but that they will be subjected to (administrators, usually, tradtionally) best attempt of discerning a rough consenus.

Do we need to distinguish CREEP from BLOAT? Explanation that is redundant or adds zero or negatively to clarity is bloat.

I think we need to first define what the paragraph is talking about. Second, who it is talking to. Third, add it, and polish. We're still at step one. --SmokeyJoe (talk) 05:34, 29 August 2013 (UTC)

SmokeyJoe: Having examined the proposed additions, I tend to agree that the lengthier wording under consideration does feel like instruction creep. For my own part I don't see any compelling reason for why the current form actually requires any amendment. ╠╣uw [talk] 10:10, 29 August 2013 (UTC)

Thanks Huw, that's quite understandable. These discussions are in multiple directions. I think the only thread I'd like promote is at Wikipedia_talk:Consensus#On_the_filed_RfC_question, where we are talking about the line (currently): "Consensus is ascertained by the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy.", where I am thinking that it would be better to specify that the statement is in reference only to closing discussions, and modifying so as to not redefine "consensus" as a judgement. --SmokeyJoe (talk) 11:39, 29 August 2013 (UTC)

Here is how it looked in 2008[2]:

Smokey, note the reference to "determining consensus", and to consideration of "strength and quality of the arguments". These fundamental ideas have been in there a very long time. As far as I know, despite hundreds of rewordings, there have been no objections to these fundamentals, except from you. --B2C 05:14, 29 August 2013 (UTC)

That's not bad. Why was it changed?

I hadn't really looked at the "determining" until noticing that it had become a subsection heading in the section that begins "Editors usually reach consensus as a natural product of editing." It is much better with "Determining consensus" sitting low at level 2. But having looked directly at "determining", I, like you, see it is not a good word. Ascertain is much better. Reading is even more better. The paragraph is imperfect, in touching formal discussion closing without distinguishing from normal editing.

I'm not objecting to any fundamentals. "determining consensus" needs better wording and distinction from achieving consensus through editing. "Strength and quality of the arguments" is fine. I expect the "strength" disappeared because it caused so many to !vote "superstrongest possible support" style trying to score "strength" points I suppose. Quality is fine, though lacks definition, and it is not helped by bolding.

"Determining consensus" is a poor section title. Do you disagree? --SmokeyJoe (talk) 05:48, 29 August 2013 (UTC)

Yes, I disagree. I think "determining consensus" is a fine section title. It's a phrase that has been in this context for years. I've never seen anyone be puzzled by it, question it, or have any issue with it whatsoever, except you. But I'm also fine with "Finding consensus" or "Ascertaining consensus", so I have no objection to using either instead. I think they all mean essentially the same thing.

The strength of an argument has to do with how well it is grounded in policy and practice. --B2C 05:54, 29 August 2013 (UTC)

"The strength of an argument has to do with how well it is grounded in policy and practice". Never seeen that definition before. We have to watch out for you trying diminish anything not written into policy. You overstate the role of policy-as-written in decision making, and seem unconcerned that much of community is disconnected from policy-as-written, whether due to poor expression or excessive content. Diminishing accessibility due to excessive content is a reason why CREEP and BLOAT are reasons to oppose per se. --SmokeyJoe (talk) 06:04, 29 August 2013 (UTC)

The section - seen in big picture

I'd like to take a (hopefully brief) side trip... to examine the section as it relates to the rest of the policy, and ask a few conceptual questions that will better help us (or at least me) understand what the goal of the section under discussion actually is.
In the section/subsections that come before the one under discussion, we talk in terms of achieving consensus (and we outline two ways: achieving consensus through editing, and achieving consensus through discussion). But then we suddenly shift focus and terminology and start talking about "determining" consensus... so my first question is: why do we make this switch? Why are we discussing how to determine consensus at this particular point in the policy? (I'm not saying we shouldn't, just asking why we do). I suppose another way to ask this is: How is determining consensus different from achieving consensus? Blueboar (talk) 12:21, 29 August 2013 (UTC)

The words 'determining consensus' can be taken a couple different ways, of course. To determine something is to finalize it, and I don't think that's the intended meaning based on the single summary sentence that opens the section. Also, that sense seems to be covered above. To determine something means to understand it, and that is closer to the meaning intended, at least the way the section opens. (Unfortunately, that link 'Closing discussion' doesn't belong there.) So, I would say that we are suffering from the poor organization of the project. The section on achieving consensus -- which is about the nuts and bolts of it -- should be separated from the later section which it seems should advise the reader on the delimiters of their thinking about consensus as they discuss and evaluate proposals. --Ring Cinema (talk) 13:10, 29 August 2013 (UTC)

(Let me add that the section No Consensus seems out of place by this framing of the article. That section is about nuts and bolts, which belongs above, not in the "understanding consensus" section that should follow.)

Also, the section on pitfalls and errors is a good one but seems to be more about understanding consensus than doing it on Wikipedia. --Ring Cinema (talk) 13:24, 29 August 2013 (UTC)

Consensus is usually obviously there or it is not. Normally, we keep trying to achieve consensus until we have it. But sometimes we want a decision either way, even if we don't have a strong consensus, so we allow an uninvolved party, a closer, to make a determination about whether there is consensus on the issue in question, and, if there is, what that consensus is. It's an expedient way to get to consensus, if you will. --B2C 03:36, 30 August 2013 (UTC)

The previous comment doesn't seem to reply. I made the point that the two main sections of the article can be separated into two parts: the nuts and bolts of consensus on Wikipedia followed (loosely) by a section of advice on the parameters of consensus vis a vis proposals and discussion. Perhaps the article would benefit if we tried to tighten that organizational pattern, since at times it is inconsistent. I made two specific proposals regarding that overarching idea. --Ring Cinema (talk) 04:30, 30 August 2013 (UTC)

I think it would be a lot easier to manage this if we just created a draft and made edits there. In those cases, the edits and rewordings/etc become less contentious, as they are not *yet* part of policy - then we don't have to spend time describing what lines we've added, and can just add them and await a revert or tweak. rapid consensus often comes best through a series of edits.--Obi-Wan Kenobi (talk) 15:07, 30 August 2013 (UTC)

I'm unconvinced of the need for a lot of new edits. This specific issue with organization could improve clarity without changing the meaning. --Ring Cinema (talk) 11:22, 31 August 2013 (UTC)

Agree with RC. It is unclear why changes to the original wording are needed. As suchOnce/If the need for & scope of such changes have been established, a clearly formulated redraft here on Talk might help elucidate the degree to which changes are neededeventual formulation. Eusebeus (talk) 12:40, 31 August 2013 (UTC)

But to what end? Without an objective in view, what is the point of offering drafts or language? Unless the changes are on the margin -- but in that case the proposal can be made here as has been done many times before. --Ring Cinema (talk) 15:45, 31 August 2013 (UTC)

Edited my original comment for clarity. Sorry about that. Eusebeus (talk) 16:49, 31 August 2013 (UTC)

removal of analysis

I've removed the well sourced, undoubtedly true content [3] because this page documents an English Wikipedia policy, a widely accepted standard that all editors should normally follow. It's not the purpose to document how bad we are at it. NE Ent 02:38, 13 September 2013 (UTC)

I disagree with your removal. The policy does document best practices, but it can also acknowledge objective reality. We should provide some context on how our abstract principles hold up in actual practice. MastCell ^Talk 03:33, 13 September 2013 (UTC)

I support the removal. Policy should describe preferred practice. We could add a note that WP:Consensus describes a preferred ideal that is frequently not the day-to-day reality.

The material would go well in a WP:essay. It is very interesting and pertinent to the project. It is worth careful reading.

The material would not be suitable for simple addition to mainspace. It is a primary source and interpretation is non-trivial. It is not "undoubtedly true". Human behaviour is hard enough by close personal inspection. Human interactions are confounded when forced through written-only media such as this. Their study was written-only, and automated. So much detail was averaged. They went looking for conflict and found it. Confirmation bias. They excluded 93% of articles to do this. They assumed a dubious measure of consensus, even if they did have a good method to make these measurements. An alternative interpretation is that the researchers focused on articles that were and are far from having achieved consensus. Has an article achieved consensus? That's an extremely complicated question, I doubt that a start has been made on achieving a consensus on what the precise question is. --SmokeyJoe (talk) 04:50, 13 September 2013 (UTC)

I'm not wedded to keeping this material on this particular page, but I think it's important for it to be somewhere relatively accessible. Ideally, we'd even acknowledge that severe problems persist at only a tiny fraction of articles, but that if you choose to edit them, you should realistically expect problems. WhatamIdoing (talk) 05:14, 13 September 2013 (UTC)

There might be a way to justifiably mention it, and link the reference, in a footnote. We should take care to not allow it to ever be misread as implicitly approving the behavior that it discusses. Or should an essay: WP:External perspective of Wikipedia consensus building be written? --SmokeyJoe (talk) 05:25, 13 September 2013 (UTC)

"a widely accepted standard that all editors should normally follow"

How can a policy be truly said to be "a widely accepted standard that all editors should normally follow" if there isn't a consensus in support of it? According to this page, if I add something to an official policy and there is no consensus to remove it, then it stays. That does not seem right to me. Removing policy points that cannot get consensus support allows for particular disputes to be governed by the particular merits of the case, I don't believe someone should be able to pull a trump card in those cases that says "but policy says..." when that policy point does not have consensus support.--Brian Dell (talk) 16:14, 3 September 2013 (UTC)

To the contrary, a change to a policy requires consensus to stay, otherwise it is removed. --Ring Cinema (talk) 18:01, 3 September 2013 (UTC)

If that were true there would be explicit instructions to get policy page edits pre-approved on Talk pages before attempting them. Anybody could then cite those instructions in order to remove material that was added without discussion. The true situation is that an addition to a policy requires nothing to stay unless there is an immediate edit war over the addition in which case there is a determination as to which side of the dispute is closer to consensus. If the party trying to roll back the addition can be successfully portrayed as the party trying to "make a change," then in the event of "no consensus" the disputed additional material remains. That's the way it is now.--Brian Dell (talk) 10:51, 4 September 2013 (UTC)

However, policies are like all other Wiki materials: they can also be changed by silent consensus. If a change is boldly made without discussion and it isn't (reasonably) immediately removed, then it becomes part of policy per this section of the policy policy (not a typo). The longer it's there without being removed, the harder it ought to be to remove it, but even long-standing changes to policy can be easily reversed if they're dramatically not in keeping with the current thinking of the community (because it shouldn't be hard to get a "loud" consensus to revert them). Regards, TransporterMan (TALK) 18:34, 3 September 2013 (UTC)

I think it ought to be possible to remove from policy material that is "not in keeping with the current thinking of the community" even if not "dramatically" so. It is not especially difficult to get a "no consensus" result if one wants controversial material retained in a policy. As it stands now, that "no consensus" result means the controversial policy remains. If there is a call to remove something and that gets a far wider discussion than occurred when the material was added, it is dubious to argue that the community supports inclusion if the result for the much more extensive discussion was "no consensus." The Chelsea Manning article reverted to Bradley Manning because the "no consensus" default there was to flip back to Bradley Manning. But while at least some sources were using "Chelsea," the Manual of Style said it doesn't matter if there are NO supporting sources beyond merely verifying the subject's wishes. A call to require at least some third party recognition of a gender change in the Manual of Style was rejected because the "no consensus" default there was to keep the policy as written. It strikes me as a contradictory result (the unlimited deference to the article subject re his/her gender identity in the MoS is upheld while even limited deference to the article subject at Chelsea Manning is overturned in favour of not deferring) and it came about because the MoS was originally changed at a time when there was relatively little scrutiny whereas the bio isn't changed because of the current high scrutiny. This essay agrees with me that "If the discussion is about a section already in the policy, that section should be removed. Policy and guideline should reflect consensus. If there is no consensus as to existing policy, then it no longer reflects that and should be removed." I disagree re "policies are like all other Wiki materials" because policies are cited to constrain editor discretion, absent a clear and current consensus to deny editors their discretion they should be allowed their discretion.--Brian Dell (talk) 07:22, 4 September 2013 (UTC)

No, Brian Dell, there would not have to be instructions of the kind you mention. Normal editing of all too, I think. Removing material is a change, so it should require consensus to remove it. I believe we are in agreement that this is the only sensible way to proceed anyway. Perhaps there is some difference of opinion about what the policy says? --Ring Cinema (talk) 00:42, 5 pages follows the procedure that change requires consensus to be accepted. That applies to Bdell's scenario, September 2013 (UTC)

Policy pages ought to be treated differently. Note that WP:CREEP refers to "policies and guidelines" as opposed to the size of the encyclopedia.--Brian Dell (talk) 03:59, 5 September 2013 (UTC)

I can't think of a reason to treat policy pages differently but of course I'm open to a good argument. --Ring Cinema (talk) 04:44, 5 September 2013 (UTC)

@Ring Cinema: I mentioned instructions? I'm confused... — TransporterMan (TALK) 13:14, 5 September 2013 (UTC)

I've done a lot of policy and guideline work over the years. This system actually works for us in practice. You really don't want people to make obvious improvements (e.g., resolving self-contradictions or explaining a procedure in plain English) and have that reverted because "Nyah, nyah, you didn't say Mother, may I? first".

We don't actually have much of a problem with the first-mover issue on policy pages. There is no prejudice against changes, which is importantly different than a prejudice against additions. I can boldly make a change that I am certain improves a policy page. You can, just as boldly, make a change that you are certain improves the same policy page. (Perhaps your change will happen to be the opposite of mine.) This system tends to somewhat favor up-to-date, ideal policies at the expense of stability.

However, once we're past the point of bold editing, any change ultimately requires a consensus, regardless of whether that's a change to add or to remove. This system tends to favor stability over ideal, up-to-date policies. Thus the longer approach tends to balance the bolder one. It is rare for the result to be a policy that is seriously out of step with the overall community viewpoint. In most cases, the discrepancy is merely one of overstating: something is presented as having substantial community support, when the community is perhaps ambivalent about it. This is another situation in which it's helpful to remember that WP:There is no deadline. The community will changes its policies and guidelines when it eventually comes to a decision on the disputed point. WhatamIdoing (talk) 05:27, 13 September 2013 (UTC)

Damn, I wish I had said that. Excellent analysis, WhatamIdoing, simply excellent. Kudos. Best regards, TransporterMan (TALK) 13:12, 13 September 2013 (UTC)

Another opinion

I have always wondered why WP does not take its lead from the millennia of experience the world has in conflict resolution. In nearly all civilised countries there are two basic forms in which conflicts are ultimately resolved, Courts and Votes.

Here we have a system that is more like voting in a failed state. People state their opinions and cast their votes then an arbitrary closer come along, looks at the votes, and then uses their own judgement to decide who wins. Nowhere else in the civilised world is a conflict resolved this way. Martin Hogbin (talk) 12:18, 14 September 2013 (UTC)

• Are you talking about (1) building and finding consensus, or (2) closing discussions on proposals, or (3) dispute resolution? I think Wikipedia does (1) fairly well in most places, (2) pretty well for a non-hierarchical decision making system, (3) poorly, and very different to real world workplace grievance resolution. --SmokeyJoe (talk) 13:02, 14 September 2013 (UTC)

(ec)Problems: First of all, absent a mechanism to hold a vote representing many editors, most discussions end up involving only three to six editors -- the concept of a "vote" in such a subset fails. Second, the Arbitration Committee specifically considers all content issues outside of its remit - thus the "court" suggestion fails. Third, any content dispute system which fails to abide by the major policies (NPOV and BLP among them) will then face the prospect of violating decisions made by much larger groups of editors. The noticeboards are where we have the rough equivalent of local magistrates - which generally works fairly well, as they attract people based on the policies involved and not primarily on the content issues. Collect (talk) 13:23, 14 September 2013 (UTC)

Thanks for your replies. I guess I am talking about SmokeyJoe's 2 and 3. I agree that the system generally works well for 1.

To give an extreme example just to explain what I mean (and I really, absolutely do not want to discuss this subject here) consider an RfC on whether there is a God or not (I know this is completely unrealistic but bear with me for a while). Many editors comment on the subject giving many well known arguments on the subject, some even claiming to be proofs. Along comes a closing admin, he happens to be an atheist, which set of arguments are they going to find has the most weight?

More realistically, there are many contentious highly polarised questions that might come up in an RfC, often as NPOV questions. Is this position NPOV? No matter what the arguments are put by the responders, the final outcome is likely to be swayed by the personal opinion of the closer, whi generally is whoever happens to be passing by.

My proposed solution to this would be to say that the closer should just count votes unless they can come up with a really exceptional reason not to, and that any decision not to go with the majority should be open to some further appeal process.

Collect, regarding your comment, '...the Arbitration Committee specifically considers all content issues outside of its remit - thus the "court" suggestion fails'. Yes, I know, that is the problem. There is no way to get a final decision on a content dispute. The policies do not help for two reasons, firstly, they are written by the same people who are supposed to follow them, in civilised countries it is generally considered a good thing to separate the law makers from he law enforcers. Secondly interpretation of a policy, like NPOV, is everything. The policy is fine, but who decides what exactly it means. Martin Hogbin (talk) 16:01, 14 September 2013 (UTC)

Juries make good decisions. While practices differ between jurisdictions, the classic US system calls for unanimity of 12 non-experts who have seen all the evidence. That method steals something from consensus but pushes it to unanimity in exchange for limiting participation to those selected by lottery. It's a pretty good method to decide a difficult, one time case. --Ring Cinema (talk) 01:51, 15 September 2013 (UTC)

Juries are used mainly where a person is accused in wrongdoing of some kind. In criminal cases most countries have a high burden of proof, usually 'beyond reasonable doubt'. Juries might be useful in WP to arbitrate on permanent blocks but I do not think they have any relevance to content disputes. Martin Hogbin (talk) 08:24, 15 September 2013 (UTC)

Actually, juries are used in civil matters as well. Usually they only require a majority vote. Read the Seventh Amendment to the United States Constitution for US law. Collect (talk) 13:26, 15 September 2013 (UTC)

You're talking about US law but around the world there are many configurations of juries, and the classic case, as I said, requires unanimity. I don't quite see the point of Martin's comment. The idea that any of this section has relevance to a WP content dispute is moot in light of the fact that we use consensus or admins. Loosely speaking, this is like leavening mob rule with non-expert arbitrary opinion. But if we want to discuss a fantasy about how we could resolve content disputes, a randomly chosen group of a dozen experts or non-experts empowered to choose between a limited set of options has much to be said for it. Of course, in the case of juries, sometimes it takes sequestration to force them to reach a verdict and that doesn't seem so plausible. --Ring Cinema (talk) 21:16, 15 September 2013 (UTC)

I was not proposing a jury, just pointing out that we currently have what you called, '... mob rule with non-expert arbitrary opinion'. Martin Hogbin (talk) 22:50, 15 September 2013 (UTC)

I think Martin Hogbin makes an excellent point, and it's also a point that you never hear expressed. When it is expressed, people can't seem to wrap their heads around it. Consensus is a terrible way to prove the truth of things -- things ought to be proven just like in a court of law: Prove your point with evidence. But of course "consensus" is all over Wikipedia -- it seems to be one of the basic tenets. So nobody speaks critically of the idea, or ever suggests that there is a "dark side" to the concept, so to speak. Of course, the idea that the world is flat was a consensus. Maybe someday Wiki will see the light. — Preceding unsigned comment added by 68.196.192.105 (talk) 23:10, 6 October 2013 (UTC)

Section arguably unjustifiably removed

Several years (!) ago, a section that I think made some extremely important points was removed from this policy in this edit: [4] I think it should be restored. Dogmaticeclectic (talk) 22:34, 23 October 2013 (UTC)

The relevant discussions surrounding that edit can be found in Wikipedia talk:Consensus/Archive 3. Blueboar (talk) 23:44, 23 October 2013 (UTC)

Section? Dogmaticeclectic (talk) 05:59, 24 October 2013 (UTC)

link to essay on sham consensus

I propose to link WP:Sham consensus, an essay, from this policy's See Also section. In the last 90 days, as of a few minutes ago, it had had 779 visits, of which probably only a couple were from me. About two weeks ago, I tried to count how many pages linked to the essay but the counts were inconclusive (I temporarily commented the essay out of a navigation template and four days later the count had dropped from 401 to 241 without transclusions and redirects before I reverted the commenting out; 241 still seems to be a lot). The essay is needed for naming a local consensus which, for example, concludes that one article should be filled with original research, when doing so would violate policy; such a consensus is a sham consensus, not to be followed. Previous discussions were at WT:CON Essays on Wrongful Consensus and False Consensus, WT:CON Sham Consensus, and WT:CON Procedurally Flawed Consensus. I'll wait a week for any response. Nick Levinson (talk) 19:18, 6 October 2013 (UTC)

Please don't. At least not yet.

Nick, I intentionally read your essay before I fully read your post here... and I have to tell you that I came away clueless and confused as to what a "sham consensus" actually is. It was only on returning to this page, and reading your explanation here that I began to understand what you are talking about (and even then, I have enough questions that I am not sure I completely understand and agree with your view). An editor coming cold to your essay would still be confused.

So... My take is that it is premature to link the essay at this point. You need to go back and work on the essay some more. Flesh it out... Explain the term in more detail, give some examples, etc. Then bring it back for a second look.

Also... I would like it if other editors made some edits to your essay. As is, it is really just a statement of one person's view (your's), and I am not sure we should link essays that reflect only one person's views. It strikes me as somewhat UNDUE. Blueboar (talk) 22:23, 6 October 2013 (UTC)

I don't mind other editors editing and I can continue doing so. A sense that I got last time was that we shouldn't name the phenomenon at all and yet it occurs and needs to be talked about concisely, so it needs a name, and the essay helps to establish a name for the phenomenon. Another issue was whether naming is useful unless we also provide a remedy, whereas remedies already exist and so the essay does not need to provide any. If new remedies are needed, I don't think anyone has said so or suggested any. And another issue was whether we should state an exhaustive list of exactly what would make a consensus one not to be relied upon, and that would make policies, guidelines, and ArbCom decisions unenforceable, since any provision in a policy, guideline, or ArbCom decision would have to be copied first into the essay (presumably while upgrading it into a policy or guideline itself), thereby increasing the maintenance burden by rather uselessly requiring double sets of edits now and in the future. The term should be copied into a policy or a guideline (although some would say it's too soon for that), and the essay might benefit from examples, but the essay should not define all of the things that can make a consensus a sham, because it is the defining of them that would double the maintenance burden, a doubling that would likely be unwelcome by editors at most policies and guidelines and likely refused, especially by ArbCom. Generally, essays in the Wikipedia namespace are not supposed to contradict policies or guidelines (they can, more or less, in the user namespace, but this essay is not in the user namespace). I've also just seen the debate at Wikipedia:Village pump (policy)/Archive 100#Linking to essays on Policy and Guideline pages, in which "a few essay[s] that are for 'further reading', which explain and are fully consistent with the intent of the policy/guideline should be welcome" appears to be the most applicable answer, was not disagreed with, and supports linking this essay, even as is and regardless of the number of editors of the essay. I'm not sure how to expand the essay without risking examples essentially overtaking the definition, but I'll think about it and am open to ideas. Phrasing like "includes but not limited to", at least in U.S. law, does not mean what people usually think it means, because it actually limits what is included. But let's try things out and see what comes of them. Nick Levinson (talk) 16:44, 7 October 2013 (UTC)

I edited Wikipedia:Sham consensus. I added the OR example to Wikipedia:Wrongful consensus because a wrongful consensus is included in a sham consensus, by definition, and therefore it would undermine things to put the example into the latter. But, hopefully, both essays are now clearer. I also added about remedies.

Are you concerned that by naming a type of consensus on which we should not rely we would be defeating the kind of consensus on which we should rely, which is the more usual kind of consensus? I don't think it would, but I addressed that in the essay on sham consensus, so as to make that clear to readers.

Please let me know what other edits you think are needed.

Nick Levinson (talk) 16:46, 8 October 2013 (UTC)

I think I have an idea where you want to go with this, Nick, but I have problems with your essay, too. It seems like the first statement of the definition is actually incorrect, incomplete and uses special terms for no real reason. Plus, it seems to say that policies and guidelines always trump the editors on a page. That's very incorrect for guidelines (they are guides, not rules) and, for reasons we have discussed here before, policies are malleable according to the requirements of a particular page. That is a complicated subject. Now, is there a true, proper, and correct concept behind what you write? To me, the correct part of your thinking is that if deceptive activity yields or blocks a consensus, there is a problem for WP because we rely on a semblance of transparency in editorial discussions. --Ring Cinema (talk) 23:56, 9 October 2013 (UTC)

The MOS is a guideline; if a local consensus is to discard the entire MOS and invent a new style without a compelling reason justified mainly by the needs of that article, then that consensus should not override MOS consensus. At the top of the MOS page, it says that as a guideline "it will have occasional exceptions", but a wholesale exceptionalism even due to one article's consensus is almost certainly going too far. The inability of an article's consensus to override a guideline's consensus is an already established policy in Wikipedia:Consensus#Level of consensus.

Exceptions to policies and guidelines exist and can be created at any time. Since Wikipedia already accepts the possibility of exceptions, not every consensus arriving at an exception could violate a policy or guideline and thus such a consensus would not be a sham consensus, since a sham consensus is only one which violates the policy or guideline. If there's no violation, the consensus is not a sham.

Deception is not the only reason that requires that a consensus not be relied upon. Openly demanding that copyright be infringed is no more valid than covertly demanding it. If substantive error is not recognized as voiding a consensus, then policies and guidelines are meaningless on substantive issues. For example, Wikipedia:Do not create hoaxes could be ignored because hoaxes can be fun to create and propagate and WP:Notability could be ignored so we could write articles about who-knows-what and keep them up. Both are content guidelines (one by its own terms and the other per a template).

In Wikipedia:Sham consensus, the first statement uses two terms that are both linked to essays defining them and other than that does not use any special terms. It uses those two terms because they are part of the definition I am offering, no one has produced a better definition yet, linking to the essays defining them is more efficient than copying the definitions over, better terms are already taken, and we need a term for what we periodically talk about across Wikipedia so we don't need to use dozens or thousands of words of running text where a name will refer to the same thing far more efficiently, in the same way that you don't need to craft a lengthy description of you or me when all of us have names (including usernames) for ourselves when names are more efficient. If the definition is incomplete, please add what's needed. The definition is not incorrect unless you have another definition for the term to which to compare it to; I haven't seen that anywhere and no one has cited or presented one.

There still seems to be an effort to prevent naming what already happens. Could someone please explain why naming is per se objectionable? Is it that local consensus should be allowed to trump higher-level consensus? Should that not be addressed by transparently editing the Consensus policy itself? Otherwise, shouldn't we accept the Consensus policy as it is and just name the phenomenon? When I brought this up months ago, it was to amend the policy to include the name; I'd prefer that, but this time it is only to link to the essay, and other essays can also be linked to, if there is a debate on the point and someone wants to write an essay in another direction. I don't mind editing this essay or anyone else doing so, but we can hardly ignore that periodically there is a consensus we shouldn't use and we need to call it something.

Nick Levinson (talk) 16:03, 10 October 2013 (UTC) (Deleted a redundant sentence and corrected punctuation: 16:10, 10 October 2013 (UTC))

As far as naming goes, we are all free to use English to name things to our hearts' content. Without anyone writing a word, we all have an idea about the meaning of 'sham consensus', etc., and more definitions don't always yield clarity. As to the hierarchy of rule, it is not settled when a policy trumps a local consensus, for reasons that have been mentioned before. Most obviously, since policies reflect editorial practice, local consensus that ignores a policy may be construed as a change in policy or a proposal to do so. The notion that a single editor would claim, per an effort to countermand what they believe is a sham consensus, a consensus of editors who have a different idea about what is best -- for any number of reasons -- is not a good idea and invites more mischief than anything else. The best way to avoid that kind of mischief is, in my view, to simply accept that when there is deceptive behavior, a new consensus can form around correcting the deception. --Ring Cinema (talk) 17:18, 10 October 2013 (UTC)

Okay, we can use the term. Now I'd like to arrive at a definition on which we agree, if we're not there yet and so that when we use the term within Wikipedia's processes we can expect to mean the same definition.

Claiming that a consensus is a sham would raise a matter to be considered, but the claim without evidence would not normally cause an adverse result, just as claiming that an article has original research without pointing to specific content that is OR when sourcing is listed would not normally cause content to be deleted. Any policy or guideline can be misused by someone claiming a violation without evidence, but we don't erase all policies and guidelines because of that risk. Instead, we consider claims and, if someone is essentially crying wolf by repeatedly abusing process we advise them to stop, and block them if necessary. We may even add detail to policies and guidelines. A problem now is that an editor can claim that consensus exists when it is only a sham and revert a proper edit, such as one tagging or removing OR. An agreed-upon definition of sham consensus would help alleviate that problem.

The relationship between levels of consensus is indeed pretty well settled by WP:Consensus, so that exceptions are only exceptions and are not the norm. It is true that an article can give rise to a change in policy, but usually it would not, as an attempt to do so would often lead to reverting or editing to conform to policy or, with a rarely-read article, would go unnoticed at the policy level, thus leading to no change in the policy. Indeed, policies and guidelines point the way to acceptable editing, so that most articles tend to follow policies and guidelines, not lead policy/guideline development, with occasional exceptions when articles do lead the way.

Coping with deception should not be limited to trying to form a new consensus. The same deception that skewed the consensus would often continue to skew it. For example, sockpuppetry perhaps should be dealt with by sanctioning a misused sockpuppet.

Do you believe that substantive issues that cause a consensus to be a sham are all by deception, for example, that a consensus to ignore a lack of notability or to use only original research, even if the consensus is openly arrived at without a procedural violation, is a deception because it is deceptive in allowing such a major violation of policy or guideline? I can see why someone might, and if that's the case then I might clarify the essay to include that interpretation, although on the face of it deceptiveness appears to have a narrower meaning that is about process and not substance. Please let me know.

Nick Levinson (talk) 14:42, 11 October 2013 (UTC)

Nick... above you give the example: "The MOS is a guideline; if a local consensus is to discard the entire MOS and invent a new style without a compelling reason justified mainly by the needs of that article, then that consensus should not override MOS consensus."... I don't disagree... but... I think what is missing from your essay is what to do what to do when there is (at least potentially) a "compelling reason justified mainly by the needs of that article".

That's exactly the type of situations where local consensus might conflict with "wikiwide" consensus (as expressed in Policy and Guidelines)... and this situation is exactly why we invented WP:Ignore all rules. What makes WP:IGNORE situations difficult, of course, is that editors may disagree over whether the reason for ignoring the MOS (or any other guideline) actually is "compelling" or not. What is compelling to one editor may not be compelling to another. Thankfully, we have a system in place to deal with this... we seek a consensus from a broader audience (by filing an RfC for example).

But... occasionally, there is agreement among local editors... they know the subject matter and agree that the reason to IGNORE is compelling. In this case, the local consensus is not a sham... it is valid even though it goes against policy or guidance. Your essay needs to make this clearer. Blueboar (talk) 15:46, 11 October 2013 (UTC)

What to do when a local consensus is better than a policy or a guideline sounds like it belongs in the Consensus policy itself, if it's not already there and/or in another policy. I'll poke around. This essay is meant as just a definition, just like we have dictionaries that don't advance national policies but mainly define the words we use and leave policymaking to legislators and others, but what you suggest could go into its own essay, if it doesn't fit something higher. WP:IAR is potentially a disaster and is mainly ignored; most edits and revertings strive to conform to already stated policies and guidelines other than IAR. RfC goes up the path that includes policy consensus, and it's not necessary to rely on a local consensus only until ArbCom tosses it out. Dispute resolution already addresses what to do; this is just about agreeing on terminology. I don't propose to create a new process, as I don't think we need one and I don't think anyone argues we need one. It may be that the concern you raise can be resolved by crosslinking more between WP:Consensus and other policies and help pages, for example. I'll probably look at stuff this weekend. Nick Levinson (talk) 17:02, 11 October 2013 (UTC) (Corrected misspelling: 17:06, 11 October 2013 (UTC))

What is the purpose of offering this definition? There is no magic to a definition; in fact, attempts at definition can as easily do harm as good. If something happens in the editorial process that is fraudulent, deceptive, illicit, underhanded, or bogus, there are plenty of available words in English to say what it is. I don't think, Nick, you are actually offering a definition as much as you are offering to name something. The definition is to say what you are naming. Now, you could have chosen a relatively neutral term, like 'green consensus' or something positive like 'bold consensus' (is that positive?). Instead, you chose something pejorative. You are saying by the term that this is something bad, it's a sham. Yet, the cases under the predicate of the term are sometimes not bad at all. So, it's not "just a definition". Your essay covers quite a bit more territory than that. --Ring Cinema (talk) 19:10, 11 October 2013 (UTC)

The purpose is to offer a term to describe what otherwise would require quite some length of prose to describe every time, that being the purpose for which most words and their meanings exist. If we can use a variety of terms, we can agree on one, as we do for many phenomena affecting Wikipedia, without stopping people from intentionally using any of a variety of other terms, such as when they wish to make a distinction (such as between wrongful and false consensus). It's hard to think of an instance of defining as doing harm unless silence is the goal or the definition is wrong; silence in this case would be harmful and across English the solution to a bad definition is usually to add a definition (dictionaries are not usually banned and many common words have multiple definitions).

Negativity is precisely the purpose of this particular definition because that is what has to be talked about from time to time. There is no need for a new neutral term; we already have consensus as a neutral or positive term (and I'm not sure what bold consensus would mean, unless it is the 'implicit consensus that results from boldly editing'). We need a term for a consensus gone bad or that was prevented for a bad reason.

The concern you raise about possible harm can be dealt with while still having and using a negative term. Suppose someone wants to create a POV article despite its being against policy and does not justify it within WP:IAR. Suppose a consensus supports going POV. That consensus is a sham and therefore should not be relied on because that particular consensus is against policy. But then the policy is changed, an exception is created, or the matter goes up the decision-making chain and the article's POV is accepted by a Wikipedia-wide consensus, wider than the consensus for the policy. Then the sham consensus would no longer be a sham. Consider an analogy to a murder called such by a medical examiner that is followed by a trial where it is determined that the death was accidental; what was named a murder no longer is. Consider a soldier who leaves the army without permission; in the U.S., at least in the past, some were classified as deserters until it was later determined that they went AWOL but were not deserters. In the case of the POV against policy, at the point in time when it is against policy the consensus for it is a sham but when the policy is changed or the exception is created the consensus is no longer a sham and if the highest-level consensus replaces the sham the new consensus is not a sham. The murder is no longer a murder; the desertion no longer that; and the sham consensus becomes simply a consensus and not a sham. The position that we can't be allowed to agree on a defined term for a negative phenomenon is tantamount to a position that we can't refer to the negative phenomenon or that we can't point out its negativity. Yet Wikipedia already recognizes negativities with defined terms, such as hoax, and a hoax would violate a guideline right now. We can have defined terms even if they're negative. The defined terms themselves are helpful to discourse. I'll clarify in the essay that consensus can change from a sham to not a sham.

Nick Levinson (talk) 19:44, 12 October 2013 (UTC)

I've edited the WP:Sham consensus essay. Please edit or offer any other suggestions you think might improve it. Nick Levinson (talk) 21:08, 12 October 2013 (UTC)

Nick, let's explore the example you give above, because I have serious reservations.

• You postulate: "Suppose someone wants to create a POV article despite its being against policy and does not justify it within WP:IAR. Suppose a consensus supports going POV. That consensus is a sham and therefore should not be relied on because that particular consensus is against policy."

OK... first question is: Who says its a POV article? Editors can legitimately disagree on that issue. For example, let us assume seven editors working on the article... five of them reach a consensus that a particular viewpoint is fringe, and thus does not merit being discussed (per UNDUE)... two editors disagree, arguing that the viewpoint is not fringe, and that excluding the viewpoint violates NPOV. The five who support omitting the viewpoint will say "Consensus says the viewpoint is fringe"... the two dissenters will point to your essay and "No, that is a WP:Sham consensus". So, who is right? Was it a legitimate consensus, or a sham?

My take is that it was a legitimate (but limited) local consensus... and like all (limited) local consensuses, the question is whether it will be confirmed or overturned if presented to a wider audience. We have to remember that no consensus is ever permanent. Consensus can change. So... let's assume that the dissenters filed an RFC, asking whether the viewpoint is fringe or not. Let's assume twenty new editors respond, and all say "Nope... The viewpoint is not fringe". This result does not mean that the old consensus saying "fringe" was a sham... what it means is that we now have a NEW wider consensus (saying "not fringe - mention it") which replaces the old (local) consensus. Blueboar (talk) 14:24, 13 October 2013 (UTC)

That appears to assume that if an article is POV (assuming it is) but a consensus supports keeping it as is, then we can't edit it to be NPOV because that would be against consensus, the solutions being limited to persuading some of those who participate in the local consensus to change their minds or going up the chain of decision-making before we can edit, in which case the decision-makers above would not normally be able to see what that edit would say and we would increase the number of cases for the upper decision-makers, burdening them. Wikipedia does not work that way.

Of course editors legitimately disagree on many occasions about many things. That is not ground for denying we can discuss them, including naming them. Fringe is a name. We apply it. We don't need to have a 20-word description that means 'fringe' because we have that very word and we have agreement on a definition of it. Likewise for due weight.

On who says it's a POV article: That's what discussion is for. And the discussion is a lot easier because we have the terminology, in that case POV, with an agreed-upon definition, so that we don't simultaneously and in one place discuss whether the article has a point of view, what a point of view is, and why being neutral should matter at all. Those are already things we can look up and we can separate those discussions, so that if someone thinks that neutrality is a crock of garbage they can take the matter up on the talk page of the NPOV policy and not on an article talk page, so the article talk page can be kept focused on improving the article within existing policies and guidelines.

Calling something a sham consensus, a fringe theory, or nonneutral is not usually the final act (unless someone abandons the matter or concedes the point). Having called it that is a way of opening or focusing discussion and editing.

I agree that a consensus having been a sham and then changed into a consensus on which we can rely would result in the consensus not being a sham anymore and said so in the essay yesterday.

And I agree that a consensus can change and have said so several times. Being able to talk about a consensus helps that process. By calling it a sham within the definition, the discussion can immediately move to whether a policy, guideline, or ArbCom decision is being violated and not stay stuck on the claim that "there's already a consensus", since the violation is why it's a sham, and the response might establish that there's no violation and therefore the consensus is not a sham and can indeed be relied upon.

That not all editors need agree that a given consensus is a sham is also already in the essay.

Nick Levinson (talk) 19:51, 13 October 2013 (UTC)

Re: "That appears to assume that if an article is POV (assuming it is) but a consensus supports keeping it as is, then we can't edit it to be NPOV" - well, you can try... but if there is a consensus to leave it "as is" your edits will probably be reverted. So, yes, if there is a local consensus to keep the article "as is", your only choice would be to either a) convince the local editors to change their minds, or b) go up the chain of dispute resolution and seek to overturn the local consensus with a broader one. That's how Wikipedia works.

But you miss my key point... the consensus in my example is that the "as is" article isn't actually POV. It may be that one or two editors think it is POV, but the other five editors agree that it isn't. The majority may well have taken the NPOV policy into account, and reached their consensus based on interpreting that policy in the context of that specific article. Now, this local consensus may be overturned when you bring in more people, and reach a new broader consensus... or it may be confirmed. We don't know that until we actually do bring in more people. My point is that the initial local consensus wasn't a "sham". It was a legitimate consensus based on interpreting the policy in a given way. Blueboar (talk) 01:43, 14 October 2013 (UTC)

I am finding myself in agreement with Blueboar here, but I would like to repeat my previous supposition. Ostensibly, it seems the only cases that are correctly described as a "sham" are cases of deception. If there is not deception, what is the nature of a disputed consensus? It seems that all the disputes on consensus absent deception are garden variety disagreements of changeable consensus. I am perfectly ready to change my mind about this because someone else may have an idea of situations that don't involve deception but could still be illicit examples of consensus bad enough to shock the conscience of WP. --Ring Cinema (talk) 03:58, 14 October 2013 (UTC)

If a local consensus is binding even against policy, as suggested above by Blueboar, then policies are meaningless without going to a higher level each time to have the article adjudged anew. That's not how Wikipedia works. That would also make Wikipedia:Consensus#Level of consensus meaningless. This is not about ignoring just any consensus. It is about ignoring (not relying on) consensus that is in violation of a policy, guideline, or ArbCom decision. We're not supposed to be relying on such a consensus anyway, named or not.

If there's a desire to get rid of or rewrite the Level of Consensus subsection, please edit it or propose the change. As far as I know, that's not pending now.

I did indeed get your point about the article being really NPOV when it's allegedly POV. That's what discussion is for and I said so.

An initial local consensus may be a sham and then become not one because of a change somewhere. That would not be held against the newly legitimate consensus. Nothing says it would be and the essay reinforces that. If the essay needs further clarity, please edit or say how so I can edit it accordingly.

If deception is the only way a policy, guideline, or ArbCom decision can be violated, then someone has a very wide-ranging meaning for deception and I don't think there's consensus that, for example, all addition of OR is always deception. A violation can be substantive and at the same time transparently nondeceptive, i.e., honest albeit a violation. If an editor adds OR and writes into the edit summary that "this article needed OR so I added it now", that's not deception but it is a violation. If the editing was on a point not previously raised at that article, the editing created a consensus, and therefore it created a sham consensus. We do not have to debate whether the OR is good/bad in order to arrive at a new consensus. We can revert and thereby replace the sham consensus with a legitimate one.

Shocking the conscience is not the standard for whether a violation occurred. That would be too extreme a standard. The policies, guidelines, and ArbCom decisions are usually sufficient on their faces for determining what is violative. If the standard should be raised, that should be brought to a policy talk page, although I don't think that one will get adopted.

All we're talking about is giving a name to the kind of consensus on which we're not supposed to rely because it is in violation of a policy, guideline, or ArbCom decision.

Nick Levinson (talk) 16:16, 15 October 2013 (UTC) (Corrected misspellings & clarified on essay & re OR example: 16:27, 15 October 2013 (UTC)) (Corrected this sig block: 16:32, 15 October 2013 (UTC))

You say, "We're not supposed to be relying on... consensus that is in violation of a policy, guideline, or ArbCom decision..., named or not." In a sense that is partly true (guidelines don't belong here at all), but the inference you draw from it is mistaken. There is no place to stand outside of consensus that judges the results of consensus. So, what is a sham or not is in the eye of the beholder. The exception, I believe, is where a deception of identity confounds the discussion. --Ring Cinema (talk) 16:43, 15 October 2013 (UTC)

Nick... Re: "If a local consensus is binding even against policy, as suggested above by Blueboar..." No... That misstates what I am suggesting. I am suggesting that no consensus is ever "binding" (except in a temporary sense)... because all it takes to overturn a consensus is the formation of a new consensus that says something different. If you disagree with a previous consensus, you don't simply call it a "sham" and ignore it... you seek a new consensus that either confirms or overturns that previous consensus. Blueboar (talk) 17:41, 15 October 2013 (UTC)

All the beholder would be doing under the essay is naming what is already there, because all the essay does is define a name for the thing being beheld, just as the beholder of an article biased toward one point of view and excluding another would likely name that quality as "POV", because that term is already established within Wikipedia.

Guidelines do belong here, because a guideline consensus trumps an article consensus. We touched on this point before and I cited as examples the consensus for WP Do not create hoaxes or WP Notability as superior to an article's consensus. Exemptions from guidelines and even from policies exist but not just because some editors proclaim one absent a good reason. But if you still disagree and if you believe that a local consensus even without a good enough reason can ignore a guideline, what do you think of what the WP Consensus policy's Level of Consensus section says? It 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 ... guideline does not apply to articles within its scope.... Wikipedia has a higher standard of participation and consensus for changes to ... guidelines than to other types of articles."

If there were "no place to stand outside of consensus that judges the results of consensus", consensus would be useless. Throughout Wikipedia and probably dozens of times a day, judgments are made about whether there's a consensus and what that consensus is. That's no different with judging one as a sham. This is not a problem of someone declaring a consensus a sham and therefore all discussion freezing. Just the opposite: if you think a consensus is a sham and no one else does, you can do exactly what you would have done without the essay, except that you would now have a concise name the meaning of which is agreed upon, even if not all editors agree that the particular consensus is a sham. That is what discussion accomplishes. Your discussion can then focus on what violation occurred or didn't occur. Then whatever remedy would apply, if any, can be applied. That's the same whether we use a variable 50-word description that's offputting for length or the concise two-word name from the essay.

You've narrowed the exception you recognized from deception to one of "deception of identity". Are you saying that a consensus because of sockpuppetry can be a sham but that no other consensus can be a sham? In other words, are you saying that we should follow any consensus except one due to sockpuppetry even if it flagrantly violates an ArbCom decision, guideline, or policy? May we use only private emails as sources for articles even though they're unverifiable as long as an article's consensus says to use them? That wouldn't be deceptive but it would violate a policy. We can't discard every policy, every guideline, and every ArbCom decision whenever a local consensus wishes to, no matter how many votes there are. We're not allowed to do that whether the essay exists or not.

In most cases, if a consensus is problematic, you address the consensus and seek a change. But in some cases, ignoring it is precisely what we must do and several cases have been gven of just that. We do not leave an article as all OR because the consensus is to leave it that way. A consensus that violates policy is not to be relied upon. You're allowed to tag or delete the OR even against a sham consensus. Throughout Wikipedia, that expectation to edit is made clear. All we're doing is acknowledging a name. Listing the essay does not change any procedure (except to recognize a name) nor does it change the norm that usually a consensus even if disagreed with is legitimate.

I didn't misstate. A position that the local consensus allows keeping the article as is even if it is POV means that the local consensus overrides the POV policy. A local consensus does not have the authority to deem an article NPOV and therefore it is NPOV. It is POV or NPOV depending on what is in the article and what is sourced but not in the article. The local consensus might legitimately be on any issue within any discretion allowed under the NPOV policy and there might legitimately be disagreement on whether the article is POV, but if a local consensus can decide that the article is POV and must stay that way, then either the local consensus is a sham or the local consensus is more authoritative than the policy. The latter is already and explicitly contradicted by the Consensus policy. The Level of Consensus subsection makes the policy consensus more authoritative than the local consensus (with occasional exceptions).

I think the main concern in this thread against listing the essay is that editors would ignore a valid consensus. In most cases, a consensus is valid even if disagreed with, and the essay says so. It's not likely that an editor would rubber-stamp every consensus they don't like as a sham just as they don't rubber-stamp every article they don't like as POV. The latter happens more often (and is easier given that there's a tag for it) and we survive that without much difficulty; someone else looks at it and either starts to solve the POV problem or removes the tag. Anything can be misinterpreted; but misinterpretation is not a ground for having no standards. It may be a ground for clarification, but not for silencing. If clarification would help, please clarify or suggest how. But letting a discussion be more focused so we can get the core issue resolved helps Wikipedia.

Nick Levinson (talk) 15:44, 16 October 2013 (UTC) (Corrected indent & misspelling, added point, & rephrased: 16:00, 16 October 2013 (UTC))

I've read the above explanations and the essay and I still don't understand what problem you're trying to solve, and I definitely think we should not link it here yet. We already have Wikipedia:LOCALCONSENSUS, and I'm not sure this adds something new to the mix. Policy is constantly changing, and consensus is also constantly changing, so I'm not sure exactly what would be a "sham" consensus. The essay as written is confusing, and remember WP:IAR is also a fundamental policy, so that even gives an out for certain cases for an article to ignore other policies or guidelines.--Obi-Wan Kenobi (talk) 15:53, 16 October 2013 (UTC)

The problem is the need to keep discussions easier to follow by having a term for what we discuss. The view here seems to be to overturn WP:LOCALCONSENSUS. I don't want to overturn it. It should stay as it is. WP:SHAM serves in following the policy and not a consensus that overturns a policy unless there's a good reason for an exception. That is already the case even without the essay; the essay just defines a name. Everything changes over time and if a consensus used to be a sham consensus but no longer is then the old state doesn't matter; what matters is what is happening at a moment of contention, which is when the obligation exists. IAR is for exceptional cases, and is not the norm.

Please tell me what was confusing about the essay. I can edit it.

Nick Levinson (talk) 16:16, 16 October 2013 (UTC)

Guidelines do not trump a local consensus, by no stretch. --Ring Cinema (talk) 17:44, 16 October 2013 (UTC)

Who said anything about overturning WP:LOCALCONSENSUS? I am certainly didn't. I totally support WP:LOCALCONSENSUS.

Where I think we disagree is what to do in situations when WP:LOCALCONSENSUS might apply. Your solution seems to be to "call the local consensus a sham". My solution is... "seek a broader consensus to either confirm or overturn the Local consensus."

I think my solution results in fewer arguments, and a more solid sense of what consensus actually is... and so is more in line with the intent of WP:LOCALCONSENSUS.

I also think the problem is that you want to create a single name for something that it too complex for a single name. Every LOCALCONSENSUS issue is unique... Policy always needs to be applied and interpreted in context... and every article has a unique context. Blueboar (talk) 18:11, 16 October 2013 (UTC)

Overturning WP:LOCALCONSENSUS is required by Ring Cinema's view in the post just before the last, a view that was stated by the same editor earlier and even earlier and that was rebutted (by me) but without an answer specific to the rebuttal.

Everything is unique but there are some common characteristics and on the basis of those we can usefully define a name. The same is true of POV; not every POV is identical, nor every article or article subject, nor every claim of POV, but the term POV defines common characteristics and so we can use the term even when we disagree on whether a particular article is POV. If uniqueness prevented naming, human language would not exist.

My solution here is to use the pre-existing solutions. The only thing I've added is an agreed-upon name for a certain type of problem, to ease communication about it. The procedures don't change and neither do the remedies. Not relying on a sham consensus is already required without any essay; we are not allowed to act in violation of an ArbCom decision, for example. I didn't propose or create a change in procedures or remedies. The essay does not challenge any procedure or remedy.

Thus, I think we do not intend to disagree on what happens if an editor has called a consensus a sham. Either no one would do anything because no one wants to pursue the matter or they would do essentially what you suggest, working on the consensus or going up the decision ladder. I've added text to the essay to make that clearer. Is there something else I can add to help that understanding?

Nick Levinson (talk) 16:29, 17 October 2013 (UTC) (Linked: 16:35, 17 October 2013 (UTC))

Re: "The only thing I've added is an agreed-upon name for a certain type of problem..." I have several issues with that... first, the name you want to use has not been agreed upon (I have serious issues with calling any consensus a "sham"... except those formed by clear cut sock-puppetry). Second, I question whether the "type of problem" that you want to apply this name to actually is a "problem". Blueboar (talk) 17:03, 17 October 2013 (UTC)

Whenever an editor is told to follow a consensus even though it violates a policy, guideline, or ArbCom decision, that's a problem, regardless of quantity. Following such a consensus requires violating policy, guideline, or ArbCom decision. We're not allowed to violate them. We're not allowed to follow a consensus that requires violating them. Therefore, a consensus that requires violating a policy, guideline, or ArbCom decision is a problem. That kind of consensus is a problem.

To the definition in the essay, I gave the name sham consensus and no one proposed any other name. There was disagreement on whether to call it anything, which does not solve the communication problem of needing a name, and a suggestion that only a positive or neutral name be given, but the definition is of a negative characteristic, and so, to be clear, the name should also be negative. But no one suggested a specific other name. I had liked false consensus but that has already been defined and merging was not accepted. So I thought about it and looked in a thesaurus and sham consensus was, in my opinion, the best term. If you prefer another term that would be clear, thus negative, please propose it or write an essay that accomplishes much the same purpose and link it.

You seem to be suggesting that we should accept any consensus except one formed by obvious sockpuppets. Putting aside that subtle sockpuppetry would be okay, as perhaps you're only concerned about misunderstanding who the editors are and you're not willing to accept subtle sockpuppetry either (as a supposedly legitimate way to get around the prohibition on misuse of sockpuppets by one editor casting multiple votes), I'd like to offer a hypothetical example. Suppose someone, whose username is also their real-world name, uncovers a religion heretofore not reported in Wikipedia and finds a hundred sources for it, writes a lengthy complex all-POV article on it, creates a talk page for it, and proclaims on the talk page that this article must be kept as a POV article. Suppose a passerby likes the religion and adds their vote to the talk page in favor of keeping it POV. No reason is given for the exception from the NPOV policy. No sockpuppetry was entailed. Let's assume that there's no dispute that the article is POV. Therefore, the NPOV policy would have been violated. The consensus says POV is good for this article. Is it necessary to seek a higher consensus before, say, a new page patroller can completely rewrite the article as hopelessly POV? If yes, that's contrary to policy and the way to deal with your disagreement with the policy is by seeking or causing a change to the NPOV policy or the Consensus policy. If no, i.e., if a new page patroller need not seek a new consensus but can revamp the hopelessly-POV article, then the consensus supporting the article being POV is thereby rejected on the spot and sham consensus describes the NPOV-policy-violative consensus that immediately preceded the rewriting of the article. What is your position on this case of the clear NPOV violation with no sockpuppetry?

Nick Levinson (talk) 16:21, 18 October 2013 (UTC)

Thats a rather bizarre hypothetical. I think this conversation would do better with specific real-life examples. I still don't see how SHAMCONSENSUS helps anything, and am also not convinced it is a problem (nor it is a sham). There can indeed be cases where for a particular article we IGNORE a policy, guideline, or even ARBCOMRULINGFROMONHIGH - because it is what makes most sense. If someone comes along and stridely disagrees, then they have only to open up an RFC, advertise it broadly, state that X violates Y, and see what the community thinks about it. If the community agrees with your proposed change, then bingo, you've established a new consensus; if they don't, you've also established that the exception to the rule was acceptable. Either way, you don't have a sham - you just have a local consensus that was broadened to either be accepted or rejected at a global level. Let me give you a specific example - there was a (longstanding) discussion about whether we could have chinese characters in titles. Normally, I would say "no"- however, in this specific case, like Li (surname), there was a strong argument to be made to ignore this, from my POV, and use a chinese character as disambiguator, as it was the only sure way to disambiguate (e.g. using itself). So while localconsensus should not override global consensus in general, there are plenty of examples where per IAR, it does!--Obi-Wan Kenobi (talk) 16:28, 18 October 2013 (UTC)

And what you describe with the Chinese character is not a sham consensus. The cases you gave regarding WP:IGNORE are already covered in the essay; it agrees with your position. (I'm not sure that can apply to ArbCom but if it does the essay covers that, too, in agreement with your position.) The must-be-POV case I gave that you describe as bizarre may be bizarre and is unusual but it would be a sham consensus. One reason a sham consensus is rare is that in the must-be-POV case editors would usually say that the article is NPOV, not that it should stay POV. An even clearer case might be one of copyright infringement: suppose we add to an article the entire (short) lyric of a hit song. A consensus is formed that the only way to understand the article is if the lyric is published and so the consensus says to publish the entire lyric. (That is not legally a good reason for copyright infringement and we'll assume it does not support fair use, as it likely doesn't as the lyric would be complete.) The copyright owner contacts us and wants it deleted. Do we need to change the consensus or seek wider community input for a wider consensus before taking the lyric down? I'm saying only that such a consensus does not need to be followed, and that's specifically because it violates a policy, and the obligation not to follow it is already true and has been continuously true before the essay was created. Because the consensus violates policy, it is, at that moment, a sham.

The terminology simply eases communication. Someone doesn't need to explain what they mean when they can more simply use a term that has that meaning. It's like you don't have to explain that you mean a point of view that excludes another point of view, and even have to explain what a point of view is, when you can simply say "POV"; anyone following that link finds a definition, which does not have to be repeated.

To give specific examples would be inviting conflict on those pages now, requiring me to spend gobs of time justifying to those editors when I may not want to do anything with that article at this time, and that would be a distraction to quite a few of us. I'm only proposing a term and a definition, not a new procedure nor the removal of an existing procedure. A term does not need as much of a case to support.

I've added to the sham consensus essay, clarifying it. Please let me know how I can clarify it further.

Nick Levinson (talk) 17:16, 18 October 2013 (UTC) (Corrected misspelling, syntax, & punctuation: 17:23, 18 October 2013 (UTC)) Added phrase at 5:23p UTC: 17:49, 18 October 2013 (UTC))

Hi Nick. Thanks for the reply - I'm afraid it's not any better, having looked at the recent additions. It's still rather confusing, and I still can't see the difference between a sham consensus and a local consensus - who is to judge? The example above re: copyright is tending into legal matters, whereby the WMF can override consensus in the interest of protecting the legal interests of wikipedia - so it's not a good example to use. Again, a local consensus on a particular article that violates a particular policy or guideline is NOT necessarily a sham (which is a pejorative term) - but there is a difference between IAR for one part of one article, and a project saying "NPOV doesn't apply to articles within our remit - which is I think the point of localconsensus wording). I just don't see this shamconsensus concept as workable or useful in spite of explanations - we would be better off expanding on localconsensus and explaining or giving examples of where local consensus can override community consensus, even if it normally shouldn't. I suggest in any case that we close the discussion here, as there is no consensus to link to it from this page, and continue it at the essay.--Obi-Wan Kenobi (talk) 17:28, 18 October 2013 (UTC)

A local consensus is presumably legitimate. A sham consensus is not. Would it help if I added that to the essay? It already says that most of the time a consensus is legitimate. The consensus is judged by anyone who has the responsibility of judging it, just as with the case of POV, and that judgment is accepted or rejected by anyone else, as with a POV judgment. With legal matters, including copyright and defamation, our right or duty to ignore a consensus to violate the law is also true of a consensus to violate policy. We all agree that a consensus to violate a policy may be legitimate when WP:IGNORE applies and the essay says so. If you have a better term, please suggest it or add an essay to that effect, a suggestion I already made. So far, no one but me has discussed it at the essay's talk page but that's open. Nick Levinson (talk) 17:49, 18 October 2013 (UTC)

How's this for a better term: "Potentially Problematic Consensus". And it can be summed up as follows:

• A "Potentially problematic consensus" is any existing consensus that an editor thinks may have a problem... whatever that problem may be. We resolve the issue of a Potentially Problematic Consensus by discussing the existing consensus at a wider level, thus forming a new consensus that either confirms or overturns the previous one.

That's it... If we need an essay, that's all it needs to say. It really is that simple. Blueboar (talk) 21:41, 24 October 2013 (UTC)

No, because it does not address what cannot be done by that means. You can write that essay, but it's different, not a replacement. An ArbCom decision is, by policy, not subject to editor consensus; therefore, any consensus in violation of an ArbCom decision is a sham and cannot be relied upon, and does not need to be the subject of a new consensus, since no new consensus is relevant. An article consensus that does violate a policy or a guideline is already invalid because it already violates WP:Consensus, which is already a policy, and therefore such a consensus would already be a sham. The definition for Potentially Problematic Consensus is too vague to be of much use since what you propose is already covered by the Consensus policy, so I don't see that the new essay would add anything, but go ahead and write it if you think it will add something. As your proposal stands now, we still need to link to the more usefully precise term, which so far is sham consensus. Nick Levinson (talk) 15:06, 25 October 2013 (UTC)

Nick, I'm sorry but you're just wrong here. " An article consensus that does violate a policy or a guideline is already invalid because it already violates WP:Consensus" - that is patently false. Articles violate guidelines ALL THE TIME, and they also violate policies - we have many policies, and many policies are in conflict, so an article could be in accord with one policy and not with another. The only case where I think you have an invalid consensus is where a group of editors decides to go against an arbcom or WMF decision, but we don't need a new article about that, we already have language in our policies describing what to do in that case. in the other cases, where you say "Gosh, this article violates WP:AT policy" - the solution is to form a new consensus around the proper title for the article. The Manning case is a great example where people said "Bradley Manning violates BLP even if it adheres to WP:AT", while others said "Chelsea Manning violates WP:AT and doesn't violate BLP" - the only solution is discussion, and naming either of these positions as a sham or invalid is just plain wrong. Per IAR, there is very little, actually, that cannot be done here, and very little that cannot be henceforth undone by a broader consensus.--Obi-Wan Kenobi (talk) 15:13, 25 October 2013 (UTC)

Then you disagree with the policy consensus supporting the WP:Consensus provision and should edit it or propose an edit to it. Right now, that policy consensus stands and therefore we cannot violate it; WP:IAR provides exceptions but not for everything all the time. Even if many articles are in violation, they need to be aligned with policies and guidelines (including IAR if appropriate) and editors are free to do so. In general, policies can be understood as a consistent whole; conflicts between them should be redressed. I haven't read the debate around Manning but what you describe sounds like it is based on that consistency, because, for example, where BLP applies it overrides non-BLP policies about delaying edits so they can be discussed, and that's not an inconsistency. Broader consensus works but it need not be awaited when the existing consensus is itself void, as some would be. Voidness, as opposed to mere disagreement, is generally rare. Nick Levinson (talk) 17:54, 25 October 2013 (UTC)

It appears to be common practice, including among editors on this page (myself included), to rely on a pre-existing higher-level consensus to overrule a contrary article consensus without awaiting a new article consensus to conform to the higher-level consensus. Most editors appear to agree that reinventing the wheel is not necessary. Therefore, by implication, for example, an article about a non-notable subject is already against consensus, a guideline consensus, and first developing a new local consensus to delete it is not necessary. The speedy-deletion process includes many general examples of that. The belief that an article consensus overrides even a guideline consensus appears to have very little support across Wikipedia and appears to be widely opposed in belief and in practice. Thus, that a consensus sometimes need not be followed is already established and agreed to, and all we need is a name for that kind of consensus. Nick Levinson (talk) 16:32, 28 October 2013 (UTC)

Re: "any consensus in violation of an ArbCom decision is a sham and cannot be relied upon, and does not need to be the subject of a new consensus, since no new consensus is relevant."... not necessarily... there may be something relevant to the specific local consensus that the ArbCom decision did not take into account ... something that might cause ArbCom to re-evaluate its previous decision, or to make an exception. Again, the consensus isn't a "sham"... it is simply a consensus that needs to be either confirmed or overturned. Blueboar (talk) 18:57, 28 October 2013 (UTC)

No. Here's the policy: "The English Wikipedia Arbitration Committee may issue binding decisions, within its scope and responsibilities, that override consensus." Therefore, consensus, even a new one, cannot override a binding ArbCom decision within its scope and responsibilities. What I quoted is a policy: Wikipedia:Consensus, the section Decisions Not Subject to Consensus of Editors). A disagreement with an ArbCom decision may go to ArbCom, and is not left to resolution with only a lower-level consensus, even if it is believed that ArbCom would rule differently than it had in the past. You may, of course, disagree with the policy, in which case you should amend or propose to amend it. But, right now, the policy by a policy consensus is that a consensus may not override such an ArbCom decision. And, right now, whether the essay exists or not, a consensus to override ArbCom is a sham consensus and cannot be confirmed except as a sham and does not need to be brought to ArbCom for overturning first. Nick Levinson (talk) 16:09, 30 October 2013 (UTC)

I've added examples to the essay. These are examples of sham consensus. The contrary kind proposed above by an editor, examples of where a local consensus would override a higher-level consensus, would be appropriate for another essay or page. We could add examples of legitimate consensus, but the overwhelming norm is legitimacy and almost any consensus found in Wikipedia would qualify, so giving examples would likely be bloat. But it appears that confusion is persisting because of a misunderstanding of what a sham consensus is because editors appear to be projecting something else altogether into the essay, and perhaps the examples will help to make the essay clearer as to what it is about. Thanks. Nick Levinson (talk) 17:20, 30 October 2013 (UTC)

!vote

Hi, !vote is currently a WP:Cross namespace redirect to Wikipedia:Glossary. I have nominated it for discussion at Wikipedia:Redirects for discussion/Log/2014 January 4#!vote.

There is currently a link to Wikipedia:Polling is not a substitute for discussion in the introduction of this page('Wikipedia:Consensus'), under the word 'vote', and there is a link to WP:!VOTE and Wikipedia:POLLING in the prose under the village pump. But no mention of 'Polling is not a substitute for discussion', 'straw poll', 'not a vote', or '!vote'. I wonder if this page might be able to mention those concepts more clearly by using these key words somewhere in the prose, and linking to WP:Polling is not a substitute for discussion without hiding the page name? John Vandenberg ^(chat) 02:13, 7 January 2014 (UTC)

Consensus building to add a nomiaciones that have been deleted from Miley Cyrus

Hello , you see, has deleted a section of the article: "List of awards and nominations received by Miley Cyrus" and I wish they could put back , since I've been talking to a mate and will not let me post it , because it says Wikipedia does not have to be a means of endless lists, but it is a very important nominations Miley Cyrus! There are few rewards anyone! They are the World Music Awards! Now, that is what has been deleted and I hope that through consensus here can be inserted into the page as expected :

World Music Awards

Year	Recipent	Category	Result
2014	Adore You	World's Best Song	Pending
2014	We Can't Stop	World's Best Song	Pending
2014	Wrecking Ball	World's Best Song	Pending
2014	Bangerz	World's Best Album	Pending
2014	Adore You	World's Best Video	Pending
2014	We Can't Stop	World's Best Video	Pending
2014	Wrecking Ball	World's Best Video	Pending
2014	herself	World's Best Female Artist	Pending
2014	herself	World's Best Live Act	Pending
2014	herself	World's Best entertainer of the year	Pending

• Thanks and I hope we get to a solution :)

I'm afraid that this is not the place to seek help with a dispute with another user. Please discuss the matter with the other editor on the article talk page first — not just through edit summaries — and if that does not work out after a thorough discussion, consider dispute resolution. Regards, TransporterMan (TALK) 17:21, 24 March 2014 (UTC)

Determining consensus section

Was there an outcome to this discussion? [5] I've seen two unrelated editors this month ignore unanimous opinion against them and massively edit-war against consensus, saying (to paraphrase, and perhaps exaggerate a bit for comic effect) "consensus is achieved through the quality of arguments and not by numbers, and as always I find everyone else's arguments to be invalid, so therefore, logically, it's actually *everyone else* who is ignoring consensus and not me." Was there anyone against the proposal that we make it clear that this applies only to closing and not to editing? Rolf H Nelson (talk) 02:51, 23 February 2014 (UTC)

Quality of argument is in the eye of the beholder and humility is hard to teach. --Ring Cinema (talk) 06:20, 23 February 2014 (UTC)

WP:CONSENSUS is more about the process of how to reach agreement ... it's not about determining who "wins" when there is a lack of agreement. I wish we could ban the word "determined" from the page... a consensus is never "determined" - it's "achieved". A consensus can eventually be achieved when all editors work together in an attempt to resolve each others concerns and issues.... a true consensus is never achieved when editors are out to "win" an argument. Editors don't listen to each other when they try to "win" the debate... and when people are not listening to each other it is actually impossible to achieve a real consensus. As soon as you stop listening to each other, and start to count up votes, you are no longer trying to achieve a consensus, you are trying to "win" an argument. Blueboar (talk) 16:30, 23 February 2014 (UTC)

That doesn't answer the question though, what do you do when one editor refuses to listen to everyone else on an article and just ignores them because what they say is "false"?Smeat75 (talk) 20:17, 25 March 2014 (UTC)

The glib (but valid) answer would be "dispute resolution." But there are other options:

• If you believe that there is a consensus and the one editor is just arguing against it, either (safer) ask for a consensus evaluation at AN or (more risky) report them for tendentious editing at ANI.
• Ask for an RFC to bring in more editors.
• Consider that truth and false are determined by verifiability; shift the discussion to one about sources.

Regards, TransporterMan (TALK) 20:47, 25 March 2014 (UTC)

Thank you,TransporterMan!Smeat75 (talk) 21:05, 26 March 2014 (UTC)

@Smeat75: Closure — aka consensus evaluation — requests are actually at ANRFC and only transcluded to AN to try to get an admin's attention. Just caught my error on that one, sorry about that. Regards, TransporterMan (TALK) 14:05, 27 March 2014 (UTC)

Lack of consensus about how to change prior version that everyone agrees stinks

The policy says: "In discussions of proposals to add, modify or remove material in articles, a lack of consensus commonly results in retaining the version of the article as it was prior to the proposal or bold edit." Is this policy true even if the lack of consensus is only about how to change the prior content, and not about whether to change the prior content? In other words, if everyone agrees that the prior version stinks, but there is a complete lack of consensus about how to change it, does this quoted policy remain true? Maybe it should say whether it remains true. Incidentally, this inquiry is unrelated to any article that I'm involved with, and furthermore I am not expressing any opinion here about what the policy should be. Clarification could prevent a lot of edit-warring, regardless of how it's clarified.Anythingyouwant (talk) 16:50, 28 March 2014 (UTC)

I would say yes, and editors should continue to work towards a compromise, unless there is an agreement to start with to remove the material, then a separate discussion could be had on what to add back.--Obi-Wan Kenobi (talk) 17:35, 28 March 2014 (UTC)

Everyone hates sentence X that's in an article, side Y wants to change it to version Y, side Z wants to change it to version Z, and no one wants to delete version X unless they can change it to Y or Z. I see this all the time, and the edit-wars that follow. Obiwan, you seem to be saying that in this case the policy requires X to remain until there is consensus to delete it or to change it to Y or Z. Correct? Would it be worthwhile to clarify this in the policy, or do you think it 's already sufficiently clear?Anythingyouwant (talk) 17:48, 28 March 2014 (UTC)

If people really can't agree to remove it, then it should probably remain, bad as it is. I'd suggest, again, seeking a compromise,even if it's not what everyone wants. Not sure if policy needs to change though.--Obi-Wan Kenobi (talk) 17:57, 28 March 2014 (UTC)

I tend to agree with you Obiwan, that keeping the universally unwanted content is already required by policy. But, when I tried to clarify the point years ago, it ran into objections against keeping a "universally unwanted" bit of content.[6] What to do?Anythingyouwant (talk) 18:03, 28 March 2014 (UTC)

Right now, the policy is descriptive; it talks about what "commonly" happens in these situations, without prescribing a specific course of action. I think it's a bad idea to change the policy to be more prescriptive. These situations are diverse enough that a one-size-fits-all policy mandate is more likely to be abused than to be helpful. In that, I agree with the feedback you got when you brought this up a few years ago. MastCell ^Talk 18:11, 28 March 2014 (UTC)

Being descriptive is fine with me. What is the correct description of what commonly happens when no one wants to delete a sentence, everyone agrees it should be changed, but there is no consensus about what to change it to? Obiwan seems to be saying that this situation is already covered by the policy, and that the universally unwanted sentence is commonly left as-is. If that's correct, then why not say so explicitly in the policy? It's a fairly common problem that causes a lot of edit-warring, which could be alleviated by a clear policy statement about common practice.Anythingyouwant (talk) 18:23, 28 March 2014 (UTC)

What are the other options? You've already stated that there isn't not consensus to simply delete, so, it seems to be "no consensus to delete the material", which is covered.--Obi-Wan Kenobi (talk) 18:28, 28 March 2014 (UTC)

What happens all too often is that one side in the dispute (or both) will say, "there's a clear consensus that this lousy sentence should be changed and therefore I am doing so". They install their preferred version, thinking that that is perfectly common and acceptable according to the policy.Anythingyouwant (talk) 18:32, 28 March 2014 (UTC)

It's sort of hard to deal with this in the abstract. Again, my generic feeling is, if both sides don't agree on removing the material pending a resolution, then the material should stay, bad as it is. If you get into a situation like that, I'd suggest trying to bring the parties to agree to remove material which is everyone agrees is bad - this is generally good for the wiki, no information is better than bad information.--Obi-Wan Kenobi (talk) 18:44, 28 March 2014 (UTC)

I agree with your generic feeling, and believe it would be useful to describe that feeling briefly in the policy. There's no reason why an eager editor would agree to delete an awful sentence if he or she feels justified in changing it to his liking, based on the correct perception that there is consensus to (somehow) change the sentence. I don't do it, but others sometimes do. You've got to admit that it is kind of absurd, or at least counterintuitive, that a Wikipedia policy would ever require editors to leave a sentence as-is even if everyone agrees it is awful and needs to be changed. But that's sometimes necessary.Anythingyouwant (talk) 19:03, 28 March 2014 (UTC)

I disagree, I don't think it's absurd, it's really the editors who are being absurd. Think of it this way, as a heuristic - editor A prefers Z to A, editor B prefers Y to A, and both agree A is terrible. However, by *not* agreeing to delete A, they are essentially saying 'nothing' < 'A' < 'my choice' - in other words, by not agreeing to delete, they are simultaneously saying that keeping it is preferable to deleting it. If you can change their mind, so that it looks like this 'A' < 'nothing' < 'my choice', then the choice to delete is clear. If there is consensus on either way, then everyone is agreed that it is BETTER to keep than to delete, thus you actually have consensus to keep.--Obi-Wan Kenobi (talk) 19:25, 28 March 2014 (UTC)

Often only one of the editors is being absurd by refusing to delete, insisting (correctly) that there is consensus to change the sentence (somehow). Or you may have a case where deleting a sentence is impractical (e.g. a picture caption or the lead sentence that defines the subject). It would be simple enough for this policy to briefly say that it is not common practice to change material even if there is consensus to do so, unless there is also consensus about how to do so. We agree the policy already implies that, but making it explicit would not take up much space. Anyway, I'm going to try to be quiet now, and see if you and anyone else comments. Cheers.Anythingyouwant (talk) 19:43, 28 March 2014 (UTC)

Contested actions by administrators.

When actions by administrators are contested and the discussion results in no consensus either for the action or for reverting the action, the action is normally reverted.

Is this still true? --Anthonyhcole (talk · contribs · email) 14:09, 25 November 2013 (UTC)

• Was it ever true? WP:DRV. WP:MR. At these venues, administrative decisions, are contested. A "no consensus" close to the review results in the contested decision standing. "Admin discretion" stands in the absence of a consensus to reverse it. --SmokeyJoe (talk) 03:28, 7 January 2014 (UTC)

  Blocks get reversed. What other examples can we think of? WhatamIdoing (talk) 16:53, 7 January 2014 (UTC)

Yeah, most cases do not. Blocks may be the only case where a no-consensus does not go in favor of the admin whose decision was challenged. If the wording here is long-standing, this is probably more of a reason to ensure that procedures that are hostile to non-admins be made less to, to conform to this policy, rather than break the policy to conform to the admin pool's clique-like default that admins are always right until unquestionably proven wrong.  — SMcCandlish ☺ ☏ ¢ ⚞(^Ʌⱷ҅̆_⚲͜ⱷ^{^})≼  08:00, 6 April 2014 (UTC)

Policies and guidelines are not "types of articles"

I fixed an obvious typo, changing "articles" to "pages", here, correcting Wikipedia has a higher standard of participation and consensus for changes to policies and guidelines than to other types of articles to read Wikipedia has a higher standard of participation and consensus for changes to policies and guidelines than to other types of pages because, obviously, policies and guidelines are not articles, which is what my edit summary said.

This was reverted by Ring Cinema, who has been debating with me elsewhere, with an edit summary that seems inadequate to me: "not sure but I believe this is better and not inaccurate -- 'pages' includes areas where consensus doesn't apply at all. If "not sure", then don't revert, open a discussion on the talk page. There are no pages, except in userspace, where consensus does not apply (well, someone somewhere can probably dig up some rare exception, maybe something controlled by WP:OFFICE lawyers; but you know what I mean). Whether pages that aren't subject to consensus may exist is irrelevant anyway. It doesn't change the fact that our internal WP:POLICY pages are not WP:ARTICLEs, but are pages (as are articles). The objection is akin to noticing an obvious change from apples are oranges to applies are fruit and reverting it on the basis that not all fruits are apples and regardless of the fact that oranges and apples are different; it's fallacious twice over. Is there any actually substantive objection to the correction?  — SMcCandlish ☺ ☏ ¢ ⚞(^Ʌⱷ҅̆_⚲͜ⱷ^{^})≼  06:23, 6 April 2014 (UTC)

Is there a word (or words) that can convey "more than articles but less than pages"? Butwhatdoiknow (talk) 12:51, 6 April 2014 (UTC)

I would suggest:

• Wikipedia has a higher standard of participation and consensus for changes to policies and guidelines than for changes to articles.

Unless there is some other type of "page" that needs to be mentioned. Blueboar (talk) 13:23, 6 April 2014 (UTC)

May I suggest: "Wikipedia has a higher standard of participation and consensus for changes to policies and guidelines." There are other types of pages: "disambiguation pages, templates, navboxes, user pages, discussion pages, file pages, category pages, help pages and Wikipedia policy pages." --Ring Cinema (talk) 16:25, 6 April 2014 (UTC)

Perhaps:

Wikipedia has very strong standards for changes to policies and guidelines, generally requiring a stronger consensus than for changes to any other pages on Wikipedia.

No need to enumerate any "levels" as this makes it a strong and broad statement as to such edits. Collect (talk) 18:25, 6 April 2014 (UTC)

To return to the fruit analogy, there is no reason to say apples are fruit but are not grapes, pomegranates or figs when apples are fruit does just fine. Please don't turn a trivial typo correction molehill into a mountain. Also, please note that the assertion that Blueboar would make with their proposed edit isn't necessarily true. Various articles are protected from editing and/or subject to discretionary sanctions in response to continued, heated editwarring, and (at least for a span of time, which may be a few days or many years) may in fact have a higher expectation of discussion and consensus-seeking than the average policy page. It's much easier to get consensus for edits to a page like WP:CONSENSUS itself than at some of the more contentious articles.

Unless there's some general class of pages at WP that categorically and consistently require a higher level of consensus than policies and guidelines, there is no reason not to just say "pages" here. Cf. WP:CREEP and WP:KISS; don't over-think or over explain it. If someone ends up somehow WP:GAMEing that simple wording, then maybe it would need additional explaining. PS: I think it would require a whole site-wide RFC for us to come up with some kind of hierarchical system of "levels" of consensus beyond what is already at WP:CONLEVEL, which is general for a reason (and came about in response to WP:RFARB cases that were dissimilar enough in their details that its important it remain general). PS: Even adding "any" to "pages" isn't necessary. As I first notec, someone can probably find an odd example somewhere of a page controlled by OFFICE or whatever that can't be changed by consensus at all, so "any pages" is highly unlikely to be 100% accurate.  — SMcCandlish ☺ ☏ ¢ ⚞(^Ʌⱷ҅̆_⚲͜ⱷ^{^})≼  02:36, 7 April 2014 (UTC)

A new proposal regarding bird article names

Those interested in the wider issues might like to look at Wikipedia talk:Manual of Style#A new proposal regarding bird article names. Andrewa (talk) 03:54, 9 April 2014 (UTC)

Semi-protected edit request on 18 April 2014

103.17.63.128 (talk) 09:59, 18 April 2014 (UTC)

NOT DONE... We can not respond to your edit request in any other way... you need to tell us what you want changed. Blueboar (talk) 12:14, 18 April 2014 (UTC)

CONLIMITED addition

Guy Macon made an addition of additional examples to CONLIMITED which was immediately reverted by Flyer22 in this edit. Just for the record, I don't think the addition was needed, that the scope is pretty well defined. (Though perhaps Guy can point us to the "multiple editors misunderstanding the scope" and I'll change my mind.) But I also think that Flyer's reason for the revert — "People often go to noticeboards and those other places to achieve WP:Consensus, as part of WP:Dispute resolution." — misses the point. CONLIMITED, with or without Guy's addition, doesn't restrict the ability of folks fighting over an edit at a single article to come to consensus at DRN or at one of their user talk pages, but does restrict the ability of folks at a noticeboard or Wikiproject or the Pump to change policy in a way that affects a bunch of different articles or editing in general. Best regards, TransporterMan (TALK) 18:10, 25 March 2014 (UTC)

Yeah, I figured that my comment would be seen by one or more editors as missing the point; still, since noticeboards and the WP:Village pump are often used to achieve wide WP:Consensus, I did not want anything misinterpreted by that addition. Like you noted, any one of the WP:Dispute resolution processes might reach a WP:Consensus that WP:Ignore all rules applies to a particular article. As for applying to many articles, a WP:RfC, for example, can have many editors weigh in and that WP:RfC just might result in a change of a guideline or policy. Such dispute resolution processes are often "the broader community." There is never any way to get all of the Wikipedia community to weigh in on a matter (and like the WP:Consensus policy notes, consensus is not unanimity). Of course...any effort to change a guideline or policy should be had that guideline or policy's talk page. Flyer22 (talk) 18:29, 25 March 2014 (UTC)

Oh, and I don't think that the WP:REFDESK counts as WP:Dispute resolution or WP:Consensus. Flyer22 (talk) 20:37, 25 March 2014 (UTC)

Examples of three users thinking that the reference desk guidelines overrule the talk page guidelines:

(In response to "I say stop removing the questions")

"If someone asks for professional advice, we're not allowed to give it. Hence the question will go unanswered. And hence the question itself need not be kept there." --Baseball Bugs[7]

"Well, you could keep it, but what's the point, when there is no possible answer?" --Baseball Bugs[8]

"The ref desks are not really talk pages, although they're kind of structured that way." --Baseball Bugs[9]

(In response to "WP:LOCALCONSENSUS clearly states, 'Consensus among a limited group of editors, at one place and time, cannot override community consensus on a wider scale.' Not editing or deleting others' comments is a Wikipedia-wide policy")

"I personally fail to see how talk pages are supposed to completely override the reference desk guidelines." --Dmcq[10]

"Go and raise an RfC then and this can be decided or you will waste even more of people's time." --Dmcq[11]

"The reference desks are not talk pages." --Dmcq[12]

(In response to "Allowed: "'"We cannot give medical advice here. Please consult a doctor. For general information about [ example ], you may want to see our [ example ] article." Against policy: deleting or editing the question.")

"We've already got a disclaimer, a statement in the top-of-page guidelines, and a template to use when we remove material, and a template to notify the OP whose material's been removed. Why are we discussing adding to this? ---Medeis (AKA μηδείς)[13]

Examples of deletions that arguably are allowed by the reference desk guidelines but are clearly not allowed by the talk page guidelines:

Deleting an answer because Medeis (AKA μηδείς) decided that it was unhelpful.[14][15] Actually, it was pretty good advice if the questioner is in the UK, PC World (www.pcworld.co.uk) does a good job at data recovery.

Deleting an editor's comment complaining about the way Medeis (AKA μηδείς) edited the editor's previous comment.[16] Moving it to talk would have been a better choice. Also labeled the deletion as a merge,

Deleting a question by a blocked sock and taking out answers from other users with it, two days after the discussion ended.[17]

Deleting a question that isn't medical advice:[18] (and what's up with the deleting and hatting?)

Editing (not just deleting) another user's comment[19]

I can provide dozens of additional examples if needed. --Guy Macon (talk) 19:21, 25 March 2014 (UTC)

Oh, now I see, but I'm not sure how I feel about this. I think an absolutely strict interpretation of CONLIMITED would, you're right, prevent some of this, but the fact is that there does seem to be some accepted exception to CONLIMITED for Wikipedia-space pages establishing their own rules which apply at those pages. It would seem that practices which would be unacceptable at an article page are accepted by the community and generally uncontroversial at those pages, such as closing and archiving discussions at DRN or removing stale and answered requests at 3O. Closing off a section at, say, Scientology or another mainspace page and demanding that it remain closed and be archived to an archive page would get you blocked or banned, but doing it at DRN is everyday practice. Maybe we need a policy just on that subject. Best regards, TransporterMan (TALK) 19:36, 25 March 2014 (UTC)

I have been thinking about that very thing. It seem to me that DRN and (most? all?) of the other noticeboards are arguably policy-compliant because of Wikipedia:Talk page guidelines#Off-topic posts. Clearly what is off-topic is a local decision, and differed between DRN, ANI, etc., but all of those venues should follow Wikipedia:Talk page guidelines#Off-topic posts when deciding what to do once it is established that something is off-topic. DRN doesn't need to delete non-harmful posts. Collapsing and moving to another section is good enough, IMO.

WP:AIAV is an interesting test case for that idea; not only are the reports limited to a certain format, they also get deleted instead of archived, which technically violates WP:TPOC. My response to that is that, while the village pump and reference desks are definitely talkpage-like, AIAV really isn't.

Whatever is decided, I would really like it to be made explicit that the reference desks are not exempt from the talk page guidelines, and that deleting a good-faith question rather than saying you cannot answer it is unacceptable behavior. --Guy Macon (talk) 20:45, 25 March 2014 (UTC)

More clarity is better than less clarity, generally. It makes the section a bit longer, but if it is likely to help reduce strife, that is more than enough reason to endorse the addition.  — SMcCandlish ☺ ☏ ¢ ⚞(^Ʌⱷ҅̆_⚲͜ⱷ^{^})≼  08:03, 6 April 2014 (UTC)

PS: The Village Pump really isn't like a wikiproject or the Ref Desks; the broadness of participation there ensures that the consensus isn't "local" there. However, it is very often poorly informed because one side of an issue will set up the debate, VP will dump comments in based on that, and the discussion already seems settled before the other side even has a time to react. VPP discussion do not always go that way of course, but frequently enough that VPP is sort of "not a reliable internal source" as it were. It is best viewed as a straw poll, a general gauge of where consensus might go in a more neutral, longer-term centralized discussion or RfC.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  20:59, 12 April 2014 (UTC)

The refdesk addition has been reverted on the basis that reference desks don't have "articles within their scope"; so some other approach to the wording would be needed.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  07:02, 20 April 2014 (UTC)

Shortcut de-capitalisation

Could someone have a look at these recent edits? Andrewa (talk) 19:15, 6 April 2014 (UTC)

Indeed. There's been a rash of this de-capitalization of WP shortcuts all over the place lately. It's not helpful and it doesn't represent how shortcuts are actually used.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  20:54, 8 April 2014 (UTC)

Thank you! Are other editors involved or just this one?

In any case I think their talk page is the next stop. They are relatively new and have also made some other edits (restringing my talk page posts) which seemed unhelpful to me. I objected on their talk page and while they did eventually stop doing it (to me at least), I didn't feel that I'd had a positive response. Would someone else like to try? Andrewa (talk) 03:51, 9 April 2014 (UTC)

I spoke too soon, they are at it again and this time also breaking the links from the heads-ups on other pages by changing the section heading. This time I have reverted. [20]

Could someone else have a look at User talk:Mama meta modal#Please do not restring my comments or change the headings of sections I create, as this looks like going to an RfC/U and I need at least one other editor involved for that. TIA. See this version for previous discussion there (but please don't edit that version of course).

Or if you think I'm being unreasonable, comments to that effect welcome too. Probably best on their talk page or mine rather than here. Andrewa (talk) 07:54, 9 April 2014 (UTC)

I fixed some more of these at WP:AT. I decline to go edit-stalking the user in question to revert more of these; I figure people will clean them up as they come across them.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  08:46, 20 April 2014 (UTC)

A summary of this page in 11 words

I propose demystifying this page with an 11-word quote added to the lead which makes this key policy easier to understand for the uninitiated:

The policy has been summarized by Clay Shirky as follows: "Wikipedia's editor-in-chief is a rotating quorum of whoever is paying attention"

Does anyone have any objections to adding this to the lead?

Oncenawhile (talk) 20:41, 20 April 2014 (UTC)

I rather think the suggestion is quite unlikely to get consensus here. It is nineteen days late. Collect (talk) 21:16, 20 April 2014 (UTC)

Yeah, doesn't seem like lead material in a policy. It is funny, though. Maybe add it to a consensus-related essay somewhere. And source the quotation.[21]  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  06:16, 21 April 2014 (UTC)