Wikipedia talk:Consensus/Archive 10

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 5

←

Archive 8

Archive 9

Archive 10

Archive 11

Archive 12

→

Archive 15

I wanted to bring up the spread out nature of this discussion. We're discussing the addition of something here, there's an almost month old conversation at

) 10:40, 1 September 2009 (UTC)

No opinion. There's still substantial work to do on understanding the positions of the current participants and getting data on what situations those positions apply best to; but it doesn't bother me if more people get involved, the discussion moves to new pages, and things get more chaotic. I'll keep up. - Dank (push to talk) 15:04, 1 September 2009 (UTC)

What we need is an agreement that discussion takes place at the most relevant page. I don't think that moving it out will help, I'm already concerned that this discussion is going on without much corresponding editing of policy to line up with what we're saying. An RfC may be a bit premature.    M   21:55, 2 September 2009 (UTC)

I think it's vital that we get this discussion centralized. It's obvious that what amounts to the same issue is being discussed in so many different places that it's hard to keep up with it, and this tends to unbalance the power into the hands of people who are able to dedicate enough time to follow all the fragmented discussion threads. causa sui× 04:55, 3 September 2009 (UTC)

One problem here is that I'm likely conflating two separate, yet intertwined, changes. I think that M may be as well. There's a whole... thing going on with

WP:NCON and the other related Naming conventions. One component of that dispute involves these discussions, which deal with how Consensus works, the role of Policies and guidelines, and the general process involved in making changes. I've been thinking about this today and, while I may be making excuses here (I'm trying to work on my Bot), M and I are likely too close to both disputes to be able to clearly summarize both disputes and to not allow our own views to slip in (I think I am, at least). It wouldn't help much if we just end up bickering about exactly what the RfC/cent is supposed to be covering.
— V = I * R (talk to Ω

) 05:14, 3 September 2009 (UTC)

I understand, Causa; I think the solution is just to notice that conversations are ongoing and reaffirm that we're not in any hurry. The goal is to make the final product reasonably acceptable to everyone, and the stress is coming from the fact that we all recognize we're far from that goal.

WP:POLICY will always be part of the mix, so anyone interested in the discussion will probably have to skim at least these two talk pages, until/unless someone wants to make a helpful summary to make the discussions easier to keep up with. - Dank (push to talk

) 12:52, 3 September 2009 (UTC)

I am sincere when I say that I believe everyone here has the best intentions: I don't believe that the discussion is being deliberately fragmented with the aim of excluding people! Being patient is a good thing, but the problem is that the longer you wait, the more fragmented the discussion is going to get and the longer it is going to get in each of its fragments, which only makes it more difficult for people like me (who only have maybe a half hour a day on Wikipedia) to keep up with it. I can't even respond to all the responses to the comments I have left in these threads, let alone read anyone else's. Bottom line: If you want to include the greatest number of people possible, you have to centralize this discussion. causa sui× 15:58, 3 September 2009 (UTC)

I'm not opposed to a page linked from here and from WT:POLICY where we discuss a particular issue, but I don't think it works to exclude a policy page's talk page from the discussion about that page; that's what a talk page is for.

I added a short infobox at the top of POLICY and CONSENSUS saying that the discussion is ongoing; I'm hoping that will get across the idea that this discussion won't be closed prematurely, and there's no need to get an edit in quickly for it to "count". If someone feels a more formal "disputed" template is necessary to get the idea across that we're not done here, we could do that, but I'm not a fan of using that template on policy pages. - Dank (push to talk) 17:50, 3 September 2009 (UTC)

I've been struggling with the snappiest way to express my objection to the current language on both pages. Maybe this is

) 12:57, 3 September 2009 (UTC)

Another perspective: when a policy page implies, "I'm great, you should definitely pay attention", I'm tempted to put a {{peacock}} tag on it. Show, don't tell. It would be great if someone could come up with some kind of metric that would confirm what a lot of us know anecdotally, that policy pages are probably your best first stop if you're trying to get up to speed quickly on a topic, that the talk pages are probably the best mining grounds for people looking to see what everyone has said on the topic, and that policy is your best defense (and your worst enemy) at Arbcom. - Dank (push to talk) 13:59, 3 September 2009 (UTC)

Agreed, agreed, agreed! Honestly, at this point I'm basically ready to walk away. I'm kind of sad about it, since I really was ready to do some serious copy editing, organizing, and clarifying on some policy pages, but... why? A) this really just isn't worth the hassle, and probably most importantly is B) I've come to realize just how much that it really doesn't matter what policy literally says. Read and really understand

WP:IAR, both angles of it, and it becomes aparent that nothing else makes any significant difference. Do what's right for the content of the articles, while working with fellow editors, and everything will work out.
— V = I * R (talk to Ω

) 16:33, 3 September 2009 (UTC)

Well, I hope you'll continue to be involved ... if nothing else, just discuss what bugs you most so we can at least get whatever that is right. I'm not sure if the point of my personal story was clear ... the point is that I think everyone has had experiences like this, where they went to someone for expertise and then something about the tone suggests "conflict of interest" and puts them off. Even when what we're saying is fine, I think it would be a really good idea to avoid the tone that policy is oh-so-important. Information that backs up the claim would be fine, though. - Dank (push to talk) 18:10, 3 September 2009 (UTC)

I share your disposition toward this problem, but I want to suggest that part of the reason the policy pages have become so horribly authoritarian and bureaucratic is that people who don't like firm rules simply ignore the policies... which leaves them in the hands of the authoritarians! I don't want to frame this as an "us vs them" kind of thing, but I'm trying to bring out that people who are attracted to rules and bureaucratic thinking are drawn to pedantic editing of policy pages, whereas people who aren't, aren't. That means that our views are increasingly underrepresented in policy pages and gradually the authoritarians can use that fact to claim that their views are winning out. We really shouldn't walk away from this- we should stay in it and continue to work to make sure our views are heard. causa sui× 18:28, 3 September 2009 (UTC)

Good point. It's not really in me to actually walk away from this sort of thing anyway. Also, I'm not really frustrated so much as I've just become... resigned to the current state of affairs. I think that I'm kind of letting the naming conventions debacle that is in progress get to me, a little bit.
— V = I * R (talk to Ω) 18:40, 3 September 2009 (UTC)

I'm concerned about the tone only, not about the editors; my wiki-philosophy prohibits me from speculating on people's motives. - Dank (push to talk) 00:51, 4 September 2009 (UTC)

That's noble, but I'm not sure how it's a response to what I wrote... causa sui× 06:57, 4 September 2009 (UTC)

I'll respond privately, then report back here on whatever we can agree on. - Dank (push to talk) 12:58, 4 September 2009 (UTC)

←Okay, Causa said, "That isn't meant to accuse anyone in particular of being an authoritarian, but just to keep [Ohms] in the discussion." He thinks it's important for advocates of some flavor of IAR to stick it out in policy discussions, otherwise opposing views will shape the policy page. My position on IAR is something like this:

• Editors don't have to keep up with changes to policy pages; in fact, most editors deal with policy the same way most citizens deal with the law, by trying to avoid doing the things that get other people into trouble, rather than by consulting an attorney for every little thing. That works most of the time.
• OTOH, if you're unlucky or not careful and things go wrong, citing IAR won't usually fix whatever your problem is. IAR is historically a poor defense at Arbcom and ANI, and it's also a poor defense if you're doing something
  running for adminship
  .
• Even when policy pages aren't perfect and even though not everyone reads and follows them, they still perform a very useful function by superceding any other pages claiming to cover the same material. "Rules" should be in places that are well-traveled, public and noisy, and policy at least succeeds at that. - Dank (push to talk) 22:48, 5 September 2009 (UTC)

• I don't disagree with any of that, to be sure. I should note that those who actually live up to IAR should never have to deal with much in the way of dispute resolution at all, and IAR is not a defense in dispute resolution simply because by that point it's already failed. The whole key to IAR is "the best interest of Wikipedia", after all. I guess that my main issue is that I just don't want to spend significant amounts of time on disputes, and especially not on minutiae. I'd really rather just work on something else for a while, and either forget about it or re-approach the issue later. The real key to everything here is that we all need to work together, and that is done through compromise and consensus building, not through rules. That's the basis of my earlier point that these documents just aren't that important.
  — V = I * R (talk to Ω) 05:01, 6 September 2009 (UTC)

  The first point is key, and I think it's what is severely lacking in our drafting of policy pages right now, so I'm glad you said it. Rule #1 for writing policy should be "Once a person has read this page and grasped the main points, she should not have to reread it." Yes it's a good idea to brush up on the policy once in awhile, and so perfection in this principle is not feasible nor required: but it's a goal we ought to strive for as authors of policy.

  However, the second point is really baffling to me, and I can hardly contain my befuddlement every time it comes up. When I bring up how important IAR ought to be in our interpretation of policy, I usually run into this objection: "But there are people who use IAR to defend their bad ideas!" This is a mystifying objection to me, because IAR was never intended to be a get-out-of-jail-free card for people who do stupid and destructive things. Only someone who doesn't understand IAR could be fooled by the arguments of people who have screwed up and are being brought to book in front of Arbcom (or ANI, or whatever) who invoke IAR to excuse themselves from their patently destructive behavior. The proper response to such antics is at best to ignore them; at worst to erupt with jeers and laughter. The proper response is not to obscure IAR or pretend it doesn't exist because it is such a "problem" that vandals and miscreants distort and abuse it in such a preposterous way, though that has been the fashion lately.

  I don't think the motivation behind IAR is difficult to grasp if we dissociate ourselves from this noise. IAR says that if you want to do the right thing, you ought to do it, whether the policy authorizes you to do it or not. That doesn't mean that you won't have to face democratic accountability if someone challenges your actions. It doesn't mean that you don't have to explain yourself to others because you can pull the "IAR card" out of your sleeve and escape peer review from your fellow editors. It's the "rules are rules" crowd who have contributed most to this grave misinterpretation of IAR because the strawman best suits their purposes in getting rid of it. I wish we could put that strawman to rest. causa sui× 00:32, 9 September 2009 (UTC)

I think I agree with your main points, but with some reservations. Most policy pages don't strike me as "severely lacking", although I'm not currently happy with the language on this one. And on some policy pages, firm rules are fine with me. For instance, at

WP:CSD, we're not saying what pages you can and can't delete, we're saying that from long experience, we know that there are only a few kinds of pages where almost everyone agrees on the deletion; if it's not one of those, then people might differ so it should go to AfD. That's not "thou shalt not", that's just another way of arriving at the same deletion result. My main reservation is that I can't jump to labeling someone as being in the "rules-are-rules crowd" if they say "I think this is the rule we should all follow". Even small societies have hard-to-decipher rules that impede outsiders, and Wikipedia is a very large society, full of competing interests and unwritten rules. Maybe the person who says "please follow this rule" knows from experience that if we don't have that one rule, it means that we'll have two rules, or twenty, all enforced in various ways by various cliques in a way that's opaque to outsiders. Some rules are anti-authoritarian. - Dank (push to talk

) 02:08, 9 September 2009 (UTC)

You make some interesting points, but again, I'm not sure what they have to do with what I said. You seem to be responding to views that I wouldn't attribute to myself. As for the CSD debate, I'm not sure how that fits in to what we're talking about either. causa sui× 17:15, 9 September 2009 (UTC)

You seemed to be expressing a distaste the "rules are rules" crowd, whoever they are. I thought I was responding to that ... can you repeat what you're saying, and give me an example of what you don't like? - Dank (push to talk) 17:39, 9 September 2009 (UTC)

P.S. CSD popped into my head as one of the better examples of where a firm rule is not that harmful, since the effect of the rule isn't "You can't do this", but "You have to take more time and involve more people if you want to do this". - Dank (push to talk) 17:52, 9 September 2009 (UTC)

Regarding just letting things go on as they have - I too am frequently discouraged, but I keep two things in mind: first, most cleanup really does remain, especially after major disputes are resolved. Second, 'people get it anyway, and will do what they like' applies only to experienced editors. Policy pages may not have much effect on them (even 3RR is overlooked for well-known editors), but policy pages do help our newer editors. One major role of policy pages is to prevent Wikipedia from becoming a Meritocracy of contribution count, where we have nebulous rules determined by whoever's around giving their opinion at some one time.    M   21:30, 6 September 2009 (UTC)

I just reverted an edit adding a section that described this essay as the "main article". Atomic Blunder, everyone is welcome to make changes to policy ... including you, including me. (I haven't been making any non-trivial edits to policy for a while because, in general, I can't be objective enough to do the

) 17:48, 9 September 2009 (UTC)

I'll give it another try. --Atomic blunder (talk) 19:58, 9 September 2009 (UTC)

Okay, thanks. Btw, I'm wrong ... tag-teaming is specifically mentioned on a policy page, at

WP:OWN#Multiple editors. - Dank (push to talk

) 20:23, 9 September 2009 (UTC)

I've removed the paragraph on tag-teaming, which is too much just a bad faith accusation. The spirit of WP:TAGTEAM is anti-consensus, an accusation of bad-faith editing against multiple editors. What little good in WP:TAGTEAM is better described on the policies and guidelines it links to, such as WP:MEAT in the lede. --

talk

) 21:52, 9 September 2009 (UTC)

Writing

I don't know who's changing what here, but the writing is really suffering. Parts of the current lead are practically meaningless. I suggest we find an earlier version to return to. SlimVirgin ^{talk|contribs} 05:18, 22 August 2009 (UTC)

I went back to a 2007 version, which I tweaked a little. We should ideally go through every section and make sure that each sentence actually says something. Much of the previous lead didn't really say anything at all. SlimVirgin ^{talk|contribs} 05:28, 22 August 2009 (UTC)

I don't disagree on the whole. I do disagree with your edit summary "has nothing to do with Dunbar's number", RE "Try not to attract too many editors at once." It does have something to do with Dunbar's number. However, for me anyway, the number that overwhelmes me on a single page is more like 5 than 150. I think "do not attract too many at once" is good advice, but I think some (any) justification should be offered. --SmokeyJoe (talk) 07:29, 22 August 2009 (UTC)

Dunbar's number is a kind of tipping point, beyond which AGF doesn't work for diverse reasons (at which point, the prerequisites for consensus no longer exist). There is a different Dunbar's number for each environment, and it can be empirically determined (though I don't think we ever tried on Wikipedia. That might be a good question for Erik Zachte!) . It can be as low as say 50 people in an online enviornment. Perhaps even less. While 5 people is quite a handful, I don't think it exceeds Dunbar's number. :-P

The use of diverse Dispute resolution techniques all typically meatball:ExpandScope. If taken too far, it is certainly possible to exceed 50+ participants using dispute resolution, and that tends to bring the house down at a particular location. At that point things become unmanageable, and we get all these governance reform types whining about meatball:CommunityMayNotScale, and asserting that consensus doesn't work. Well yeah, it works just fine, as long as you don't deliberately break it! :-P

There are other ways to expandscope too far too fast too. For instance, the

proposal ultimately failed because at one point there were as many as 880 participants. That alone was sufficient to kill it.

At the moment, there are several places where people are advised to draw in more participants "to get a broader consensus", whatever that means :-P . This can lead to broader wikidrama instead, if people blindly follow that advice, and forget about Good Old Dunbar.

So we're going to have to mention this someplace. Any idea where? --Kim Bruning (talk) 08:50, 22 August 2009 (UTC)

Kim, not to rehash old debates... but ultimately the ATT proposal failed because it took the broader Wikipedia community by surprise. We had a core group of perhaps 20 policy wonks (with occasional comments and tweeks by around 300 others) who toiled away for months arguing and reaching a local consensus... but the broader community knew nothing about it until it was suddenly implimented. What killed ATT was the ham-fisted way it was implimented, and the (understandable) knee-jerk reaction of the broader community to waking up one day to find three of the core policies suddenly "GONE" and replaced with something (gasp) NEW. (I know this wasn't the intent of those involved... but that was how the broader community saw it). Blueboar (talk) 13:59, 22 August 2009 (UTC)

That's exactly what happened, from my perspective. --SmokeyJoe (talk) 21:31, 22 August 2009 (UTC)

That's not what happened, Blueboar. It had been discussed on the mailing list, posted to the various policy pages, and announced on all the public pages we could think of. People were emailing to say the content policies were clear to them for the first time. It was Jimbo who singlehandedly overturned it, even though he had also been told about it weeks in advance. We then got a wiki-wide majority for it, even though Jimbo had opposed it; it just wasn't the two-thirds majority we needed, which is very difficult to get when you involved hundreds of people. SlimVirgin ^{talk|contribs} 19:25, 22 August 2009 (UTC)

You miss my point SV... yes, all that took place... but why was Jimbo brought in to oppose it in the first place... because people were blind sided by it. Most of the community did not know that ATT was even being contemplated until it was implimented. It took them by surprise and they complained to Jimbo. All else followed from that. Blueboar (talk) 19:49, 22 August 2009 (UTC)

Jimbo wasn't brought in to oppose it. He was arguing with someone that using primary sources in BLPs is OR. He was saying it was, and the others were saying it wasn't. NOR backed him up (as did ATT). He therefore went to quote the NOR page, and saw that someone had added a tag to it saying "superseded" or something, which really it would have been better not to do, because the concept of NOR was not superseded at all. So he undid the tag, and came to ATT arguing against it. Until that point, it had been working extremely well as a policy. People liked it. No one had complained to Jimbo. SlimVirgin ^{talk|contribs} 19:57, 22 August 2009 (UTC)

I get the idea that Jimbo deliberately expanded scope. He does know all the old wikis, and he is an expert, albeit a tad rusty these days ;-) --Kim Bruning (talk) 20:04, 22 August 2009 (UTC)

You get the wrong idea. I know what happened, and that wasn't it. SlimVirgin ^{talk|contribs} 20:06, 22 August 2009 (UTC)

I was there too. I predicted the poll outcome on the day it opened. ;-) --Kim Bruning (talk) 20:13, 22 August 2009 (UTC)

SV... I agree that you were plugged into the back channel discussions, and I was not... so I will bow to your "insider's knowledge" on the Jimbo issue. However, this does not change my assertion that the average editor felt blindsided by ATT's implementation. There were a shit load of comments saying effectively "Whoa... Nobody told me about this... I want my WP:V!" For that part of it, I was there. Blueboar (talk) 20:31, 22 August 2009 (UTC)

Yes, they didn't expand scope gradually enough, they exposed the whole community to it from one day to the next, and therefore there were suddenly hundreds of new people discussing from one day to the next.

Same situation, same numbers, just a slightly different way of saying it, and the different viewpoint allows drawing of extra conclusions, IMHO.

Otherwise it sounds like we're pretty much in agreement here? :-) --Kim Bruning (talk) 14:35, 22 August 2009 (UTC)

Back to the writing

Could people please try to keep the writing tighter and more readable? There's a lot of material on the page that makes no sense. For example, what does this mean? "Minority opinions typically reflect genuine concerns, and their (strict) logic may outweigh the "logic" (point of view) of the majority." SlimVirgin ^{talk|contribs} 19:32, 22 August 2009 (UTC)

URK. I'm of to wikimania as of tomorrow. How about we look into some of that when I get back? I totally agree that we need to do some tidying.

I'll admit I'm not entirely AGF-ing, and I know you're experienced and everything and you won't really do anything evil, but I'm just a little worried about coming back and reading things like "Consensus is non negotiable" or "You cannot edit consensus without consensus" ... Do I need to be afraid? ;-) --Kim Bruning (talk) 20:01, 22 August 2009 (UTC)

Perhaps someone else can explain what it means? I don't know who wrote it originally. SlimVirgin ^{talk|contribs} 20:02, 22 August 2009 (UTC)

Well, say there are 20 people (majority) saying that 1+1=4, and 10 people saying that 1+1=2.

Even though the opinion is outnumbered two-to-one , it is still true that 1+1=2.

Hmmm, have you ever watched

12 angry men ? One person convinces all the others that he is right, and in the end he gains consensus for his position. It's interesting to see how the movie portrays it. :-) --Kim Bruning (talk

) 20:11, 22 August 2009 (UTC)

What is "strict logic." What is "'logic' (point of view)"? SlimVirgin talk|contribs 21:39, 22 August 2009 (UTC)

The very reason why those twelve men were so angry... :)

Dreadstar ☥

21:52, 22 August 2009 (UTC)

Hmm, I think Strict logic would be Logic, while "logic" (point of view) is simply a point of view, and not logic at all (strictly not logic ;-) ) . Hmph. That's fairly twisty wording.

How about putting it this way?

"Minority opinions typically reflect genuine concerns. Just because someone is in the minority, doesn't mean that they are wrong. It is important to evaluate every argument based on its merits, not based on who it comes from, or whether they currently are in the minority or in the majority.". Is that clearer? --Kim Bruning (talk) 22:46, 22 August 2009 (UTC)

Less is more when it comes to writing -- almost any kind of writing, but particularly policy. "Minority opinions may reflect genuine concerns" is all that's really needed. SlimVirgin ^{talk|contribs} 23:35, 22 August 2009 (UTC)

Hmmm, you have a point about being succinct. But the proposed wording is not strong enough, I feel.

The point is that the minority has just as much chance at being right as the majority does. We want the best ideas to prevail, not the biggest numbers. Does that make sense?

Let's see. How about: "Ideas are judged on their merits, not by whether they are held by the majority or by a minorty". Will that do? --Kim Bruning (talk) 23:44, 22 August 2009 (UTC)

Again, you need to decide whether you believe "descriptive not prescriptive." As a matter of fact, ideas on Wikipedia are rarely judged on their merits, but by how many people believe something. You can say that a vote is not a vote until you're blue in the face, but it usually is, especially for anything contentious. So: do you want this policy to say what is the case, or what ought to be the case? SlimVirgin ^{talk|contribs} 00:07, 23 August 2009 (UTC)

Neither. I want to state what we have found to work *best* in reality. If we tell people just what happens a lot, we'd also have to tell them to vandalise the wiki ;)

Over time, we've found that the best results are found through discussion and editing. Ironically, voting actually disenfranchises people.

There's also another reason to discourage voting: Voting on encyclopedic content is disastrous (imagine a majority agreeing that 2+2=5 ;-) ). So we need to use consensus in the main namespace. By encouraging use of consensus elsewhere, we can ensure that people don't get tempted.

numerous people on wikipedia therefore act to discourage voting, and they explain to people why straight voting is a bad idea.

I've also seen many people use polls properly to gauge consensus . I hope you're one of those people! :-) --Kim Bruning (talk) 00:20, 23 August 2009 (UTC)

Kim says, "The point is that the minority has just as much chance at being right as the majority does."

Kim, this is simply not true -- not in the real world, and not in the average editor's experience of Wikipedia's working. For example: if you compared every modern evidence-based pharmaceutical drug to the best "alternative medicine" counterpart, you will not find that the "minority view" is better in 50% of the cases. Similarly, if you've got many editors saying "This former head of the IAEA is clearly an expert on radiation safety" and one claiming that anyone that's been hired by a utility company is a hopeless shill (as we did at RSN this week), the minority view does not have equal odds of being correct simply because it's a minority view.

Editors of an enterprise that purports to be factual would be well-served by remembering that minority views are usually in the minority for a sound reason. WhatamIdoing (talk) 21:02, 24 August 2009 (UTC)

Hmm, I may have misspoken, I did not mean to imply 50/50 odds. What I do mean can easily be illustrated as follows:

Example:

I always had the impression that -for instance- homeopathy has a rather large number of adherents, despite lack of evidence in its favor. It is quite possible to be the single scientist in a room full of homeopaths, for instance. ;-)

So I'm a scientist, and say I walk into a room full of homeopaths. I am now greatly outnumbered. Does that make me automatically wrong, and will homeopathy suddenly start to work? ;-) --Kim Bruning (talk) 14:53, 11 September 2009 (UTC)

Of course you are wrong... no scientist should ever walk into a room full of homeopaths! :>) Blueboar (talk) 20:55, 11 September 2009 (UTC)

The page implies that the "Wikimedia founding principles" have some sort of bearing on our project. This appears to be a freely editable page at Meta, with various language versions that are not necessarily synchronized, and whose contents (and changes thereto) are probably ignored by a good 99.9% of the Wiki supercommunity. In other words, whatever's written on that page probably doesn't mean anything profound. If I'm right about all of this, can we get rid of the paragraph here that refers to that page?--Kotniski (talk) 16:44, 14 September 2009 (UTC)

Well, yes, but I'd do it for a different reason: because there isn't any debate that I know of about whether consensus is how we operate, so we don't have to take the WMF's word for it. We owe them and all that for getting a good thing started, but we've gone way past the original conceptions. OTOH, if it's true that all of the WMF projects follow a similar approach to consensus, it wouldn't hurt to mention that. - Dank (push to talk) 18:50, 14 September 2009 (UTC)

The page I'm talking about isn't the WMF's word anyway, as far as I can tell. (And didn't Wikipedia exist before the WMF?)--Kotniski (talk) 08:44, 15 September 2009 (UTC)

• In response to Dank's remark, I just want to say that there's no need for a debate to establish that consensus is how we operate. That's already well-established, and besides, the act of holding a debate about whether consensus is how we operate would, itself, prove that we operate via consensus, so it would be tautological thinking.

  Besides which, a bit of experience with Wikipedian debates tells me that if we debated the matter, we'd end up with a consensus like this:

  "There is a consensus that Wikipedia operates by consensus, except for BLPs where there is no consensus whether or not we operate by consensus, flagged revisions where there is a consensus not to have a consensus, and discussions about consensus, where there is a consensus that consensus is not how we operate but there is no consensus about what replaces consensus, except for copyright violations and breaches of personal privacy, oversight requests, arbcom, and office actions."

  For the sake of sanity, please, let's just assume we operate by consensus.  :)—S Marshall ^Talk/_Cont 11:26, 15 September 2009 (UTC)

That's funny and scary at the same time. - Dank (push to talk) 12:09, 15 September 2009 (UTC)

Since no-one's providing any reason to continue referring to the page I mentioned, I'm going to remove it for the moment.--Kotniski (talk) 11:31, 15 September 2009 (UTC)

Kotniski, I've got some questions about your edits today. Here's my position:

1. You can land at ANI for not following certain policies even once, for instance
   WP:OUTING
   . But there are very few sentences in policy pages that have that kind of clout. Other policies are likely to land you at ANI if you ignore them 10 or 20 times. But infringements of most of the sentences on policy pages will never put you in danger of a block by themselves (though they make it more likely you'll get reverted). The worst that can happen is, if you're already in trouble for something else, policy infringements will give people extra ammunition to shoot you down with, so if you like to skate on thin ice, it's a very good idea to keep up with the relevant policy.
2. If you're trying to develop a good reputation, trying to give the impression that you keep up with developments in, say, content-related matters, then reading and either following or working to change content policy pages should probably be first on your list.
3. If you're trying to get up to speed on some new area, you'll probably wind up regretting it if you avoid the relevant policy pages; it's generally good stuff.
4. If you create a guideline of some sort that doesn't square with policy, then it's very likely that your material will get ignored and/or reverted. Even when policy isn't perfect, even when people don't follow it, there's still a strong consensus that policy pages are the place where the action is supposed to happen (which is why we are generally very reluctant to promote pages to policy status ... it's a high bar).

I don't know of any other downsides to not following policy, and I don't understand what it means to say that people have to follow it. We also get a lot of flak for "thou shalt" ... it gives the appearance of COI, which lowers the credibility of the document. - Dank (push to talk) 12:38, 15 September 2009 (UTC)

That all sounds very true... but I guess I'm not understanding what it has to do with my edits?--Kotniski (talk) 12:46, 15 September 2009 (UTC)

This one. I don't disagree with what you're thinking, the problem is that some editors have a bad reaction if it sounds like "You can't", "You must", "You'll be sorry", or "We know best". (I'm not saying that's what you're saying, I'm saying that's what some people hear.) Individuals and groups of editors can and do decide not to follow policy; it happens all the time. - Dank (push to talk) 13:25, 15 September 2009 (UTC)

Well, all I was doing was basically rearranging the information for clarity and logic - I don't see that I added anything that wasn't there before.--Kotniski (talk) 13:55, 15 September 2009 (UTC)

"A wikiproject cannot decide that ..." wasn't there before. - Dank (push to talk) 14:06, 15 September 2009 (UTC)

It was, a bit lower down (and worded ever so slightly differently, but "cannot" was certainly there).--Kotniski (talk) 14:12, 15 September 2009 (UTC)

Sorry, I didn't notice the wording further down. I still object ... that's why I'm watchlisting here and at POLICY, to try to get people to tone down the current wording and be more specific. - Dank (push to talk) 15:03, 15 September 2009 (UTC)

Hammersoft and I are having a discussion at

) 17:31, 16 September 2009 (UTC)

Talk page character

I just saw this recent edit summary: "Talk pages are also subject to consensus." and I wanted to start a conversation about it. I should probably point out that I agreed with Dank's earlier change, and I don't necessarily disagree with this one. I'm simply slightly confused by the reasoning given in the edit summary, so I would be interested in seeing that thought fully expressed.
— V = I * R (talk to Ω) 01:11, 26 September 2009 (UTC)

Chillum removed "non-talk" here: "Someone makes a change to a [non-talk] page, then everyone who reads the page has an opportunity to leave it as it is, or change it. When two or more editors cannot reach an agreement by editing, consensus is sought on

WP:BOLD says: "Be bold and drop your opinion [on the talk page]", it doesn't recommend bold editing of other people's comments. The other two removals of my addition of "non-talk" don't look right to me either ... on the other hand, "non-talk" is a non-word. If anyone has something snappier, feel free to edit. - Dank (push to talk

) 01:49, 26 September 2009 (UTC)

Okay, reverting since we're getting close to the end of the month, but feel free to revert and discuss. - Dank (push to talk) 12:31, 27 September 2009 (UTC)

Btw, the changes so far this month seem reasonable to me, but changing for instance "expected to discuss substantive changes" to "typically expected to propose substantive changes" might generate some discussion. Since the recent changes seem relatively uncontroversial and the page has more stability than it had last month, I'll assume for porpoises of the Update that I can use the current version, unless there's discussion that suggests otherwise. - Dank (push to talk) 02:21, 26 September 2009 (UTC)

I haven't actually looked at this for a couple of weeks now, but I don't think that anything has significantly changed. There seems to be a faction of people who feel that even minor changes to policy or guideline pages are significant (and changing a single word certainly could be significant; I will happily dispute the correlation → causation perception there, though), but I don't think that has actually occurred recently. All edits appear to be straight copy editing (which is a very good thing!).
— V = I * R (talk to Ω) 03:57, 26 September 2009 (UTC)

There's been some discussion in the past on whether it's okay to revert edits by banned sockpuppets on sight, by analogy to

) 03:30, 26 September 2009 (UTC)

For what it's worth, I agree with your removal. As a matter of fact, I'm actually tempted to nominate that page for deletion, which is extremely unusual for me. It's one saving grace is that it is an essay... that and deleting things here is just bad on principle, in my view.
— V = I * R (talk to Ω) 03:57, 26 September 2009 (UTC)

What's the deal with these changes? HarryAlfa seems to show up every now and then and put the text (about working to a "system of good reasons") in; someone else reverts; and there seems to be no discussion about it. On the face of it the change seems reasonable - we shouldn't be counting heads, but comparing reasoning - what's the objection?--Kotniski (talk) 10:16, 4 October 2009 (UTC)

"non-reasoning nay-saying is anti-Wikipedian" is not a good language. Ruslik_Zero 11:05, 4 October 2009 (UTC)

Can't we improve the language, instead of just deleting it? Don't we agree with the sentiment?--Kotniski (talk) 11:13, 4 October 2009 (UTC)

I do not like the whole sentence: If reasons are offered to support a position, but that position is opposed, then counter-reasons should be given or different conclusions should be explained; non-reasoning nay-saying is anti-Wikipedian.. It is pointy, and HA seems trying to solve his own problems rather than to improve the policy (remember discussion on Talk:Link). Ruslik_Zero 11:53, 4 October 2009 (UTC)

Yes, I certainly remember that... but this principle seems pretty sound (people saying "no consensus for this" or "I oppose this" without giving reasons don't exactly help the editing process along). Maybe the point is already sufficiently made on this page though.--Kotniski (talk) 12:09, 4 October 2009 (UTC)

I object to the language like "should be given". It is a volunteering project, and nobody "should" do anything. It is another matter when well argumented opposes or supports carry more weight than "per smbd". But the latter is already covered in Wikipedia:Consensus#Community_discussions_and_polls (and the language is much better!). Ruslik_Zero 13:10, 4 October 2009 (UTC)

The title of this section is from the text of the previous stable version of the document before this major re-write. I've made two edits, one restoring the text & one addition of my own which I believe reflects Wikipedian ethics. HarryAlffa (talk) 13:54, 16 October 2009 (UTC)

Does anyone else think it's ironic that we've got a slow-burning edit war over whether slow-burning edit wars (as opposed to reasoned argument) are bad? - Dank (push to talk) 16:32, 16 October 2009 (UTC)

Is it edit war or is it wikihounding by Ckatz and Ruslik? HarryAlffa (talk) 20:04, 16 October 2009 (UTC)

I thought it a supreme irony when Ruslik described an edit as "is not a good language". HarryAlffa (talk) 20:07, 16 October 2009 (UTC)

Another irony; when on this page, "people saying "no consensus for this" or "I oppose this" without giving reasons don't exactly help the editing process along", with Ruslik and Ckatz edit summaries of: "no consensus for this changes"; "rv. undiscussed changes"; "no consensus for this change". HarryAlffa (talk) 20:39, 16 October 2009 (UTC)

I'm adding a principles template at the bottom like a lot of other policies as this is a major policy. I don't think this is controversial but there s some discussion on

) 11:22, 27 October 2009 (UTC)

CONEXCEPT

"Some exceptions supersede consensus decisions on a page. Declarations from Jimbo Wales, the Wikimedia Foundation Board, or the Developers, particularly for copyright, legal issues, or server load, have policy status."

Hang on. Does that mean that Jimmy Wales' opinion on anything from whether a specific article should be kept to whether paid editing is prohibited goes without question, overriding consensus? Or does the 'declaration' have to be made in some more official manner? Fences&Windows 02:10, 30 October 2009 (UTC)

Examples where Jimbo's word has been shut down an idea include Wikipedia:Bureaucrat removal (others might be able to come up with more). The fact that Jimbo may overrule consensus has been understood to be true since essentially the site's creation. Some of the old-timers might remember more specific than the one I just mentioned. NW (Talk) 04:37, 30 October 2009 (UTC)

But the question is whether Jimbo's word overrules consensus by virtue of its being Jimbo's word, or only if he says that he's overruling consensus. I believe it's the latter (i.e. Jimbo doesn't always get his own way, unless he states explicitly that he's having his own way). So perhaps the wording needs clarifying. Also it's not really that they have policy status - they have a higher status than policies.--Kotniski (talk) 10:58, 30 October 2009 (UTC)

We need to avoid treating his comments as though they were legal precedents. In his comment on adminship, he said that the community are not the only ones to decide who becomes an admin. We don't need to interpret that to mean anything else.

The background to this is the dispute at

WT:PAID. If what Jimmy Wales says is policy, then all the wrangling can stop and we can have a simple statement of his policy, namely: "No editor may sell their services as a Wikipedia editor, administrator, bureaucrat, etc. Paid advocates may never directly edit Wikipedia". Fences&Windows

23:42, 30 October 2009 (UTC)

The relationship between things that we consider true because they seem to have consensus and things that we consider true because someone said so (the lawyers, Jimmy Wales, the WMF, ArbCom) is a subject that just never dies. This would be a good topic for the policy report in next week's Signpost. - Dank (push to talk) 17:07, 1 November 2009 (UTC)

The Policy Report in the Signpost two weeks from today will feature this page. See other Policy Reports to get an idea what we're looking for (here's tonight's, for instance). - Dank (push to talk) 20:13, 7 December 2009 (UTC)

Note: Policy Report cancelled for this page, see below. - Dank (push to talk) 04:32, 13 December 2009 (UTC)

Your comments on our WP:Consensus policy will be featured in the Policy Report 9 days from now in the Signpost. If it helps, monthly changes to this page are available at WP:Update/1/Conduct policy changes, July 2009 to December 2009. I'm leaving messages for everyone who has edited this talk page over the last two months, but all responses are welcome.

A paradox of modern democracies is that voters generally have a low opinion of national politicians, but tend to trust and re-elect their own representatives. I think the same thing goes on with policy pages ... some people^[who?] distrust policy pages in general but like the pages that they're working on. That's the point of the weekly Policy Report, to let people look at policy pages through the eyes of the people who work on the page.

To get an idea of what kinds of questions and answers the community is interested in, see the archives of this talk page or the previous surveys at

) 15:36, 12 December 2009 (UTC)

Responses by Kotniski

I think what's written on this page is a compromise between the philosophies of the relatively few editors who are interested in it, and probably doesn't have any significant impact on what happens throughout the project. It does quite a good job of reflecting the vague norms that exist, but doesn't go so far as providing useful concrete information. The subject of the page, "consensus", is of course left undefined, which is both a strength and a weakness. In practice Wikipedia muddles on; anything that is produced and remains reasonably stable is said to represent consensus, so to say that consensus is our model for decision-making is perhaps little more than a tautology.

Responses by Fences&Windows

The debate at WT:Paid editing (guideline) might have benefited from a wider realisation that Jimmy Wales and the Wikimedia Foundation can unilaterally decide policy. Wales stated that it is policy that no editor may sell their services as a Wikipedia editor, administrator, bureaucrat, etc., which makes efforts to reach community consensus on this particular point redundant. Compared to the processes described at 1% rule and the Pareto principle .

Responses by Proofreader77

Commented on civility , but must raise the issue of "consensus" in the context of highly contentious, current-events inflamed articles. Most editors who know enough to make a fair judgment of consensus (in the context of all relevant policy) often know better than to waste time in a "barfight" among new and more experienced users who may nonetheless not constrain inner POV in a "passion-filled" arena. Now, into this barfight strolls a, "current-events wrangling" Wikipedian with the experience (in general+topic specific) to have "good judgment" of what "consensus" of policy-conscious (and non-POV inflamed) editors would tend to be. Badge, please. :-) Proofreader77 (talk) 01:07, 13 December 2009 (UTC)

Responses by next user

Responses by next user

Responses by next user

Procedural policy

After looking at the responses here and at

) 04:32, 13 December 2009 (UTC)

That sounds wonderfully meta ... I like meta ... Will ponder this. :-) Proofreader77 (talk) 04:36, 13 December 2009 (UTC)

P.S. Another option that would work for me would be to change the 7th section over at

4th pillar. We'd have to be careful, though, that "principle" doesn't come to mean "inviolable, uber-policy" if it's being applied to this policy page, since this policy page isn't above any other. - Dank (push to talk

) 04:50, 13 December 2009 (UTC)

Probably just a repetition of my comment above, but I actually hadn't given any thought to how these kinds of "things" fit together in a conceptual structure. What you (Dank) have just done (which may or may not be trivial to you) makes this "something to think about." Usually that means something really important may be in there ... like a pony? (Sorry, silly drive-by little-horse humor ... Did I mention I'm thinking about Westerns? :-) Proofreader77 (talk) 05:02, 13 December 2009 (UTC)

Okay, there's been no objection over at WT:POLICY, so I'm going to try an experiment that I expect won't do any harm and might be helpful ... without fiddling with the policy status, I'm moving this page from the conduct policy subcat to the

WP:5P is in. Feel free to revert, and please see the discussion at WT:Policies_and_guidelines#Agreed. - Dank (push to talk

) 19:22, 13 December 2009 (UTC)

Pointer to RfC

Please see

) 12:47, 17 December 2009 (UTC)

Heads up: This page never defines consensus

• This page never defines consensus: is it number of arguers, or weight of arguments? Don't even dream of weaseling out by saying "both." • 
  Ling.Nut
  03:54, 8 December 2009 (UTC)

I would define it as "number of editors"... after they have considered all the arguments and given them due weight. It isn't simple majority... but it is not unanimity either. Blueboar (talk) 04:18, 8 December 2009 (UTC)

But your answer resolves to simply "number of editors." Everything you wrote after that is (albeit unintentionally) meaningless noise, because POV gangs do not consider arguments or give them due weight. Always and everywhere, this is Wikipedia's foremost fatal flaw. • 

Ling.Nut

04:23, 8 December 2009 (UTC)

Could you point me to a policy page where the POV gangs have enforced their POV? - Dank (push to talk) 04:29, 8 December 2009 (UTC)

Policy? I'm talking about content. I'm sure there's gang-editing and tag-team 3RR and similar on policy pages, but I wouldn't know where. • 

Ling.Nut

04:42, 8 December 2009 (UTC)

It would be astonishing if none of our editors ever colluded on their favorite articles, and that's a shame, but it doesn't seem to be fatal to our mission, judging from our growth and reputation. But is it true as you claim that consensus is meaningless, because the gangs get to have their way? If that were true, would they put up with policies that make it dangerous to bite newbies, make personal attacks, sock, publish their own opinions, wield authority unilaterally, etc? I would think these rules would chafe a successful, self-assured gang of bullies, but I don't see evidence that such a gang has been able to enforce their will in any of the policies. You would think they'd win at least one policy fight, if they existed. - Dank (push to talk) 05:04, 8 December 2009 (UTC)

(undent) Do you actually edit in mainspace these days, Dan? No offense. Seriously asking. And let's not forget the start of this thread: this page doesn't define consensus. • 

05:10, 8 December 2009 (UTC)

None taken. I'll make a very bad showing in the

Wikicup if I don't. - Dank (push to talk

) 05:12, 8 December 2009 (UTC)

I'm happy with consensus as defined in the policy. Do we need to tie it down for the wikilawyers so they can exploit it to cause trouble?

Dmcq (talk

) 11:00, 8 December 2009 (UTC)

Consensus is never defined. • 

Ling.Nut

11:45, 8 December 2009 (UTC)

This policy describes how the wikipedia community uses consensus in building wikipedia. This is not a dictionary, it is a policy. If you think the community does something different then the policy should be updated. However I think it is a fair and reasonable description. Have you got an instance of a problem you have with it? Were you looking to it to resolve some content dispute for instance?

Dmcq (talk

) 12:22, 8 December 2009 (UTC)

I had a look at your talk page to see what the problem is and it looks like you're in an edit war about climate change. I'm afraid blaming consensus for your problems there is a bit wide of the mark. It would be better to cite verifiability for anything about climate change, there's lots of people with an axe to grind and expecting them to agree by consensus is rather pie in the sky. The debate over creationism versus evolution might be a better model for your debate though I think it is more akin to those denying men walked on the moon.

Dmcq (talk

) 12:32, 8 December 2009 (UTC)

Verifiability is out as a useful tool. People can easily find malicious statements by pro-AGW people characterizing global warming skeptics as denialists. Then they can easily stick them in an article and hide behind WP:V, even though the original text is harshly POV. See my recent comments re: Barbabra Boxer as a reliable source. NO, basically, Wikipedia is utterly useless for controversial issues. Its articles ALWAYS assume a POV. NPOV is a polite fiction we all mumble to ourselves and pretend it exists; it's a

Ling.Nut

10:17, 10 December 2009 (UTC)

Do you have any suggestion as to how it might be defined?--Kotniski (talk) 10:49, 10 December 2009 (UTC)

(Note, I agree/amplify Kotniski somewhere below, but disagree with their choice here to deflect the negative characterizations and request a positive response — "negative" perspective is valuable.

• re: "Wikipedia is utterly useless for controversial issues." [Absolutely ofttimes true. :-)]
• re: "magical thinking" — Let me take this throw-away and bank shot into my saying: "Consensus is a magic word." A "magic word" wielded by those who have the "wiki-juice" to make it stick — or are simply yelling it as if they did.

Clearly that needs elaboration, but at the moment ... it's Christmas. :-) Proofreader77 (talk) 08:47, 25 December 2009 (UTC)

(undent) Several things need to be changed, but only one of them would take place here. First, consensus should reflect weight-of-arguments, not number of people shouting "Ditto!" on one side of the issue (as is currently the case; try not to deny it). Second, and perhaps most importantly, the dispute resolution process is f*cking useless. It should be drastically cut and streamlined, with much more advice and much more useful explanations, far fewer labyrinthine forums accessed via a jumbled jungle of cryptic links, and many more useful admins (note here that any number would be greater than the current number == zero). Third, articles should adhere to NPOV in their naming conventions etc., as an offshoot of the (completely undefined! at his moment) concept of Consensus. • 

11:29, 10 December 2009 (UTC)

• "Consensus" is a dynamic and fuzzy thing that defies a precise definition. --SmokeyJoe (talk) 11:33, 21 December 2009 (UTC)

Consensus might be ill-defined but its objective is clear: it's a process that leaves an article in a stable state. That aside there is a numerical definition as well. The 3rr rule prevents more than three reverts in a 24 hour period, so in that sense in a "4 against 1" debate then the majority can enforce a consensus. Four different editors can each make an edit that can only be reverted up to three occasions by the other editor. The 3rr rule prohibits edit warring but you can't engage in an edit war unless you make more than one edit, so 3rr does define consensus as a body of opinion with an 80% majority. Betty Logan (talk) 20:41, 21 December 2009 (UTC)

Consensus is actually fairly easy to define - I've done it a few times in different places (including this page) on wikipedia. The problem is that a variety of editors don't really want consensus (because consensus is difficult, time consuming, and requires editors to be patient and cooperative with other editors that they frankly disrespect), and so there's always a strong effort to prevent any definition of the word from being instituted. There's a a large core group of wikipedians who would prefer a policy-driven authoritarian structure which would allow 'disreputable' editors and viewpoints to be excluded quickly and efficiently, without the long, drawn-out political efforts that are currently required to guarantee article content and drive away undesired editors, and the ambiguity around consensus allows them a lot of authoritarian leeway they might not otherwise have. sad, but true. --Ludwigs2 22:41, 21 December 2009 (UTC)

Would you please link to some of these definitions? --SmokeyJoe (talk) 20:40, 23 December 2009 (UTC)

Do you think that having a definition would really be the best solution for this problem? I'm worried that if you clearly define consensus then you will have created a safety zone for exactly that kind of behavior. For example, refer to the ad-hoc firing squads at

WP:AN/I where if you happen to get a certain number of yes votes from whoever happens to be watching the page for a 15 minute time window, you can get someone "community banned" due to "consensus". If you define consensus clearly, then you've given policy wonks and authoritarians a clear goalpost to cross, past which they can say "we already got consensus for this, so shut up." causa sui×

23:10, 21 December 2009 (UTC)

(Rhetorical note/hear hear: re causa sui:) "the ad-hoc firing squads at

WP:AN/I where if you happen to get a certain number of yes votes from whoever happens to be watching the page for a 15 minute time window, you can get someone "community banned" due to "consensus". [End quote by causa sui] -- Proofreader77 (talk

) 08:24, 25 December 2009 (UTC)

no, that's not the case. these kinds of things happen because people can redefine consensus at will to include or exclude only those people they want to count in any given instance. I can't count the number of times I've seen people pull the "we have consensus even though lots of people disagree" game, which is a purely authoritarian move designed to silence opposing views. they can only do that because consensus is so ill-defined that they can assert that they have it, and there's nothing in policy that proves them wrong. Having a good metric of consensus in policy is important if you actually want a consensus system, otherwise you'll inevitably end up with a shout-ocracy (rule by the loudest and most obnoxious). --Ludwigs2 01:00, 22 December 2009 (UTC)

Hmmm, what I usually see is one side saying "You don't have consensus for that" countered by the other side saying "no you don't have consensus"... "Do too"... "Nuh-uh"... Etc. Until someone runs to an adult and the page gets locked.

My point is that most consensus related debates usually center on accusations of a lack of consensus rather than claims that consensus exists. Blueboar (talk) 21:11, 23 December 2009 (UTC)

Blueboar, we must be editing the same encyclopedia.  :-/

I'd add that "secret" side discussions on at least two user pages are de rigueur, plus, whenever the dispute involves irreconciliable differences over mutually exclusive options, like "This article must mention ____ in some fashion" and "This article must completely exclude any mention of ____ whatsoever", the addition of an editor who sincerely (but delusionally) believes that it is possible to simultaneously include and not include ____, to lecture everyone else about the importance of cooperating. WhatamIdoing (talk) 22:38, 23 December 2009 (UTC)

(Rhetorical note:) Hear hear! (magnificent!) WhatamIdoing is now in the running for best dif of the holiday season (perhaps all of 2009). Perfect. Proofreader77 (talk) 22:48, 23 December 2009 (UTC)

lol - well, we can argue about the best way to slice this particular holiday pie till the cows come home, but any way you look at it you end up with groups who have painfully different ideas of what constitutes consensus in a discussion. at least if there was some consistent metric of what consensus was, people would stop making bogus claims that they have consensus (or the others don't), and move on to making bogus claims that they meet (or don't meet) the requirements for consensus. might as well get people scamming the system on a more sophisticate level... --Ludwigs2 00:48, 24 December 2009 (UTC)

There doesn't need to be a definition, but there needs to be the possibility of getting a decision from a neutral party as to what the consensus is (and if there is no consensus right now, what the best interim solution is, given that the interim might in practice mean forever). One of the problems is that it's those with extreme (and often illegitimate, from the encyclopedia-making point of view) positions who end up battling it out - those in the middle (or simply the sensible ones) naturally tend to just state their positions and not stick around for the fighting - and it's the latter group rather than the former who are the ones who could produce a meaningful consensus.--Kotniski (talk) 07:56, 24 December 2009 (UTC)

(Note: I have yelled strenuous complaints about Kotniski for something in the past, so if I am responding beneath them, they must be saying something right :-)

Yes, consider the case of high-contention current-affairs-inflamed articles (e.g., Roman Polanski where sensible editors stay away — "consensus" is not defined by the most belligerent POV warriors who attach themselves to the page ... but resides outside the page.

Which means of course, that "consensus" is a "magic word" which we wield but has no actual meaning other than who has sufficient "wiki-juice" to make their call of "consensus" stick. Proofreader77 (talk) 08:27, 25 December 2009 (UTC)

(outdent) Ah, well... the scientist in me says there will never be a solution to this problem without a definition of consensus. If you see it as a measurement problem (how to know when a decently objective approximation of consensus has been reached), you need a definition against which to measure any given claim. if you see it as a political problem (how to construct debates to ensure some relatively fair form of consensus in the outcome) you still need a definition of consensus to know how to structure the debate. relying on neutral parties either shifts the consensus problem a step back (how could we reach consensus over who is neutral in a given case?) or opens the system to massive amounts of gaming (as non-neutral parties line up to assert themselves as neutral). If I wanted to be idealistic, I'd say that a sufficient number of neutral parties involved in a debate would actually start to work, but then you're getting close to the proper definition of consensus anyway, so why not just go ahead and define it?

fact of the matter is, people who are not at all interested in consensus (on both sides) use the term frequently, because the ambiguity of the term (and its status as policy) makes it possible to attack other points of view more or less at will with an appearance of authority. people who are interested in consensus are usually prevented from using the term in its proper context, because people who are not interested in it will attack any definition of the word that doesn't serve their purposes. that behavior will continue until a definition of consensus is given in the policy itself. --Ludwigs2 23:52, 24 December 2009 (UTC)

• [<--Outdent]
  Ling.Nut, you have spoiled a great outstanding WP joke by point this out. Indeed, like the Middle East peace-process, this page only describes how the process of arriving at "concensus" ostensibly works. And if it always did work, that would be enough-- the definition would be an operational one. [It would be like defininig "scientific method" as whatever scientists do. Is this scientific? Sure, look at the guys with natural science Ph.D.'s doing it-- it's scientific by definition. :) ]

  Of course, concensus on WP has a dirty secret: it doesn't always work according the process described by the article on it, even though the article doesn't tell you that. Sometimes (in fact often) the PROCESS just grinds to a halt, without any result, like a locked-up computer. What happens then, is somewhat like what happens in the rest of the world when politics fails: there's wiki-violence, the wiki-riot police and national guard, the wiki-tribunals and ArbCom, and finally some people are booted off the discussion or blocked, and the situation is resolved by means of wiki-juice: which group of editors have the most wiki-power; that is, most administrators willing to make the needed blocks, etc. This is outside the purvue of your present article on WP:CONSENSUS. Sorry. Consensus in large groups works no where in the world, without ultimate resort to force. Why should WP be different? Well it isn't. Although it may be one of the few places left where it pretends that it is. For example, this article pretends that concensus always works, and you just have to watch it work, to see what it is. Except when it doesn't, of course. SBHarris

  01:01, 25 December 2009 (UTC)

@Ling.nut, re: "Sorry. Consensus in large groups works no where in the world, without ultimate resort to force." actually, that's not true by a long shot. Consensus is the norm in most cases (in fact, the terms are philosophically complementary), and only fails where one or more individuals/groups decide that their own goals and desires trump normal civil interaction. unfortunately, internet communication enhances self-serving behavior and diminishes normal civil constraints, so you end up with a lot more unpleasant behavior on wikipedia than you would in the real world, but even so it still is abnormative behavior. --Ludwigs2 01:28, 25 December 2009 (UTC)

Well, well, if it's the norm, you should be able to give me three good serious social issues that were settled by large groups (more than 1000, say) without resort to force (see court order which threatens police violence against the losing group if they do not comply), or outright violence (riots and wars). My examples are the creation of the United States, the end of Slavery in the United States, and the withdrawal of the United States from Vietnam. On a smaller level, my own city could not decide how much to pay garbage men (sorry, sanitation workers). They were ordered to go back to work, or go to jail. Consensus I not see. Okay, your turn. SBHarris 06:20, 25 December 2009 (UTC)

Just three? Ok, off the top of my head, let's go with with basic property rights, simple traffic conventions, and stable community structures. Somehow we all manage to get through each day (for the most part) without stealing or being stolen from, without half the population driving on the wrong side of the road, without any confusion about who is related to whom or what each person is and isn't entitled to do. Any good sized city might have hundreds of thousands (or millions) of people, all of whom share a broad and ever-present consensus about what is and isn't correct behavior for many, many situations (including broad tolerance for certain areas of non-normative behavior and broad agreement about appropriate collective punishments for those who violate the norms excessively). And yeah, I do get your point: you've noticed this-or-that 'important' moment in history that did require force to change a given state of affairs (recorded history is the history of violent upheaval, mostly). However, those obvious moments blind you (and most people) to the fact that people get on and get along in countless, tiny, insignificant ways every day. Even on wikipedia that's the truth: the vast majority of articles have no editing drama to speak of, because the editors there have a consensual understanding of basic principles of civility which allows rapid consensus on editing tasks. in fact, I'd go so far argue that the generic idea behind many policies and guidelines is to push people away from all of the petty little 'moments' they can get hung up on, back to the moment-by-moment task of getting the encyclopedia done (

wp:NOT

is a great example, if you read it as a list of things that people can all too easily obsess over).

Honestly, consensus discussions aren't about how to achieve consensus, exactly, but rather about how to get back to the normal state of consensuality that exists pretty much everywhere, but has broken down in one place or another because of personal or interpersonal issues.

so, note: the fact that your city can't decide how much to pay sanitation workers is an interpersonal issue (people on all sides being a bit greedy, most likely), but clearly there is a broad consensus that someone ought to pick up the garbage, and that those people who do ought to get paid for doing it (neither of which is foregone - e.g. cities could just as easily have decided that citizens need to take care of their own garbage, or could have decided to press minor criminals or traffic offenders into garbage duty as a condition of punishment). force is only needed, or used, where unspoken consensus falls apart and spoken consensus is thwarted. it's a riff on Clauswitz (or maybe Arendt would be better) - violence as nothing more than an effort to reestablish a broken peace. --Ludwigs2 07:51, 25 December 2009 (UTC)

I asked you for three examples where force or the threat of force was not needed. And you wrote back as though you'd never heard of cops, highway patrol, and law enforcement in general. What part of force is it that you don't understand? Remove all police and how long would your nice concensus society last, before some group of thugs simply came through your front door (please no hairy chested explication of how much personal firepower you have-- some group can always besiege and outwait you).

"Consensus" in the US is when 39 senators vote against a health care bill, and 60 vote for it. And after that, everybody gets what only the majority deserve, demonstrate and whine as they will. Fighting city hall is not an option, since the government always outguns you. You WILL comply. Calling this "consensus" is actually ridiculous. It's rather the simple tyranny of the majority, and yet bad as it is, there's no system better for large groups. At least nobody's been able to think of one. SBHarris 20:34, 25 December 2009 (UTC)

really... how many times has law enforcement been to your house today? how about this week? this month? I think I might have spoken to a police officer once in the last year, and that wasn't in an official capacity - last time I had a law enforcement encounter was maybe 5 years ago when I got pulled over for speeding (which happened because I wasn't paying attention to what I was doing, not because I habitually speed). I can honestly say that no cop has ever had to use force of any sort in my presence. And while you didn't actually mention threat of force, even if you had I can honestly say that I've never had an authoritative threat of force used on me (which is a good thing, because I get uppity when people threaten me, and I'd have gotten myself in big legal trouble). are you really trying to argue that the vast majority of people in the world behave civilly (day in and day out) because they are scared of being punished? Not even Hobbes was that cynical - sounds more like Freud on a bad hair day.

I gave you three examples where force (or its threat) are not needed; not in the vast, vast majority of cases. did you want something else, or were you just hoping I'd take it as a rhetorical question?

as for the rest, If you'd like to start up a political discussion we can do that, but I'm not sure I'd disagree with you. I don't think the US Congress or the US electoral system is a particularly good example of a consensus system (or even of a democratic system, for that matter). But since I'd never advocate that Wikipedia model itself after US politics, I suspect the point is irrelevant. --Ludwigs2 00:45, 26 December 2009 (UTC)

The threat of force IS a type of force (assult). Can you imagine anybody getting any slack on a rape case by claiming they didn't use actual force, but just the threat of it? A pistol pointed at your head with a threat, is "force," even if it doesn't go off. Even if (unknown to you) it's not loaded.

As for how many cops have been to my house lately, none. But if it weren't for the threat of force I'd drive as fast as safe road conditions permitted, and I'd pay no attention to the patrol officer trying to flag me over. After all, he's just following what the slim majority of legislators has decided this year, which may well be different from what it was last year, and what it will be next year. I once got a ticket for driving 65 mph on a beautiful morning on a highway which had been designed for 70 mph, but the limit had been dropped trying to save energy. Nevertheless I was ticketed for driving unsafely. Today the limit on the same stretch of highway is 65 again. This is why the threat of force is necessary-- the legislature is composed of idiots, and if they didn't back up their decisions with threats of action by armed men, nobody would pay them the slightest mind. SBHarris 01:48, 26 December 2009 (UTC)

is that threat of force the same reason you don't cheat on your spouse? the same reason you don't steal your neighbor's houseplants? is it really why you choose to drive on the correct side of the road? look, if you want to continue claiming that you never do anything out of a simple recognition of civil-consensus, but only because you feel threatened by the punishments of society, I'm not going to argue with you about it. either you're not admitting the obvious (which is a cheap but marginally effective debate strategy), or you actually believe what you're saying (in which case you have my sympathies, because no one else I know lives in the state of perpetual anxiety that you're describing). However, I am going to dismiss your argument, because if what you're saying is heartfelt then you clearly have no interest in consensus (which you would view as an impossibility) and so you have nothing of value to contribute to this discussion. frankly, if you don't believe consensus is possible, then you must believe that any definition we establish will prove useless; so why are you bothering to debate it? might as well just let us institute any definition of consensus we want, with the cynical assurance that it will ultimately fail. --Ludwigs2 03:02, 26 December 2009 (UTC)

• Drive-by comments:

  (1) re Ludwig: "no one else I know lives in the state of perpetual anxiety") Clearly Ludwigs2 has never been dragged to ANI and subjected to an "ad hoc firing squad" (see causa sui's comment above) ^;^

  (2) General logical note (rare for me lol:) regarding current phase of discussion: Category error: Conflating REAL with VIRTUAL "violence" etc.

  (3) (Idealized) "Consensus" definition: What the summation of the wise application of policy would yield if the wise were to converge on this page and implement said wisdom (by all within-policy means necessary)

  (4) (Un-idealized/operational) "Consensus" definition - Magic word with indeterminate meaning, but with the rhetorical implication that it could be enforced with policy, if push comes to shove, and don't dare shove me, or I'll show you what consensus means, buddy! (smiling with holiday spirit, but not joking)

  (5) Remember I said "drive by" lol For clearer thinking see: S.B. Harris's remarks ^;^ Happy holidays. -- Proofreader77 (talk) 03:59, 26 December 2009 (UTC)

Oh, I've suffered enough wikipedia cluster-clucking to last me a lifetime, thankyouverymuch. It ain't always the cream that rises on this project, and I've had more than one person try to pour their crap into my coffee (if you follow my drift) maybe because I'm not (in case you happened not to notice) particularly demure about my opinions.

So once more, just to make the discussion clean, let's define consensus. Consensus on wikipedia starts from the presumption that two or more editors who disagree on a given point can (by discussing the matter through the lens of wikipedia's main purpose of building an encyclopedia) arrive at some common agreement (consensus) about what best to do with the point. This presumption is really two inter-related presumptions: (1) that each participant is reasonable enough to recognize that ideas they cherish may not have encyclopedic value, and (2) that each participant is capable of balancing their own conception of the truth of a point against the limitations of verifiable knowledge. These presumptions are - believe it or not - valid in the vast majority of cases - most editors have little difficulty self-moderating and most articles are non-contentious topics, so in most cases editors manage to produce more-or-less encyclopedic content that most other editors would consent to (or make more-or-less encyclopedic revisions to). Expanding this to an abstract level you get the idealized and idealistic meaning of consensus on wikipedia: a point has consensus when any given editor who meets our presumptions (from the entire population of potential wikipedia editors) would agree that a particular way of handling a disputed point is a properly encyclopedic way of handling the point. In fact, this is the idea behind RfC's and similar arrangements: bring more voices into the discussion (increase the sample of wikipedia editors) to determine a handling of the point that the largest possible group of editors would accept. Again, this isn't intended as a voting system, but as a discussion system. the thought is that a discussion between 30 people is more likely to produce a broadly acceptable way of handling a point than a discussion between 3 people

Now in practical situations this ideal form is hard to apply, though it has its uses. For instance, it's clear from this ideal that any page which has a long-term dispute is suffering a failure of consensus - whichever side is dominant is failing to convince the other side why its preferred version is an acceptably encyclopedic way of handling the contested point. This is always a failure of discussion: either the dominant side is refusing to explain itself or to enter into discussion fairly, or the underdog side is failing to listen to or engage explanations that are offered. This leads to the practical/procedural definition of consensus: consensus is the process of trying to craft an explanation of the encyclopedic value of handling a point in a given way, one which will be accepted by as large a range of editors as possible. In most cases this is the default, where most editors will accept a particular edit without anything more than an edit summary. In contested cases this is the best way to keep things from spinning out of control, and also the best way to produce something that will gain support at RfC's. It's also the approach most consistent with

AGF

, because it assumes that the people on the other side of the debate will respond positively to a well-reasoned explanation (and if they don't, it assumes that the rest of the wikipedia editing population will). it's also damnably hard, because it requires careful thought, reasoned detachment, and attention to the nuances of the argument on the other side (because you can't really craft a good explanation unless you have a good idea what the other side might respond well to).

of course, this is geared to article editing, but it could easily be adapted to other consensus situations as well. but... Merry Christmas! --Ludwigs2 06:33, 26 December 2009 (UTC)

I suppose I shouldn't take the lack of commentary on this last as a sign of consensus, but I am sorely tempted to anyway - lol. --Ludwigs2 23:35, 30 December 2009 (UTC)

If it would amuse you, I could perhaps make a barnstar-like note for you that says something to the effect that "WhatamIdoing hereby disagrees with you, always, on any point, unless she specifically and directly says otherwise". You could then invoke it whenever you were uncertain of agreement.  ;-)

Actually, you and I do agree on some points, but you so rarely agree with the majority of editors that I think that you would be quite safe routinely assuming disagreement in the absence of positive evidence to the contrary, as you have done in the above message. WhatamIdoing (talk) 00:53, 31 December 2009 (UTC)

re: "I suppose I shouldn't take the lack of commentary on this last as a sign of consensus..."

Correct... I oppose a Merry Christmas. Bah, Humbug! A Happy New Years, on the other hand,.... :>) Blueboar (talk) 02:33, 31 December 2009 (UTC)

lol... The odd thing is that I don't seem to disagree with other editors anywhere near as much as they seem to disagree with me. I'm still trying to figure that one out; some kind of relativistic effect, maybe. I often find myself trying to figure out what some given editor is disagreeing with, either because don't believe I've said anything that much different than what they themselves have implied, or because I don't think I'm saying anything that anyone could possibly find objectionable. Perhaps it's an attitude thing: I'm usually sure of myself (often - though not always - with good reason; and yes, you can exchange the word 'full' for 'sure' if that makes you happy), and I'm not particularly politic about things. that may come off a bit heavy-handed in text.

at any rate, no barnstar is needed; I think we've all figured out that you and I are star-crossed. Vive le difference! --Ludwigs2 02:41, 31 December 2009 (UTC)

I favor merry Christmases (and I wish you all a very happy seventh day of Christmas today), but I don't get to be excessively contrary very often, so I decided to take my opportunity.

Ludwigs, I think you may be on to something with your idea about not seeing how their view differs from yours or how anyone could possibly disagree with some things. Some conversations on Wikipedia seem to be one editor saying things like "Copyright violations are bad!" and someone else responding "No, I disagree with you: I believe that copyright violations are bad!" It must be very frustrating for editors that encounter such disputes regularly. Unfortunately, I don't think that identifying this problem leads us to a solution. WhatamIdoing (talk) 18:40, 31 December 2009 (UTC)

no, I don't think so either. The default for people everywhere (and maybe particularly on Wikipedia, given the limitations of the medium) is to hear our own inner dialog first, and what's actually being said later (possibly on the belief that we already know pretty much what the other person is going to say, so therefore don't actually have to listen to them?). The only real-world-practical solution I've found is to be calmly persistent - just keep floating the idea until people start to see the difference between the inner and outer dialogs. there are better approaches, but they require specialized contexts, such as each side agreeing to certain conversational ground rules. this approach doesn't always work (it fails where one side or another is so attached to their inner dialog that they literally can't accept anything else as real), and it's difficult to apply (the 'persistent' side is easy, but keeping a calm tone is a real trial; everyone gets frustrated with this). but you gotta work with what you have. --Ludwigs2 21:24, 31 December 2009 (UTC)

again

well, this debate seems to have gone silent, so I'll ask it straight out. after the above discussion, does anyone have any objection to my editing in a decent definition of consensus, per above? --Ludwigs2 19:33, 16 January 2010 (UTC)

Sounds like a major change, to me. How about proposing something here, and attempting to reach consensus? (No, I'm not be sarcastic; I can't define consensus, but I know it when I see it.) --Pi zero (talk) 21:37, 16 January 2010 (UTC)

I think it's unnecessary, but I don't really care one way or the other whether the thing is done (at all) -- although if it's done, then you can expect me to whinge about the details. You could always make a concrete proposal, if you're interested in prior feedback. Having said that, Consensus as a result of directly editing the page is every bit as much "consensus" as the kind that comes with date-stamped comments filed away in the archives.

The most important (to me) thing that I wanted to say is that this type of situation seems to occur on other pages, too.

corresponding encyclopedia article is an unreferenced stub with five tags applied (and more deserved). WhatamIdoing (talk

) 21:49, 16 January 2010 (UTC)

well, I appreciate constructive 'whinges', and I'm more than happy to go over to the POV page and add a definition there too - no sense justifying the failings of one page by referencing the failings of another. but for a concrete proposal, start with this (just taking it as a revision of the lead for the moment - it would obviously have to be worked in more subtly in the long run):

Consensus is an description of the way in which most content discussions on Wikipedia should be resolved. In its idealized form, consensus would mean that any Wikipedia editor who viewed the article or changes in question would agree that the material is handled in a balanced, encyclopedic manner. In practice, this universal agreement is usually assumed as the result of the natural process of article editing and

neutrality and verifiability

) to achieve greater consensus.

not perfect, and only deals with article content (need something separate for administrative consensus) but it's a start. --Ludwigs2 23:03, 16 January 2010 (UTC)

I gotta tell you, I think Wikipedia:What is consensus? does a pretty good job of offering the information you and SMcCandlish seem to be wanting here. Maybe we could use that page as a starting point for what needs to be added to this page, or to link to it more prominently from within this page.--Father Goose (talk) 00:41, 17 January 2010 (UTC)

Editors are invited to examine the following discussions. Additional comments are welcome.

• Wikipedia talk:Manual of Style#Defining consensus
• Wikipedia talk:Manual of Style#Achieving consensus
• Wikipedia talk:Manual of Style#Recording consensus
• Wikipedia talk:Manual of Style#Lapses in consensus
• Wikipedia talk:Manual of Style#Wikipedia:Consensus and Wikipedia:Manual of Style
  

-- Wavelength (talk) 21:17, 7 January 2010 (UTC)

Yikes... talk about instruction creep. Blueboar (talk) 23:03, 7 January 2010 (UTC)

I am posing this challenge to all viewers of this discussion page. Please provide (if you can!) links to examples of Wikipedia discussions where decisions were reached by consensus. Please be prepared to explain (if you can!) what consensus means in each example, and how we can be certain that it was actually achieved. You can help to organize this section by arranging your examples under new subheadings as follows, substituting your user name for the words in square brackets, and using "=== ===".

• [First editor]'s example 1 of Wikipedia consensus
• [First editor]'s example 2 of Wikipedia consensus
• [Second editor]'s example 1 of Wikipedia consensus
• [Second editor]'s example 2 of Wikipedia consensus

-- Wavelength (talk) 18:22, 8 January 2010 (UTC)

My favorite is still

WP:5P

, and we can be pretty dang sure of the latter. ;-)

Consensus was reached due to everyone cooperating and building on the work of the previous person, so that the concept rapidly became more polished over time.

Another example is the discussion to remove the delete tag on the talk page on Autonomous_robot during febuary 2004, between Psb777 and a then -as yet- anonymous editor. I wonder who that could be? ;-)

Both editors advanced arguments, and carefully considered the position of the other person. In the end one side became convinced that the other position was actually more valid, and thus ended up endorsing it. In the end, both positions had converged to one single position, so we can be certain that consensus was reached. I remember this particular discussion because of the huge amount of fun I had, and how it really drove home to me what NPOV and consensus were, and how they worked.

It's interesting, looking back, that the discussion was a lot calmer and more leisurely, and more focussed on actually writing the article than they tend to be today, where people often open with a

WP:WOTTA barrage ;-) . --Kim Bruning (talk

) 18:41, 8 January 2010 (UTC)

Filibuster/unanimity point is missing

There definitely used to be a very crucial point in this policy, that is now somehow missing, that consensus != unanimity, and that a single, or proportionally very small number of, hold-out editors cannot filibuster the consensus building process.

The point about WikiProjects used to tie into the lone-holdout point, but now seems to have replaced it. It is weaker as a result (it is a big non-sequiturial now), as is

WP:CONSENSUS

as a whole. The point of both the project point and the filibusterer point was that WP-wide consensus is not to be turned on its head by fringe views/demands, and even at the minor article talk page level, consensus is not formed by one or a tag-team of editors browbeating others into submission or refusing to go with the flow and thereby supposedly causing a "no consensus" result. Frankly, I see more and more of this happening (and more and more editors letting it happen), and I can't help but wonder if its because this policy page has been whittled away at too much, losing one of its most important observations.

The not-a-unanimity point was important because consensus often requires compromise, and editors who will brook no compromise have essentially put themselves outside the consensus building process, whether they admit it or not, and whether the editors around them, whose collaborative progress is thereby held up, realize it.

I'm also about 90% certain that there was something in here, very clear, that made the point that consensus is not a vote either. It's creepy to not bother to read a policy page for a year because I have it mostly memorized, and then come back to it and find that it has so little in common with what I remember of it. If this were some random essay, sure, but this is one of the core policies. How can it drop vital points on seeming whim? Isn't anyone watchlisting this any more?

— SMcCandlish Talk⇒ ʕ(^Õل^ō)ˀ Contribs. 17:51, 16 January 2010 (UTC)

Yes, people are watchlisting it, but exactly the same process you're complaining about on pages has been being applied to this policy page. basically (IMO) it's a conflict between some big minority POVs that have been trying to stamp out some little minority POVs by precluding them from consensus discussion (and like most POVs they have little concern or appreciation for collateral damage). I've been tempted to rewrite consensus from scratch as an essay, and then once it's done correctly submit it as a site-wide RfC to replace this page entirely - much easier than trying to argue with people who are more interested in their private little wars than the bigger picture. --Ludwigs2 19:30, 16 January 2010 (UTC)

The problem is -- how does one force people to use consensus? Our enforcement options are very limited. That's not to say that seeking consensus shouldn't be strongly encouraged, and people overtly defying the consensus process can eventually be hauled before arbcom. But in between, quite a bit of coercion takes place on the site.--Father Goose (talk) 22:03, 16 January 2010 (UTC)

I think that the people most likely to hang out here, and therefore to have an effect on the direction of the page, are people who think that consensus (with the lowercase letter) is the be-all and end-all of Wikipedia. If you think, for example, that the primary purpose of consensus is not to determine "Do we all agree to make this change to this article?" but "Does this change make the article better conform to

WP:NPOV

?" then you'd hang out at NPOV, not here.

I, too, think that this page needs to have information about consensus not requiring unanimity restored. It's not just the 'compromise' nature of some changes in articles: no adequate compromises are possible in some disputes. You cannot simultaneously display, and not display, images of Muhammad.

As for 'not a vote', someone recently pointed out that four to one is essentially a dissension-proof margin. If you are opposed by four editors willing to remove your changes just once, then you'll be blocked for

WP:3RR. So while consensus is not defined as majority rule, overwhelming majority positions aren't irrelevant to it, either. WhatamIdoing (talk

) 22:10, 16 January 2010 (UTC)

I apologise for not being present as often as I used to be. Basically a group of people have been rolling back discussion and application of consensus across the wiki, they've already pulled it out of

WP:POLICY

almost entirely, for instance. (Existing consensus is no longer a source of policy, according to many, apparently). I've managed to hang on by my fingernails for ages, but I can't invest as much time as I used to, and I'm off doing other things, now.

More strictly on topic:

• The easiest way to create a filibuster-in-fact is to simply induce a page to be locked (never allow a page to be locked).
• The
  12 angry men
  approach to consensus building is sometimes misidentified as "filibuster". When people then attack this, they sink the consensus process.
• I agree that all parties must be open to compromise and *show* that they are open to compromise, and *check* that they are open to compromise. I'd like to get people to think about 4 questions before they make an edit or take some action:
  1. A personal reason why you are in support of the action you are taking
  2. What would theoretically need to be said or done to make you reverse that position (where are you willing to compromise?)
  3. A reason why you think others (will) support you (ie, why is this likely to gain consensus)
  4. What likely would need to be said or done to reverse or modify the position of the other members of the community. (where are others willing to compromise?)

--Kim Bruning (talk) 15:10, 26 January 2010 (UTC)

I like those ideas, Kim, but unfortunately Wikipedia is sensitive to the 'one bad apple' effect. one person who refuses or is incapable of doing what you say will destroy a consensus discussion entirely, and there are just too many advantages in refusing. --Ludwigs2 16:08, 26 January 2010 (UTC)

Tom Lehrer said... "I'm sure we all agree that we ought to love one another and I know there are people in the world that do not love their fellow human beings, and I hate people like that!"

It is almost impossible to force someone else to work towards a consensus. All you can do is take the high ground and work towards consensus yourself. Blueboar (talk) 21:01, 26 January 2010 (UTC)

Well, the theory behind consensus decision-making relies on the belief that mild-mannered, common-sense viewpoints in bulk will outweigh even the most aggressive partisan viewpoint. that actually works moderately well in real-world situations (those which aim for consensus, at any rate) but unfortunately wikipedia practice has developed to privilege aggressive non-communicative acts. consensus only works on wikipedia on small backwater pages where a handful of editors are disinterestedly working on articles; on any page that's even slightly contentious, consensus discussions are out of the question - anyone who tries consensus editing will (a) have any article changes reverted with a demand that they discuss changes in talk, and then (b) find all discussion in talk to be non-communicative, tendentious opposition (assuming anyone bothers to communicate at all). for example, I just made a relatively minor change to

Alternative Medicine

the other day. It was reverted twice, the pov tags I added after that were also reverted twice, and all of this happened with virtually no discussion in talk, despite my requests. well, no discussion from one editor, who told me to discuss in talk before making changes but never actually contributed to talk; the other editor has responded with a series of mild insults but is only now getting to anything like a meaningful discussion of the changes. we'll see how that turns out...

frankly, I no longer see much value in pursuing consensus. I still try, in the idealistic hope that someone will actually respond to those overtures, but generally I come to pages with the presumption that I will ultimately need to browbeat people in order to make any effective changes to the the article, because in my experience that's the only effective strategy. C'est la vie. --Ludwigs2 21:56, 26 January 2010 (UTC)

p.s. - the non-communicative editor finally did contribute to talk - by refactoring my last talk page comments. it is to laugh.

You need me to drop by and do some User:Georgewilliamherbert/DefendEachOther ? ...or is someone already present and helping with that? --Kim Bruning (talk) 16:51, 27 January 2010 (UTC)

A little while ago, I had an 'experienced editor' revert a change (that had even been discussed on the talk page) and tell me that I was required to follow

WP:BRD (as if it were mandatory) ...and that she refused to allow me to discuss her reversion with her, even though that's exactly what BRD demands. WhatamIdoing (talk

) 02:17, 27 January 2010 (UTC)

Yeah, that's the thing. when a FOAC (I just made that neologism up - Fresh Off Auto-Confirm) editor does stuff like this, it's one thing. but when editors who are by all accounts respectable, productive, long-term contributors do it, it speaks to an entirely different universe of problems. I'm going to stop talking about it now (because I'm obviously suffering form a cynical moment), but I do hope this discussion floats a bit. --Ludwigs2 05:57, 27 January 2010 (UTC)

Well, there's AGF to start with, but you're not locked in that mode (you should still assume as much good faith as possible, but an unwillingness to discuss is prima facie bad faith, because "good faith' *includes* working towards consensus) ); but when people are being uncommunicative, you can start escalating. First you should assume ignorance, of coure and try to explain what they're doing wrong. If they then continue to be unhelpful (rule-of-thumb: 3 posts; that's why there's 3 templates for warnings too), there's no need to remain *perfectly* nice indefinitely.

For instance, if someone isn't following BRD (whilst claiming they are), explain how brd works, if that doesn't work... be bold... if they revert you with silly demands, that's when you can start asking other editors to drop by and explain to them to behave, and revert them back. When that stops working, it's time to start with (minor) sanctions (which incidentally should NOT start with locking the page, use a page ban for the troublemaker). After that, you can ramp all the way up.

Unfortunately, you need to ramp up more often than you used to, and it costs more time than it used to, but you can still do it.

I *am* starting to get worried about the large numbers of admins that are utterly clueless though. The RFA requirements have not "gone up", they have been reduced to straight seniority requirements, with only a modicum of clue required.

So in the process of ramping up, if you encounter a clueless admin... *then* you might be toast... :-/ --Kim Bruning (talk) 16:48, 27 January 2010 (UTC)

well, unfortunately, I'm not an admin, so I am limited in my ability to warn other users, and incapable of sanctioning anyone. The problem with this approach from the perspective of a normal user is that it basically leaves us walking a thin line between various blockable offenses. I've taken approaches like this before (and still do, to the extent that I can within my permissions) and I generally get accused of edit warring,

canvassing, tendentious editing, violations of consensus and other fun and completely unjustified things. If I'm lucky I can bowl another editor over on the basis of sourcing and logic or in an RfC, if I'm less lucky I can make enough of a stink to attract lots of uninvolved eyes to the page (which puts everyone else on their best behavior), but often the situation devolves to a stalemate, where loosely associated teams of editors blindly revert each other and steadfastly refuse to talk about content on the talk page. I suppose I could run for adminship and try to get the tools I need to enforce consensus editing (though I imagine that as long as I keep making statements like that my prospects for adminship would be fairly dim ), but... you see my point. --Ludwigs2

18:12, 27 January 2010 (UTC)

I handed in my admin flag ages ago, to see if I could still enforce consensus without it. I never asked for the flag back, so I guess it worked ok. :-) --Kim Bruning (talk) 19:28, 27 January 2010 (UTC)

lol - ok, I guess I'll have to study your techniques. --Ludwigs2 19:43, 27 January 2010 (UTC)

I don't think that it's truly "consensus" unless it is consensual. When you have to "enforce" a consensus, it quits being a consensus.

Back on the point, we do need to get the anti-filibuster language back into this page. A single tendentious editor must not be allowed to suck up hundreds of hours of other people's time in an effort to get their favorite pseudoscience or conspiracy theory presented favorably. It is not possible for editors to form a consensus on Wikipedia to flagrantly violate

WP:BLP in an article, no matter how long they talk about it. WhatamIdoing (talk

) 20:03, 27 January 2010 (UTC)

you can't enforce consensus, but you can enforce consensus procedure. Debates in parliament, or in the US house or senate, have strict and strictly policed rules about who can say what and when they can say it, because if they didn't have and enforce such rules government just simply wouldn't work (not even to the minimal extent that it works now)

with respect to your filibuster point, however, you're making it sound like yet another means of quieting a dissenting opinion without actually dealing with it. I've never met a fringe advocate I couldn't handle with a short discussion about the practical limitations of an encyclopedia, and I've never met a fringe advocate who would ever give up through being ignored or repressed. these people "suck up hundreds of hours of other people's time" because they get frustrated with what must seem to them to be irrational obtuseness; there is no surer way to get someone on their high horse than to indicate that you have no interest in listening to their side of the story. --Ludwigs2 20:56, 27 January 2010 (UTC)

Perhaps you haven't been dealing with the same kinds of editors, then.

exactly this kind of behavior

and with whom no number of "short discussions" had any apparent effect on their grasp of basic facts, like "Wikipedia is not the place to promote your beliefs about your health".

At some point, it is in the best interests of the encyclopedia to quit "dealing with it" and ban editors whose goals are completely incompatible with the project's goals. WhatamIdoing (talk) 23:18, 27 January 2010 (UTC)

So I guess that means Ludwigs and you agree with me? Start with AGF, stick to a logical procedurelike glue, and calmly escalate. Should the other party fail to amend their behaviour, one inevitably escalates to the level blocks and bans. --Kim Bruning (talk) 00:05, 28 January 2010 (UTC) In my case, I'm often somehwhat more forgiving than others, so I usually start out defending a person's right to speak, and tend to "escalate" in the form of slowly withdrawing my support for the (weaker) party.

@Kim: actually, yes, I agree with you - I just that for me, the more it needs to escalate, the more I see it as a failure of the process. It reminds me of when I was a kid, watching my parents make my little brother finish his green beans. such an unpleasant test of wills, and one that (if it goes past a certain level) becomes more a game of chicken than a functional interaction.

@WhatamIdoing: well, I don't doubt there are people out there who are completely immune to reason. I think I have an easier time with people who are trying to push a POV because they have an incentive to communicate - If you give them the opportunity to explain themselves they will, and then you can whittle off the stuff that's just unencyclopedic and work to include whatever little bit is usable in what they want. they feel heard, their perspective is either incorporated a little bit (or at least they have a better understanding of why it can't be), and the whole process tends to soothe tempers. I have a much harder time with editors who establish a POV on a page and then act like clams; hard to make any progress when no one is willing to talk at all. --Ludwigs2 01:58, 28 January 2010 (UTC)

I would say that it's not a failure of the process, but a failure by the person in question to adhere to the process. You yourself pointed out that some people are not here to (paraphrasing) reach consensus and build an NPOV encyclopedia. We assume good faith at first, but once we have established that the assumption is unwarrented, those people need to leave.

If a person really is here to build an encyclopedia and build consensus, we usually find out, and we never get to the point where we need to get all too mean to them. --Kim Bruning (talk) 03:55, 28 January 2010 (UTC)

"failure of people to adhere to the process"? That's a bit like the old 'the operation was a success but the patient died' line. a process is (loosely) a systematic system for achieving a functional end. in an assembly line one cannot always blame a missing screw for the failure of the product; sometimes one has to blame the process for failing to put the screw where it needs to be. on wikipedia one cannot always blame a troll for being a troll; more often than not, I suspect, one has to blame the system for making a structure where people feel like they have no recourse except to be trolls.

too few people on wikipedia have an understanding or regard for consensus, and there are no effective models of consensus here (that I've seen, anyway - every noteworthy attempt at establishing it I've run across broke down into spiteful partisan haggling). without some mechanism for maintaining the process of consensus, that will continue to be the norm. I like the idea of anti-filibuster wording, but only to the extent that it reigns in disruptions to the process of achieving consensus, not as a tool to short circuit discussions without any actual discussion. can we create wording like that? --Ludwigs2 17:15, 28 January 2010 (UTC)

By now you're probably right. Consensus requires people to learn about how to use it. For years I've been saying "we have acculturation issues, we have acculturation issues", now it's becoming "The community is no longer working the way it used to"; the one is the logical consequence of the other. If I could figure a way to round up Ye Olde Wikipedia Cabal, and set them to teaching, now that would be grand. ;-) --Kim Bruning (talk) 17:47, 28 January 2010 (UTC)

lol - I'm torn between visualizing you as Paul Revere or Chicken Little. possibly a mix of both, or Paul Revere riding Chicken Little? maybe we should all get together and write a poignant essay on everyone's responsibility to teach and promote positive wikipedian values (which I don't mean sarcastically, but which I think everyone will assume is sarcastic). --Ludwigs2 19:45, 28 January 2010 (UTC)

The trouble with any supposedly egalitarian, democratic, or meritocratic system is that certain people or groups inevitably manage to amass more power -- through a progressive loss on checks on power, by massing forces, or by getting really good at manipulating the system. The filibusterers are a major instance of the third.

Wikipedia has done an amazing job at holding back each of these forces, but it's still inevitable that bit by bit the site falls more and more prey to "ownership" of one type or another. There is already a large chasm between site regulars and newbies. I'm not sure I'd have joined if the site then was as it is now. I got in just when the culture was becoming far less open, and it's closing more and more by the day.--Father Goose (talk) 04:20, 29 January 2010 (UTC)

I found this article (on changes to the rules regarding the Filibuster in the US Senate) pretty interesting, and wondered if some of the same ideas might be applied here. In particular, this suggestion sounded promising:

Senator Harkin suggests that the cloture requirement (currently at 60 votes) could be lowered gradually, the longer a measure under consideration is debated. Thus, for example, a cloture sought early in deliberations over a bill or other measure might require 60 votes to pass, whereas the requirement for cloture on a measure after the Senate is already days into the debate, might be re-set at 57 votes. Days later still, the requirement could be lowered to 54, and so forth.

Now, I realize WP doesn't conduct votes per se, but there is an informal measure of support taken into account when trying to identify the consensus view. As an argument drags on, perhaps it should become more acceptable (in some sense of that word) to go with the majority view. That would give members of a vocal minority time to make their case, but if they can't convince other members of the community in a reasonable time, then at some point they shouldn't be allowed to hold up the process any longer.

Has this idea been discussed here before? What are other people's thoughts on it?

--Lewis (talk) 19:56, 13 February 2010 (UTC)

No, what you're proposing is still just a form of voting. Long-term disagreement is frustrating, but it's still in every respect better than coerced outcomes (and "majority rule" is a form of coercion).--Father Goose (talk) 00:26, 14 February 2010 (UTC)

The issue I'm trying to address is when a small minority of participants (or even a lone dissenter) effectively "filibuster" a proposal while refusing to listen to reason. Clearly it's good to have healthy debate, and even a lone dissenter can prove to have the stronger position, in the end. But sometimes that's not the case, and a small group can effectively block change by refusing to budge. I'm wondering if some of the ideas discussed in the article I mentioned might be used to formalize some guidelines about how to deal with this type of disagreement. "Consensus" would start out with its strictest meaning of the things that everybody agrees to, but if the discussion becomes "stuck" and progress stops, then after a time (perhaps a long time, I don't know) the threshold could be lowered progressively. I suspect that this is what many administrators of the site do anyway, but in an informal manner. I'm suggesting we make it a bit more formal, or at least discuss it explicitly as a principle to be considered in determining how to proceed when "true" consensus can't be reached, yet some kind of decision is still needed. --Lewis (talk) 00:59, 14 February 2010 (UTC)

Most of the discussions on this page relate to that problem. Voluntary consensus is the ideal. Sometimes the ideal cannot be implemented, for reasons that range from incompetence to invincible ignorance to willful disregard of community standards to mental illness.

A small number of editors at this page have, over a very long period of time, removed even such common-sense statements as "consensus is not unanimity" (which is also a form of "voting") and opposed statements that indicate that, e.g., the community's consensus that everyone must follow BLP outranks any claim of a consensus to violate it.

Wikipedia's official process for dealing with this situation can be summarized as "inflict as much pain and stress on the good editors as possible, while endlessly protecting the dissenter's rights, until a sufficient number of people are so irritated that the dissenter is entirely ejected from the community". In my experience, it usually takes between four and six months of dull and difficult work (work that, a anyone that has ever filed a user RfC can attest, the community will almost completely ignore) to achieve this end -- and even then, a few contrite words are likely to result in a long series of second chances.

If you want a practical answer, however: Because we draw the line at

WP:3RR, you need a four-to-one ratio of editors to enforce any proposed change. Any four editors can, in practice, impose any change they wish if they're willing to keep reverting (once a day) and leaving perfunctory messages about the evils of edit warring against consensus on the talk page. WhatamIdoing (talk

) 03:29, 14 February 2010 (UTC)

That is a very bad application of 3RR. Just because they are not steping above 3 reverts each day does not mean that they are not edit warring. Tag-teaming is frequently seen as disruptive and may get you blocked.

Taemyr (talk

) 07:51, 15 February 2010 (UTC)

Depends on whether additional users agree with the four. If a group is trying to force something that would not hold up to broader scrutiny, then, yes, it's frowned upon. If their actions are seen by others but nobody speaks out against them, that's a form of silent consent (notwithstanding the lone dissenter).--Father Goose (talk) 09:43, 15 February 2010 (UTC)

Agreed, but the critical issue is, or should be, approval of the broader community. Not having the nubers to "win" an edit war.

Taemyr (talk

) 12:15, 15 February 2010 (UTC)

I agree that it's undesirable, uncollegial, and disruptive behavior; it's also functional behavior, in the sense that any four non-admin editors really can force a page to reflect their views. Single actions by the "tag team" will not result in a block for any of them during the first day of edit warring, but a block for the other editor during that day is highly likely. WhatamIdoing (talk) 23:49, 15 February 2010 (UTC)

A lone dissenter rarely gets what he or she wants, so that's not much of an issue. When it's a close split, like 3 to 2 or something, that's "no consensus", and the parties will have to work out a compromise between themselves, or solicit more input via an RfC or whatever. And if the split remains inconclusive, the course of action remains unresolved as well. That's frustrating, but it's also as it should be. Consensus more or less means reasoned agreement on Wikipedia, and if the agreement isn't there, there's no consensus. If the agreement disregards an objection that is reasoned enough to attract a substantial minority of dissenters, that's also not a reasoned agreement; it's

steamrollering

.

In the majority of cases, though, I'd say a compromise position is possible, as long as one or both parties are not completely Balkanized (in which case they'll eventually wind up before ArbCom).--Father Goose (talk) 05:25, 14 February 2010 (UTC)

Certainly that is usually true if all parties are acting in good faith. Of course there are a few occasions where, no matter how much good faith we have, there is no compromise position to settle on. Where there is a clear black or white choice between position X and position Y. Blueboar (talk) 15:36, 15 February 2010 (UTC)

A dispute involving demands for "A" and "not-A" simply cannot be resolved through consensus. There is no possible method of simultaneously displaying, and not displaying, images of Muhammed.

It is possible for these editors to develop a consensus that displaying the images more closely aligns with Wikipedia's stated policies than not displaying the images would, but that's not the same thing as getting the "absolutely no images" person to voluntarily agree that displaying the images is the best choice, or even an acceptable one. WhatamIdoing (talk) 23:49, 15 February 2010 (UTC)

WhatamIdoing, I think that's a great example, and pretty much the point I was trying to address. I think in those situations we clearly need to go with the majority, and the "gradually lowering threshold" method might serve as a useful mechanism for identifying those situations, or at least providing a consistent rationale for dropping the requirement for strict consensus in certain limited cases, while not requiring any sort of "judgment calls" based on the content of the dispute. --Lewis (talk) 03:53, 16 February 2010 (UTC)

Even in that case, there are other options -- linking to the images but not displaying them inline, for one. That's not the route we chose, but in a great many situations, something other than a strict binary choice is possible. Similarly, in an AfD, improvement, removal of offending content, merging, etc. are often options. That's not to say that there are no yes-or-no choices (RfAs are a good example), but, really, there are intermediate positions for far more issues than one might realize.--Father Goose (talk) 08:26, 16 February 2010 (UTC)

Which is not to say that the intermediate position is a good one. There are already too many fudges on Wikipedia - we reach a well-argued and well-supported position on something, but a piqued minority insist on having "their viewpoint" taken into account and we end up with an illogical compromise (for example, we decide not to add any new image placeholders, but forbid people from removing those that already exist).--Kotniski (talk) 08:40, 16 February 2010 (UTC)

I think that's a very reasonable compromise. If the community really is split over whether we should have placeholders, with neither side having the One True Approach, then a compromise like that that at least gets them to stop warring over what is ultimately not that important. That's better than forcing one side's approach on the other, and if a given person, or even an entire side happens to be too inflexible to accept anything but "total victory", that person or persons should reconsider whether being part of a mass-scale collaborative project is really right for them.--Father Goose (talk) 05:11, 17 February 2010 (UTC)

If I understand the concept correctly... I disagree with idea of a "gradually lowering threshold". I would rather go with a "gradually expanding consenus"... ie post RfCs and other notifications to get more people involved in the conversation. The best way to resolve a dispute that centers on consensus is to demonstrate that one side actually has a broad consensus beyond the regular editors on the page. Blueboar (talk) 04:28, 16 February 2010 (UTC)

Blueboar, if this page still said, as it once did, that consensus did not require unanimity, then I might agree with you. As it stands, if one person holds out for 'no images of the prophet whatsoever,' then we do not appear to have a consensus according to this policy, no matter how many thousands of editors are involved. WhatamIdoing (talk) 04:34, 16 February 2010 (UTC)

I will confess upfront that I often get frustrated with the misunderstandings of the concept of consensus I run into on wikipedia. Consensus does not have anything to do with numbers, or majorities, or voting, or filibustering, or anything of that nature. Consensus means that we come to a collective agreement (maybe mutual understanding is better) about the topic under debate. even with binary (yes/no) decisions consensus is perfectly possible - it doesn't mean that everyone will come to believe the same thing, of course, it just means that everyone will come to understand why one choice is the better choice for wikipedia's purposes. that can always be achieved with careful explanation and discussion.

The problem here is not that a true consensus system is impossible, but that wikipedia has institutionalized itself so that there's no value in trying to get consensus. Most everyone who edits wikipedia realizes that the system is designed to privilege uncommunicative, intransigent, whiney editors - the secret to success on wikipedia is to get your preferred edits in, block any changes, refuse to discuss anything, and be the first to scream to a friendly admin when it gets close to 3rr time. sucks for anyone who doesn't know any friendly admins, of course... The reason it is that way is because that's the way a lot of editors want it; they have themselves set up to take advantage of that system, and any real consensus efforts are threatening (they'd actually have to express themselves and communicate their ideas, and they have no practice at doing that; it would cost them power). And here you seem to be proposing that the way to fix the supposed consensus system on wikipedia is to make it even less consensual.

You people, I swear... --Ludwigs2 05:39, 16 February 2010 (UTC)

So bitter! I've had greatest success at getting what I want on Wikipedia when I have been open-minded and communicative. Doing that requires skill, however, and a specific personality. Are there abuses on the site? Plenty. But there may be fewer here than other, more regimented, systems I have seen (such as your average government, "democratic" or otherwise). There are still opportunities to reform our methods and to strengthen our use of the "consensus" methodology. But maintaining our dedication to consensus will require "an eternal vigilance".--Father Goose (talk) 09:06, 16 February 2010 (UTC)

I don't know what others are proposing (or have proposed) but all I'm suggesting is that we try to formalize what's probably already done in an informal manner in certain situations. Administrators sometimes seem to make decisions based on their assessment of what the consensus viewpoint is, and like it or not, that's often based on the number of people on each "side" of the debate. Even worse, it seems to be (in my admittedly limited observations) based on the administrator's own personal biases, as often as not. I'm suggesting that we use something like this "gradually lowering threshold" guideline to automatically and objectively formalize a process for stepping back from unanimity to majority decision-making. Or not.. I'm not wedded to the idea, I just found the article interesting and wanted to discuss it. I can see some flaws already, since it would seem to encourage gaming of the system by people in the majority, since they could just "wait out" the minority and eventually get their way without having to convince anybody. Then again, isn't that what people do already? --Lewis (talk) 05:58, 16 February 2010 (UTC)

Yeah... see, I would tend to go in entirely the other direction with this. I don't think we should be looking at unanimity or majority, and I don't think that administrators should ever step in to make de facto authoritative decisions about content (except in the obvious BLP/Vandalism type cases). In my conception of a perfect wikipedia, administrators would step in mainly to quash disruptive, off-topic behavior and maintain an environment where people could actually discuss the material. I can't tell you how many times I've seen decent, delicate, but productive negotiations on some point destroyed by some numbskull who wants to shout about whatever idiotic thing is stuck in his craw. --Ludwigs2 06:37, 16 February 2010 (UTC)

Having admins act as arbiters in content disputes is a disaster. We must do everything in our power to avoid embracing turning the site into laborers and overseers. The more the community is subject to admin caprice, the more it will fracture, and giving admins power like this inevitably means it will be abused. At AfD, the most prominent example of where admins have such power, the rulings are becoming less and less objective over time.

Any approach that disregards or even marginalizes reasoned objections will politicize the site and discard the dialectic ideal. That we sometimes fail to live up to that ideal does not mean in desperation we should adopt an approach that drags us even further away from it.--Father Goose (talk) 08:39, 16 February 2010 (UTC)

That's true in a sense, but it also leaves the main problem unsolved - if we don't trust admins to arbitrate matters of disagreement, then who do we trust? Because someone has to (that's clear with AfDs; it ought to be clear with other matters as well, unless we just want content disputes to ramble on for ever and ever). We need to recognize that in some cases, arbitrating the result of a discussion is not just a simple matter of one person turning up and making an off-the-cuff decision.--Kotniski (talk) 08:46, 16 February 2010 (UTC)

Arbitrators are tasked with addressing editor behavior only (and they are usually wise enough to limit themselves to that). Their role, and the role of administrators with respect to consensus should be to make sure that everyone is playing fair -- actually striving for consensus, not coercion (through majority rule or the dreaded "filibuster"). If we ever turn admins or even arbitrators into people who get to declare how things will be done (regardless of whether they're claiming to "interpret consensus" or "interpret policy"), suddenly we have a power structure, and consensus suffers for it.

We do need better enforcement of our consensus methods -- more willingness to sanction people who don't play fair. It takes a very even-handed person to do so justly, though, which is why we rely primarily on the Arbcom to do it. Unfortunately, there's far too much bureaucratic baggage that goes along with an Arbcom case, which means we only resort to it in the most dire of situations. Maybe some day we'll strike upon a less high-load means of ensuring consensus methods are held to while still maintaining fairness.--Father Goose (talk) 04:11, 17 February 2010 (UTC)

I don't think it's that hard - (1) people try to agree as to what is the right thing to do; if they can't (2) they try to agree about what is the consensus thing to do; if they can't (3) the matter is settled by a trusted neutral like an admin; if that person's decision is contested (4) the matter is referred to the community of admins and they try to agree about what is the consensus thing to do; if they can't agree (5) it goes to ArbCom. Note that from (2) onwards the question is no longer "what is the right thing to do" but "what is the consensus thing to do", based on the discussion held at stage (1). (So admins and arbs aren't making the content decisions; they're observing what decision the community has taken.) And the need for the particular stages falls off, so that only a tiny minority of cases ever require stage (5). And at no stage is it necessary for anyone to be unpleasant to anyone else.--Kotniski (talk) 09:15, 17 February 2010 (UTC)

Ludwigs, your idea that "everyone will come to understand why one choice is the better choice for wikipedia's purposes" is lovely, but I tried to get that concept included in this policy, e.g., last winter, repeatedly, and it was rejected. Apparently, the goal is not to get everyone to agree that a given choice serves the project, but to get everyone to agree, full stop. WhatamIdoing (talk) 09:00, 16 February 2010 (UTC)

Even assuming there's any difference, both goals are unattainable.--Kotniski (talk) 09:05, 16 February 2010 (UTC)

@ WhatamIdoing: as I said, there's an unfortunate degree of misunderstanding about the nature of consensus on wikipedia (well, that's the polite interpretation, anyway; I've been known to use stronger words...). really, the problem with real consensus is that it takes thought, consideration for others, and patience, and most wikipedians think in terms of shortcuts. If you want to try again, though, I'm happy to pitch in and help with the project - we just need to work out a perspective we can both agree to

@ Kotniski: nothing's ever perfect, but things are only as difficult as you make them. --Ludwigs2 09:33, 16 February 2010 (UTC)

Don't understand "things are only as difficult as you make them". If it's important, can you explain?--Kotniski (talk) 11:49, 16 February 2010 (UTC)

sorry, that was really just a philosophical rejoinder. I'm reading your comment to mean "here's ideal X, which is just way too high for us to ever reach, so there's no point in trying". well obviously, if you insist on not trying then getting up out of bed in the morning is an unattainable ideal. Consensus (like 'love your neighbor' and 'be kind to animals') is not an ideal that will ever be fully realized, but that doesn't mean we shouldn't strive for it. trying and missing, we'll still have a better wikipedia. --Ludwigs2 16:07, 16 February 2010 (UTC)

Well, I agree it's worth trying to get unanimity on the issue to start with; and if that proves impossible, then trying to get unanimity as to what decision has been taken. Worthwhile aims, indeed, but they shouldn't be a necessary condition for doing anything. Sometimes we have to do things "by consensus" even if some people are disputing (obstructively, or even sometimes legitimately) what the consensus is.--Kotniski (talk) 18:09, 16 February 2010 (UTC)

Interesting... it seems we have no consensus on what a consensus is or how to reach one. But we all agree that consensus is how Wikipedia works. This explains a lot. Blueboar (talk) 15:01, 17 February 2010 (UTC)

Well, at least we accept "consensus" as a working name for the process by which Wikipedia works (or perhaps by which it should work). If someone can come up with a better name (more understandable and less misleading to newcomers), I'd happily stop calling it "consensus".--Kotniski (talk) 15:21, 17 February 2010 (UTC)

Actually, blueboar, I don't think that's true. I'm pretty sure we all know what consensus is - I mean, I think if we asked people to give us their considered opinion of what an ideal consensus was they'd all say something like "Everyone agrees that [whatever] is the right thing in that case". the problems come when we try to cut corners. e.g. we have a fear that editor(s) X will never agree to anything we consider to be 'right', so rather than trying to convince them, and rather than considering that we might not be quite as 'right' as we think we are, we simply look for various means to nullify their input - not realizing that when we do that we gut consensus completely. honestly, anytime I hear someone say 'That's against consensus' I have to laugh. that phrase is only ever used by people who (a) don't understand what consensus is, and (b) are trying to retain their preferred version of something by referring to some abstract authority. it's nonsensical. --Ludwigs2 16:40, 17 February 2010 (UTC)

I think this section would benefit from some "ought to be unnecessary, BUT..." advice about using the relevant talk page. That is, if you want to discuss changes to X, you should use Talk:X, but an agreement at Talk:X can't be imposed on page Y over the objections of editors at Talk:Y. It's sort of an extension to the "your off-wiki 'consensus' doesn't count" rule: your off-page 'consensus' doesn't count at this page.

I seem to find myself regularly needing to remind editors that "Discussions at THIS page do not change anything at THAT page," particularly for pages in the Wikipedia namespace. But I can't think of a suitable (non-aggressive) way to say it. Does anyone have any suggestions? WhatamIdoing (talk) 05:38, 19 February 2010 (UTC)

hmmm... actually, that an interesting question. part of me wants to say that sometimes overarching decisions are useful (for consistency, when you have issues of two pages that are very similar), but I do see your point. I'd suggest something like: "Consensus does not (necessarily) transfer between articles: a decision reached on one page does not carry any weight on any other page, except to the extent that the same arguments might be reapplied and rediscussed in the new context." would that work, for a start? --Ludwigs2 05:54, 19 February 2010 (UTC)

It should be in a different section, maybe under Improper consensus-building. —

talk

) 06:01, 19 February 2010 (UTC)

PS: It could also be part of a new section about where discussions should take place. If a conversation veers toward a general subject, it should be moved to a policy or guideline page. If a subject of conversation on an article's talk page affects a group of articles, it should be moved to the page of the smallest category that contains all of the affected articles. If a different article is mentioned, any discussion of it should be moved to that article's talk page. There should also be a guideline about how heavily decisions on general pages are allowed to control more specific pages in the same general area. —

talk

) 06:22, 19 February 2010 (UTC)

It doesn't really matter where a decision is reached; the important thing is that users watching any page which is going to be affected are made aware (and if it's going to affect articles generally, then it should be announced at the central locations). All within reason, obviously - for example, it's been concluded in the past that we don't need to announce a proposed categorization change on every page belonging to the affected category(ies).--Kotniski (talk) 07:19, 19 February 2010 (UTC)

I agree with Kotniski... The key is notification and communication... when you have an issue that relates to multiple pages, it makes sense to have a centralized discussion (this is one reason why we invented wikiprojects). As long as editors at the related pages are informed that the discussion is taking place, and where it is (thus having a chance to be involved in the consensus building process) then I think a consensus at one page can transfer to another. Blueboar (talk) 15:52, 19 February 2010 (UTC)

Yes, that's the nuance that needs to be kept. I think that it also needs to be written in a way that doesn't appear to authorize, much less encourage, bouncing discussions from page to page. The problem is actually pretty limited: As an example, imagine someone changing

WT:N..." isn't going to earn him any sympathy. WhatamIdoing (talk

) 18:29, 20 February 2010 (UTC)

Yup... this happens all the time between WP:V and WP:RS. Consensus at one is shot down by Consensus at the other. We do need to stress that when an issue involves multiple pages (be they articles or policy pages), editors should create a centralized discussion and notify those at all related pages that the discussion is taking place. Blueboar (talk) 18:51, 20 February 2010 (UTC)

I have some sympathy for the problem at WP:V and WP:RS, or (currently) WP:MOS and WP:REFPUNC, because there's actual overlap between the pages. I have less sympathy for people that propose changes to WP:RS at

WP:EL, and almost none for people that propose changes to, say, Cooking at Raw foodism

.

A general "make sure all the stakeholders know about your discussion" line might be useful. (Of course, all of this is predicated on the idea that anyone reads the directions, which may be entirely false.) WhatamIdoing (talk) 03:36, 21 February 2010 (UTC)

Aye, there's the rub... A lot of editors see a title and think they know what a policy or guideline says... when in fact it says something very different. Getting them to read past a Policy page's title is often hard, and a problem that is not limited to just this policy. I sometimes think we should put incentives in the title: "Wikipedia:Consensus (free ice cream offer hidden in policy text... read carefully)"

An example of not reading the page is probably

WP:NTEMP, which is sometimes cited by those arguing the opposite of its meaning. Fences&Windows

01:57, 22 February 2010 (UTC)

My personal bugbear is "per

WP:TRIVIA" when people are trying to delete something they feel is "trivial".--Father Goose (talk

) 03:18, 22 February 2010 (UTC)

I think misunderstandings happen because we believe that other editors know what they're talking about. So you hear one person assert (credibly, but falsely) that "All high schools are notable", and you believe them, and then you repeat it... and end up totally surprised when you get to

the relevant guideline

and find out that the guideline requires exactly the same level of evidence for high schools as it does for restaurants.

I've made the same mistake: I once believed that "nobody reads the directions" was a direct quotation from

WP:CREEP. (IMO, it should be...) WhatamIdoing (talk

) 03:31, 22 February 2010 (UTC)

So how about a sentence like this:

Whenever you're discussing changes that might affect more than one page, make an effort to notify and include all of the stakeholders at the other pages.

I could expand it, but perhaps concision is the better choice. What do you think? WhatamIdoing (talk) 03:31, 22 February 2010 (UTC)

Works for me.--Father Goose (talk) 04:15, 22 February 2010 (UTC)

Is "stakeholders" intended to mean anything other than people who are watching the other pages?--Kotniski (talk) 07:44, 22 February 2010 (UTC)

Good point. Alternative:

Whenever you're discussing changes that might affect more than one page, put a notification of the discussion on every relevant talk page. If many pages are involved, a posting on

Village Pump

, or other centralized location is appropriate.

O'course, that's not quite as concise, what with the extra sentence I added.--Father Goose (talk) 09:32, 22 February 2010 (UTC)

"Every relevant talk page" creates

WP:CANVAS

problems, especially if nobody reads (all) the directions. How about:

When you're discussing changes that might affect more than one page, invite editors at the other pages to join the discussion. If many pages are involved, a posting on

Village Pump

, or other centralized location is appropriate.

Kotniski, IMO under normal circumstances, the stakeholders are the people at Talk:X and Talk:Y. However, the list might reasonably include a WikiProject or guideline page, which is why I was being vague. For example, the current RfC at

WP:IBX might be able to provide valuable advice. WhatamIdoing (talk

) 19:26, 22 February 2010 (UTC)

Revised version is fine.--Father Goose (talk) 00:45, 23 February 2010 (UTC)

It appears that some folks at

("Consensus among a limited group of editors, at one place and time, cannot override community consensus on a wider scale") as meaning that WikiProjects must go along with officially approved policies, but are perfectly free to overrule mere guidelines on the basis of exactly the sort of 'local consensus' that this policy consistently opposes.

I have added the words "or guideline" to address this apparent failure to communicate. WhatamIdoing (talk) 01:16, 28 February 2010 (UTC)

An update: we're still apparently having a few problems at WT:WikiProject Composers/Infoboxes RfC. The latest is an assertion that blind reversions and systematic edit warring in the mainspace is an appropriate response to 'losing' a long discussion about the merits and demerits of infoboxes. See, e.g, "If this infobox makes it out of draft, it would be an entirely reasonable response of concerned editors to revert its use systematically as it appears."

Overall, I think that the conversation has been successful in resolving most of the issues -- the consensus seems to be that the WikiProject doesn't actually own the articles in its scope, that infoboxes shouldn't be spammed into articles indiscriminately, but perhaps added to a small number of carefully selected articles (depending on local consensus), and that nobody foresees great demand for the emerging infobox -- but if a few more uninvolved editors happened to watchlist this discussion with an eye towards educating editors about basic policies, it might be positive. WhatamIdoing (talk) 00:44, 16 March 2010 (UTC)

On this page Wikipedia:Consensus#Consensus_building_in_talk_pages it is advised to be 'bold in editing'. I argue that this statement should be made more like: be bold in making improvements (additions). Do not be bold in deleting content. I see examples of editors routinely deleting half or whole pages or large numbers of source links because they do not find them good enough. They leave it to others to improve. In my opinion, this does not help Wikipedia forward. Wikipedia was not built on deleted pages. Policy should be to never just delete anything, but only improve content. The described practice (on the page) of one editor deleting content, and another putting it back, is a destructive, negative process. Bold deleting demotivates editors from adding content. Improve boldly, don't delete. --BalderV (talk) 12:59, 8 March 2010 (UTC)

It depends on the material being deleted and why it is being deleted. Sometimes boldness in deletion is just as important as boldness in addition. Material that blatantly violates our policies should be boldly deleted. Blueboar (talk) 16:21, 8 March 2010 (UTC)

I agree with Blueboar: Boldly removing poorly sourced information is not only desirable, but in some cases

actually required by Wikipedia's content policies. WhatamIdoing (talk

) 19:41, 8 March 2010 (UTC)

The policy that addresses these points is the

WP:PRESERVE section of Wikipedia:Editing policy.--Father Goose (talk

) 20:13, 8 March 2010 (UTC)

It is also discussed by the

WP:V. Blueboar (talk

) 21:14, 8 March 2010 (UTC)

HANDLE is an important balance to PRESERVE: Bad information should be removed, not "preserved". Good information that happens to be poorly presented should be preserved (and the presentation fixed), not removed. WhatamIdoing (talk) 23:11, 8 March 2010 (UTC)

well said. Blueboar (talk) 23:26, 8 March 2010 (UTC)

It seems to me that the policy needs to put a bit more emphasis on what consensus isn't.

To take a silly example, if a bunch of wikipedians decided that the sky was purple, does that mean that no references would be needed?

So aren't there restrictions on what consensus can decide here?

The problem is that the wikipedia is not

we're not trying to create a shared reality, we're trying to match the articles to the references and to the guidelines and policies, and the guidelines and policies to the principles of the Wikipedia.

I just think that the text needs to emphasis this. At the moment, I've seen this policy abused a few too many times; peoples attitude tends to be 'we can decide anything about anything' but that's not really what this is for, it's more fine grained than that, it's about deciding whether references actually do match the text and so forth. I don't think that people really can make the Wikipedia into a novel, or rewrite the word 'rocket' to mean a type of daffodil; but this policy implies that that's theoretically OK.- Wolfkeeper 18:27, 10 April 2010 (UTC)

Discussion (soon to be RfC) is at WT:Wikipedia is not a dictionary. -- Quiddity (talk) 22:21, 10 April 2010 (UTC)

No, that's a different policy, although the actions of some editors there does tend to support what I'm saying, but this is a far more general issue.- Wolfkeeper 14:19, 11 April 2010 (UTC)

This may sound bizarre, but I have had problems with people trying to define 'glider' to mean something like 'sailplane, and only a sailplane', and the space shuttle (which NASA says is a glider) kept getting removed from the article, (and they removed an article I tried to create on the general idea of gliders by consensus as well) and other types of gliders also got removed. But the consensus of nearly everyone (i.e. lots of sailplane pilots) was that that was perfectly proper, and they pointed to this policy saying they could do anything by consensus. It got sorted out eventually to everyone's satisfaction, but things like that have happened several times, and I don't entirely blame them; peoples POV tends to get in the way and trump objective facts, I think it's a problem of the wording of the policy.- Wolfkeeper 18:27, 10 April 2010 (UTC)

Search wolfkeeper glider for background. -- Quiddity (talk) 22:21, 10 April 2010 (UTC)

Wolfkeeper, whenever anyone claims that "consensus is king", then you can point them to the

WP:CONLIMITED

section of this page. This policy explicitly rejects the idea that you can claim "consensus" to violate major content policies.

However, I think that your arguments at

WP:NOTDIC seem weak -- and consensus is king on WP:Policies and guidelines. Since you've had the same discussion at several pages (always with strong opposition), you're also at risk for complaints about forum-shopping. WhatamIdoing (talk

) 23:39, 10 April 2010 (UTC)

To be perfectly frank, I find that inaccurate and offensive.- Wolfkeeper 22:54, 11 April 2010 (UTC)

In fact a change along these lines was already actioned at wp:policies and guidelines, I don't really think that it constitutes forum shopping if people agree to make changes along the lines you suggest and then you go somewhere else and suggest similar changes there too.- Wolfkeeper 22:54, 11 April 2010 (UTC)

That's not going to get you the full story; in the end an article was created at:

gliding (flight), and it seems to be generally agreed that that was a good idea in the end, and people ended up contributing to all of them.- Wolfkeeper

14:23, 11 April 2010 (UTC)

The problem is that the consensus policy, as it stands, has no value system inherent in it. If consensus consists of just agreeing, you have to ask what they are trying to achieve by agreeing; is it OK, if a bunch of people working for a company all join the wikipedia and then make 'consensus decisions' in the same direction (not necessarily organised thing by the company). Clearly not, if they have an ulterior motive, or if their values aren't aligned with those of the Wikipedia. Nor would we want a religious group to edit their pages to change from what would be considered neutral.

So I just think we need to make some small amount of text (a sentence or two) with some emphasis on values of the Wikipedia; without shared values consensus is far less likely to form, and it helps if people are encouraged to articulate how something they're agreeing on aligns with those values.- Wolfkeeper 14:19, 11 April 2010 (UTC)

lol - I've done this several times in the past, and it's always been reverted. whaddayagonnado. --Ludwigs2 15:24, 11 April 2010 (UTC)

It comes down to this... The long standing consensus is that there is a consensus as to what constitutes a consensus, but no consensus as to what that consensus actually is. Of course consensus can change. However, we must be careful that any consensus that claims consensus has changed is not merely a reflecting a local consensus, but one that reflects a broader consensus as to what the consensus is. got it? :>) Blueboar (talk) 15:49, 11 April 2010 (UTC)

I'm not sure I agree with that, but then I'm not sure I disagree, either. can we agree that we neither agree nor disagree, or should we discuss the matter further?

It's not about what consensus is, it's about defining what we need consensus on. For example, if we're at AFD we're looking for consensus that the article does or does not meet the deletion policy, rather than simply whether we think it should be kept (unless you apply IAR). So if somebody simply says I LIKE IT, then their vote shouldn't count, right?- Wolfkeeper 23:02, 11 April 2010 (UTC)

That is the point of CONLIMITED: The community (meaning thousands of editors) has an existing consensus to ignore "I like it" !votes at AfD. The community's consensus is documented at

normally ignore such !votes. WhatamIdoing (talk

) 01:29, 12 April 2010 (UTC)

"what we need consensus on" - this is straight-forward, I think. in article space, we need broad consensus that the article presents the topic in a fair, complete, neutral, and balanced manner. basically (the way I think of it) every editor should be writing towards the goal that some large majority of wikipedia editors would agree that what s/he's writing in fair, neutral and balanced - writing for the world (rather than writing for the enemy as it's sometimes put). --Ludwigs2 02:55, 12 April 2010 (UTC)

Right, but why fair, complete, neutral and balanced? It's because that's what the policy says, and the policy says that because the policy is best consensus on ways to try to fulfil the principles of building an encyclopedia within a Wiki.- Wolfkeeper 20:09, 12 April 2010 (UTC)

But I've also seen cases where groups of people decide they don't need to worry about (for example) completeness or neutrality or whatever, and then they quote this policy. That's the concern; it's written too much to support such use right now. I just think it needs to be quite a bit clearer.- Wolfkeeper 20:09, 12 April 2010 (UTC)

This is an overdone, POV-driven change that has gotten no support on this talk page. And that's because the wording is utterly circular: consensus on a policy can't "override" consensus on the five principles, which instruct us to... "find

consensus" on things such as policy. Oddly, the 5 pillars also tell us to "avoid edit wars", which is exactly what Wolfkeeper is doing here, so in trying to defend the principles s/he's breaking them. Wolfkeeper, please revert yourself and we can discuss how to improve the wording, but edit warring needs to stop immediately.--Cúchullain ^t/c

14:29, 13 April 2010 (UTC)

The mistake you're making is that 5P aren't policies, they're principles; they're values that you are supposed to respect in your editing. There's no circularity at all; the policies are listed and indeed exist to try to explain in more detail what these principles are really taken to mean. But those principles are what have, and are being used, to build the wikipedia; if the principles were different we would have a different thing here, perhaps a wiki covering just places or people or whatever, many different things. That's why they are incredibly central.- Wolfkeeper 14:44, 13 April 2010 (UTC)

Do you or do you not agree that consensus on principles can and should override consensus on policy? Because if you don't agree with the principles what are you doing here? Even IAR doesn't override the other principles.- Wolfkeeper 14:44, 13 April 2010 (UTC)

IMO "can and should override" is the wrong approach. I believe that policies, guidelines, and the local consensus will ultimately and inevitably agree with the five principles... and I believe that this will happen, no matter what we put on this page. WhatamIdoing (talk) 17:30, 13 April 2010 (UTC)

I agree that that ultimately that does occur, and that's why the fact that this policy doesn't point it out is so very painful. Because people so very often take the attitude 'hey, I/we can do anything, we're freeee- consensus!!!', until they eventually get run over by the principles. The whole point of policies is really to reduce unnecessary contention, and at the moment WP:CONSENSUS flakes out by not pointing out that that oncoming train that will inevitably run them down.- Wolfkeeper 17:42, 13 April 2010 (UTC)

I dislike the entire concept that X override Y. Our guidelines, policies and principles should all work in harmony. If a policy and the 5P conflict, then we need to examine why they conflict. It could be that the policy has been tinkered with to the point where it is out of sync with broader consensus. On the other hand, it could be that the policy reflects the fact that the broader consensus has changed regarding the 5P. The same is true when guidelines and policies conflict. Blueboar (talk) 14:59, 13 April 2010 (UTC)

Principles aren't rules; they're values. And, at 5P, the IAR principle says that if a policy gets in the way of the principles of the wikipedia, then you can ignore it. This is very fundamental stuff.- Wolfkeeper 15:10, 13 April 2010 (UTC)

If the broader consensus really has changed then you need to change the principles. I've never seen any example of that happening; even the most critical things that consensus has changed on, like BLP, have sat neatly under the principles (in that case NPOV).- Wolfkeeper 15:10, 13 April 2010 (UTC)

Consensus on the pillars, several of which are policies, tells us to find consensus across Wikipedia as a whole. Your wording is vague and circular and would be easy to misuse, as in this hypothetical case:

“

User1:"The consensus is that that change is unproductive." User2: "Your consensus can't 'override' the principle that Wikipedia is an encyclopedia!" User1:"Well, what do the 5 pillars offer towards resolving this abstract dispute? Oh I see: by finding consensus. Well, the consensus is that you're misinterpreting WP:NOT can't 'override' the principle that Wikipedia is an encyclopedia...

”

And so on. Also, I'll mention again that the pillars also tell us to avoid edit warring...--Cúchullain ^t/c 15:14, 13 April 2010 (UTC)

Your point is not well made at all. The policies on dispute resolution and handling edit warring in the Wikipedia are very different; they don't require consensus.- Wolfkeeper 15:47, 13 April 2010 (UTC)

Wolfkeeper, I think you're approaching this from entirely the wrong direction. The problem you're concerned about is that people misuse/misunderstand the nature of consensus, and what you are suggesting (apparently) is that we overrule consensus with a semi-authoritarian appeal to principles. That will just make the problem you are concerned about worse. The policy of consensus is (should be) an extension of the principle of consensus, and the trick here is to get people to understand the principle (in which case the policy will be applied correctly), not to put the policy and principle at odds with each other. --Ludwigs2 16:06, 13 April 2010 (UTC)

I don't think they do misunderstand it; consensus is what this page says it is; and this page studiously avoids any attempt to emphasise consensus about how things align with values, and never, ever mentions principles (except in a couple of sidebars). But we want people pulling in the same direction; the policies are too randomly built for that.- Wolfkeeper 16:20, 13 April 2010 (UTC)

I'm leery of any change to policy wording that would embolden editors who deem themselves right and everyone else wrong on matters of policy. "Consensus does not override policy" (or Arbcom / Jimbo / 5P / the Foundation / the Mission Statement) is a good axiom, and it can be a useful argument in support of a position. But it's just that, an argument. It's not a trump card. Ludwigs2 is right. Editors do become authoritarian when they attempt to overrule consensus. Depending on their persistence, standing, and recourse to administrative tools they can become abusive on the one end, or they may find themselves frustrated or sanctioned. If the issue is minor and not in one of Wikipedia's hot spots there are many self-correcting mechanisms to deal with transient local misapplications of consensus. An article built on original research, containing BLP violations, that is a dicdef of a neologism, etc., will eventually get sorted out. The higher the visibility of the article, the faster it tends to attract the eyes of the wider community, which for the most part does believe in the 5P and most of the policies. On the other hand, policy pages and content hotspots already do have the eyes of the community, and ignoring consensus in favor of personal conviction that one is right is a lot more likely to result in trouble than working things through. - Wikidemon (talk) 19:08, 13 April 2010 (UTC)

I agree. The pillars are supposed to be the underlying principles our policies should help us strive towards; they should not be used as some kind of superior court that trumps discussion and all else. The suggested wording would just give wonks an easy way to turn every content dispute into a policy dispute.--Cúchullain ^t/c 19:44, 13 April 2010 (UTC)

before there was references to

23:31, 11 April 2010 (UTC)

Jimbo as an exception

Does anyone still realistically consider Jimbo's current words to be policy ? Policy forming most definitely, but no longer policy by definition. This part of his "benevolent rulership" was mostly in the forming stages of Wikipedia and was later largely moved to Office actions. In my opinion Jimmy's words in current debates have lost much of it power in the view of the community. His actions on the wiki still exceed that of normal admins, and his words carry much more weight, but to say that they are still policy by definition in this time is not realistic I think. —TheDJ (talk • contribs) 13:08, 12 April 2010 (UTC)

That really depends on the situation. Sometimes Jimbo is commenting as a member of the community. In these situations his words are an opinion, no different than anyone else's... but sometimes he speaks "officially", and his words are more than just an opinion, they are a dictate from the foundation. In these situations his words can set policy and overturn consensus. Blueboar (talk) 13:22, 12 April 2010 (UTC)

In such a case, he speaks as the foundation board which is covered separately. As such the addition of his name indicates "personal authority" separate from the foundation, which might have hold ground up to a year or two ago, but no longer I think. —TheDJ (talk • contribs) 13:45, 12 April 2010 (UTC)

There is currently a discussion at

23:45, 6 June 2010 (UTC)

RfC

Conundrum

A rather confusing situation has developed within the series of biographies on Canadian governors general: prompted by a dispute over date formats at

WP:CONS

states that "silence can imply consent... if there is adequate exposure to the community," and there seems to have been ample traffic at those pages. Does it mean then that the changes to the date format, though against MOSDATE (a guideline), now require a consensus to be changed back, per CONS (a policy)?

A more central discussion on this matter has started at

MIESIANIACAL

02:26, 10 July 2010 (UTC)

I hate to put the it quite this way, but I can't think of anything cheerier, so here goes: no one cares what date format you use (except you and whomever you're having a dispute with). Assuming you're not using actually different dates, then you can't really see this as a matter of 'silence=consent', but more a matter of 'silence=hunh?' IMO, MOS trumps consensus here hands down, since MOS clearly represents a community consensus of some sort around styling issues, and there's really no reason to believe there is any other consensus involved in this. --Ludwigs2 02:56, 10 July 2010 (UTC)

Well, no matter how you put it, I agree: I don't care what date format is used. What I am concerned about is the ambiguity in this particular situation, which I sense has landed somewhere between WP:CONS and WP:MOSDATE, and I'd like to see it settled in one place before changes are made so that an edit war doesn't break out across dozens of articles and the consistency I spent months working on isn't lost. But, if you think the MOS guideline trumps the CONS policy, that's fine. It's opinions like that which I'm seeking. --

MIESIANIACAL

05:12, 10 July 2010 (UTC)

No MOS page has ever had adequate exposure to the community; very little of it is consensus - or good advice. This case is typical; a dozen editors sat down and devised a Rule to impose on all Wikipedia, and have ignored and revert-warred against such objections as have actually found that obscure page.(Few, if any, of them were Canadian; I think the real question was whether or not to isolate the United States. There was some discussion of imposing the "logical" format DMY worldwide, but that was never close to a majority.) If MOS disagrees with other guidelines, always ignore it. Septentrionalis PMAnderson 01:42, 22 July 2010 (UTC)

PMA, I think that the existence of MOS, and most of its sub-pages, are well-known, and generally supported. It happens that you personally were on the losing side of some MOS battles, but that doesn't mean that the community is unaware of these pages' existence or contents. Frankly, it is beyond belief to claim that MOSDATE, which spent so much time being argued about in front of ArbCom, and with banner-advertised RFCs, somehow didn't have "adequate exposure to the community". It may well be the single most "exposed" page in the Wikipedia namespace. WhatamIdoing (talk) 01:58, 22 July 2010 (UTC)

Their existence is well-known. Their contents are neither well-known nor consensus. And I am speaking of many discussions in which I played no part or prevailed; indeed, the Canadian wording was one I supported against bitter opposition. Septentrionalis PMAnderson 02:01, 22 July 2010 (UTC)

Nor did the ArbCom case show otherwise; it merely showed, to ArbCom and others, that those who wrote the mess were revert-warriors and bullies, pushing the Rule they wrote on Wikipedia. Septentrionalis PMAnderson 02:06, 22 July 2010 (UTC)

I can never quite work out why you continuously vilify the MOS people for trying to impose uniformity by force, while the royalty and peerage people do the same thing with your full support.--Kotniski (talk) 08:58, 22 July 2010 (UTC)

Because it's not the same thing; if consensus on an article is to rename peers, as Lord Byron may be renamed shortly, the peerage people will adjust their guideline to suit; they don't vote against the article at FAC - nor do other useful -er- editors - nor do they write bots to impose their whims. Septentrionalis PMAnderson 13:29, 22 July 2010 (UTC)

The other difference is the peerage people are choosing among what they find in sources, on a question where we must choose: an article must have a title, and can only have one. All too much of MOS lays down rules where we could easily differ, and does so on the basis of unenlightened and provincial ideas of what English is and what would be best for it, not what English usage is or style guides say. Septentrionalis PMAnderson 13:44, 22 July 2010 (UTC)

My suggestion... date format goes with spelling (which is usually, but not always, set by the creator of the article)... so if the article uses UK spelling, use UK dating format... if it uses US spelling, use US dating format. And if a consensus of editors at the article changes the spelling, change the date format with it. But don't edit war over the issue. Simple. No conflicts. Follows MOS, Arbcom, and all other Policy/guidelines. Blueboar (talk) 02:46, 22 July 2010 (UTC)

That makes far too much sense for MOS to have anything to do with it. Septentrionalis PMAnderson 13:55, 22 July 2010 (UTC)

Canada-related pages can use either mdy or dmy. Very few use dmy. If it comes to later arguments about the format in an article (really, who would care?), it is supposed to come down to the way it way it was in the initial stages. "Now,

WP:MOSDATE seems to indicate (though not terribly clearly) that the first date format used should be the style always used." It is quite clear. However, if consensus can be built to change the existing format, that's fine, and please move on. Personally, I would opt for stability and not bother changing a Canadian article, but it's up to the editors who do care. As for Anderson's continual warring over the status of the style guides, it has been his campaign for many years to weaken the coordination of style at WP, and usually begins with the accusation that the style guides are ruled by a few power-hungry editors. Tony (talk)

22:12, 22 July 2010 (UTC)

I didn't say anything about power-hungry editors. Where did Tony come up with the idea? Observation? Septentrionalis PMAnderson 02:38, 23 July 2010 (UTC)

I think

needs to be reworded slightly. I don't claim to write to professional standards, but here is my attempt at it:

Consensus among a limited group of editors, at one place and time, cannot override community consensus on a wider scale. For instance, participants in a WikiProject cannot decide alone that a generally accepted policy or guideline does not apply to articles within its scope. When a consensus that contradicts an established Wikipedia convention is reached, participants should normally seek the opinions of a wider group of editors, for instance at a

Request for Comment

. Edit warring over such content is unacceptable.

In the case of policies and guidelines, Wikipedia expects a higher standard of participation and consensus than on other pages. In such cases, silence can imply consent only if there is adequate exposure to the community.

Thoughts welcome. Regards, --WFC-- 09:52, 29 July 2010 (UTC)

That seems like a reasonable expansion. I'd like to see a specific mention of "editors at a specific article", because I'm seeing signs that at least a couple of editors aren't getting it. It'd be silly to say that a dozen participants at a WikiProject can't override widely supported "rules", but that three random editors at an individual article can do exactly that. WhatamIdoing (talk) 03:38, 13 August 2010 (UTC)

I was going to revert the removal of your recent change (because I support the concept that policies/guidelines should spell out what they mean), but when I looked at it more closely, the first sentence ("Consensus among a limited group of editors, at one place and time, cannot override community consensus on a wider scale.") does make your addition redundant. Further, your change slightly broke the "for instance" which talks about "articles within its scope" – that's fine for the existing example, but needs some rewording to point out that obviously a few editors at an article cannot decide that some policy does not apply to the article. Johnuniq (talk) 04:42, 13 August 2010 (UTC)

Of course, we would like to think that was the case (subject to IAR), but is it actually true? In the absence of any mechanism to enforce policy (except a few that the powers-that-be have decided to take seriously), I think a few editors at an article can in practice decide anything they like, unless and until a larger number of editors can be brought in to bring the local consensus back into line with the global one.--Kotniski (talk) 14:26, 13 August 2010 (UTC)

Johnuniq, I agree that my change needs copyediting.

Kotniski, while it's true that just about anything can be done -- temporarily -- at an underwatched article, the fact remains that there is, and can be, no actual "consensus" to vandalize the encyclopedia or deliberately violate community standards. "But the three of us agreed to spam this article!" is not an acceptable or effective justification. Spam is still spam, and unsourced scurrilous attacks on living people are still unsourced attacks, (etc) even if you get two or three people to "agree" to violate the community's standards on a given page.

Editors at an individual article can (and must) use their best judgment to figure out how to implement our content policies, but figuring out whether a statement complies with BLP (for example) is not the same as declaring that you and your two best friends have a "consensus" to deliberately violate BLP. WhatamIdoing (talk) 18:44, 13 August 2010 (UTC)

Yes, BLP is one that tends to be enforced even in the face of apparent local consensus, but the same is hardly true of all policies (nor would we want it to be, hence IAR). You won't get an admin to intervene to enforce a neutral point of view, for example - they'll just say it's a content dispute, to be resolved by discussion, and the policy will only be "enforced" once enough non-POV editors have been attracted to the page to arrive at a neutral wording - whereby the consensus is duly changed.--Kotniski (talk) 08:19, 14 August 2010 (UTC)

Not always: People do get blocked for POV pushing, and many of ArbCom's biggest headaches have centered on this sort of problem.

On a related note, I think that it might be useful to write a short section, "When consensus isn't possible." What do you think? WhatamIdoing (talk) 06:12, 15 August 2010 (UTC)

I've never heard of anyone being blocked for POV-pushing when their edits have been supported by local consensus, though I suppose it may have happened. Anyway, as to your suggestion, most definitely yes - this policy shouldn't just consist of idealistic eyewash, but should give specific and useful advice on how to deal with the difficult problems that regularly recur.--Kotniski (talk) 07:30, 15 August 2010 (UTC)

I've started the suggested section at Wikipedia:Consensus#When_consensus_isn.27t_possible and would be happy to have other editors improve it. WhatamIdoing (talk) 02:09, 25 August 2010 (UTC)

It looks like Wolfkeeper has declared this to be a bad-faith effort. WhatamIdoing (talk) 05:54, 25 August 2010 (UTC)

It would be interesting to know why. I thought the section as written was a pretty good start (though if it's the external links thing that's objected to, we could leave that out).--Kotniski (talk) 12:51, 25 August 2010 (UTC)

Presumably it is his primary concern (and yes, three examples might be a bit much). But I suspect that he invokes this page as his get-out-of-jail-free card in more circumstances than in trying to preserve an external link to a favored comic strip. WhatamIdoing (talk) 16:25, 25 August 2010 (UTC)

So CONSENSUS is my own personal get out of jail free card is it? I wrote it and installed it in the Wikipedia did I? Oh wait, no, that's ELBURDEN, and that's precisely what you did. You make me sick to my stomach.- Wolfkeeper 17:00, 25 August 2010 (UTC)

I agree that your summation is an accurate description of the current state of policies/guidelines.

However, I also somewhat agree with the concerns raised at Wikipedia:Village pump (policy)#WP:External links 'default removal' of any link that is contested (more or less) by anyone (basically, that the EL guidelines cater to exlusionists, who are often the primary debaters at that guideline's talkpage).

Fwiw, I would consider myself a moderate External Link inclusionist (ie this is too many links (a current Good Article) – although the quality of the links is fine, there are simply too many – but this is a perfectly acceptable number (recent Featured Article)). In contrast, many EL talkpage debaters have stated that they believe there should generally be very few, if any, external links. I agree that both mindsets are necessary, in order to keep Wikipedia functioning in all its aspects (ie, conservatives and liberals balance each other out, ideally (and ideal situations are as rare as commonsense...)).

I don't really have any suggestions to resolve this, at the moment. (Partially because I'm nervous about get my first ever fillings in 2 hours. People! Floss thy teeth!)

Perhaps a reinforcement somewhere of the advice to "move disputed links to the talkpage, instead of just deleting them". ? HTH. -- Quiddity (talk) 20:04, 25 August 2010 (UTC)

The 'when consensus isn't possible' section

This section was badly written.

1. Consensus is always possible except where editors decide (for reasons ranging from advocacy to emotional attachment) not to seek it. Please don't legitimize dysfunctional behaviors as standard and accepted.
2. The wording made it sound as though wp:BURDEN and wp:DEL were dispute resolution processes, when in fact they are authoritative processes that bypass both DR and Consensus.

The section may or may not be a good idea (certainly some pointer to DR processes would be useful), but can we find a more pro-consensus phrasing for it? I don't see the point of adding a section to the policy which essentially shoots the policy in the head. --Ludwigs2 14:58, 25 August 2010 (UTC)

Really? Consensus is always possible? Then why is ArbCom so busy?

Oh, I see: You said, "Consensus is always possible except" when it's not. So since you admit that consensus is not actually always possible in practice, do you mind if we have a section here that acknowledges it and points people to

WP:DR, i.e., what to do when consensus isn't possible in practice -- something that doubtless comes up when dealing with the fraction of our 50,000 editors each month who turn out to be flawed humans rather than virtuous paragons? WhatamIdoing (talk

) 16:23, 25 August 2010 (UTC)

Yes, I was about to write something similar - indeed, even if editors were virtuous paragons, there would still be situations where consensus was not possible, simply because there are some issues on which no middle solution exists and there are too many editors on both sides for it ever to be possible for one side to persuade the other. In fact I would merge this page with DR and possibly one or two others - I don't think there's a clear division between the consensus-seeking process and the "dispute resolution" process.--Kotniski (talk) 16:39, 25 August 2010 (UTC)

My point is that if people maintain a degree of neutrality, detachment and civility, they can always talk through an issue to reach some sort of consensus. This is the basic aspiration on which all of the project's core principles are founded, so it's probably not a good idea to flout it outright. ArbCom and ANI are busy, of course, because there are a lot of editors (as Kotniski notes) who cannot or won't consistently maintain that necessary degree of neutrality, detachment and civility. The fact the fact that editors sometimes act like bighorn sheep in rut is not a condemnation of the consensus process, but rather a morality tale about the evils of misapplied passions.

So, if we want to include a section like this (which I don't object to on principle), I ask that we (1) maintain the pleasant fiction that people can reach consensus as a rule, and that non-consensus is an unfortunate departure from the norm, and (2) give useful advice about consensus-building and consensus-determining processes rather than instructing people about what they can and cannot unilaterally delete. would that be OK with you guys? --Ludwigs2 16:53, 25 August 2010 (UTC)

Well, isn't that what the rest of the page is doing? This section would just be a small part of the page overall, so it seems reasonable for it to focus on the topic that it's been written to address.--Kotniski (talk) 17:01, 25 August 2010 (UTC)

I have no objections to including phrases like "In rare circumstances" to remind people that intractable conflicts aren't the universal, constant experience. Most people don't realize that normal, everyday cooperative editing is consensus-based work, so they don't give it credit for working.

When they have a reason to look up this page, they usually need help. I admit that my beginning effort didn't really give them much help -- but now we're giving them basically zero help, which IMO is worse. Ludwigs, perhaps the thing to do is for you to restore that section and improve it, instead of removing it entirely for being imperfect. WhatamIdoing (talk) 17:11, 25 August 2010 (UTC)

Also, I'd like to say that we already "give useful advice about consensus-building". You may have perhaps noticed that large section,

WP:CON#Consensus-building? It's more than 50% of the current policy (by character count). I do not think that we need to repeat that large section in the tiny section about what to do after you've followed all the advice in #Consensus-building and it didn't work. WhatamIdoing (talk

) 17:15, 25 August 2010 (UTC)

I'll re-add it with revisions later this afternoon (need to do stuff for a bit). I don't think the "It's already there" argument quite works, though; consistency is important. --Ludwigs2 18:02, 25 August 2010 (UTC)

Seems like it just needs a lot of rewriting. --

talk

) 18:15, 25 August 2010 (UTC)

well, I restored it and did a rewrite, but as I was rewriting it really seemed entirely redundant - shouldn't we just merge this into the consensus building section? --Ludwigs2 07:51, 26 August 2010 (UTC)

(bit off topic) Everything should really be merged - as it is we have various different sections and different pages addressing the same topics, saying different things that don't really serve any purpose and don't mesh together into a coherent whole. How about one single policy on "Resolving disagreements"?--Kotniski (talk) 09:42, 26 August 2010 (UTC)

It's not a bad idea, but speaking purely of the practical issues (i.e., the excessive bias towards inertia where policies are concerned), I bet that the community won't accept it, even if the merge proposal had no downside and massive upside. If you want to pursue this, you might mention it at

WT:DR. WhatamIdoing (talk

) 19:29, 26 August 2010 (UTC)

I think it's a good idea too. maybe we should start by making a list of all the conflicting places this issue is addressed so that we can rationalize them all. In the meantime, I'll be bold and merge the two sections on this policy. --Ludwigs2 16:42, 31 August 2010 (UTC)

well, I went and took a stab at this, but I'm not entirely pleased with the result. I had to reorganize a bit to make it all make sense, and now the last section - the 'pitfalls and errors' bit - seems much more like it belongs in a guideline than in policy. what fo you think? --Ludwigs2 01:37, 9 September 2010 (UTC)