Misplaced Pages

Revision as of 00:24, 4 November 2011

The project page associated with this talk page is an official policy on Misplaced Pages. Policies have wide acceptance among editors and are considered a standard for all users to follow. Please review policy editing recommendations before making any substantive change to this page. Always remember to keep cool when editing, and don't panic.

This is the talk page for discussing improvements to the Consensus page.

Put new text under old text. Click here to start a new topic. New to Misplaced Pages? Welcome! Learn to edit; get help. Assume good faith Be polite and avoid personal attacks Be welcoming to newcomers Seek dispute resolution if needed

Archives: Index, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24Auto-archiving period: 30 days

Archives

Index Archive 1 Archive 2 Archive 3 Archive 4 Archive 5 Archive 6 Archive 7 Archive 8 Archive 9 Archive 10 Archive 11 Archive 12 Archive 13 Archive 14 Archive 15 Archive 16 Archive 17 Archive 18 Archive 19 Archive 20 Archive 21 Archive 22 Archive 23 Archive 24

This page has archives. Sections older than 30 days may be automatically archived by Lowercase sigmabot III.

"Consensus is a partnership between interested parties working positively for a common goal." -- Jimmy Wales

“Respect the past consensus” and “consensus can change”

Whilst a paragraph in Process section (beginning with “Some articles go through extensive editing…”) advises us against “forcing everyone to rehash old discussions without need”, Consensus can change section reminds us that “consensus is not immutable.”

I think we should always keep good balance between “Respect the past consensus” and “Consensus can change.” I suspect that giving only one message without the other may, in some cases, influence the editor’s behaviour in a negative way. I believe these passages should always be read together.

In this regard, I would like to suggest that the paragraph in “Process” section be moved to “Consensus can change” section, which will then read:

"Some articles go through extensive editing and discussion to achieve a neutral and a readable product. Similarly, other articles are periodically challenged and/or revised. This is a normal function of the ongoing process of consensus. It is useful to examine the article's talk page archives and read through past discussions before re-raising an issue in talk – there is no sense in forcing everyone to rehash old discussions without need.

However, consensus is not immutable. Past decisions are open to challenge…"--Dwy (talk) 21:21, 14 September 2011 (UTC)

• Agree with this proposed change. I find the wording of the "Consensus can change" section rather chatty and thus not optimal for a policy statement. However, I think your proposal is a step in the right direction. If there is general agreement and the change is made, perhaps we could then try to tighten up the wording of that section. Sunray (talk) 00:10, 17 September 2011 (UTC)

Encouraged by Sunray’s support, I implemented the proposed change. I would appreciate it if Sunray or anybody else could make further revisions to tighten up the wording or otherwise improve the section.--Dwy (talk) 10:01, 22 September 2011 (UTC)

Section: Consensus-building pitfalls and errors

The section WP:FORUMSHOP may need revision. Here's the verbatim text:

"Forum shopping, admin shopping, and spin-doctoring. Raising the same issue repeatedly on different pages or with different wording is confusing and disruptive. It doesn't help to seek out a forum where you get the answer you want, or to play with the wording to try and trick different editors into agreeing with you, since sooner or later someone will notice all of the different threads. You can obviously draw attention to the issue on noticeboards or other talk pages if you are careful to add links to keep all the ongoing discussions together, but best practice is to choose one appropriate forum for the consensus discussion, and give (as much as possible) a single neutral, clear, and objective statement of the issue. See also Misplaced Pages:Policy shopping."

The language here is assumptive, that if people state similar ideas on various discussion pages they are trying to reach a consensus through manipulation. This sets a poor precedent, because oftentimes there is topic similarity. This also limits discussions to one page that may be similar on other pages, in which case an editor has to delay posting a comment on a discussion page and search through many other pages to find the most relevant page to base a topical discussion upon. Again, oftentimes topics are highly similar. An example is pages about the topic of deleting articles. Should a user who wants to posit suggestions to improve the AfD process limit their discussion to one page which fewer people will view, or post the information on multiple pages to encourage the building of a true and actual consensus? Many Misplaced Pages users don't view discussion pages, and it seems that the wording of this part of the policy may limit and inhibit the actual building of consensus. As presented, the policy is assumptive that users want to manipulate Misplaced Pages, which is quite likely what only a small minority of users intend to do. Northamerica1000 (talk) 10:42, 22 September 2011 (UTC)

Normal practice is to start discussion in one place, and to post links to it to all other relevant pages. This is not forumshopping, it causes no delay, it makes sure that all interested editors can participate, and it makes it possible to reach one consensus instead of multiple, possibly conflicting ones. For more background on this particular case, please see Misplaced Pages:Administrators' noticeboard/Incidents#Please stop User:Northamerica1000. Fram (talk) 10:49, 22 September 2011 (UTC)

Discrepancy between this page and Policy

This page seems to require talk page discussion before changes to policy pages. The policy policy however allows the bold editing of policy pages. I personally feel that bold editing should be allowed and the wording here modified, although I agree we should admonish caution when making changes. I think an outright ban on editing policy pages(eliminating the possibility of bold, revert, discuss and totally contrary to WP:BOLD) is a bad direction to take the policy of policy, after all, that is how all these pages were created in the first place. Simultaneously (somewhat ironically) I am tepid enough to talk about this discrepancy first, rather then just editing the page. Thoughts? Crazynas 04:16, 12 October 2011 (UTC)

I do not see any conflict—they both say the obvious, namely that normal procedure would be to discuss a proposed change to a policy rather than jumping in and substantively changing it. However, this is not a bureaucracy and there may be exceptions, so "no prior discussion" is not reason to revert a good change. Johnuniq (talk) 06:20, 12 October 2011 (UTC)

I don't read this page or any page as prohibiting the bold editing of policy. If you can improve a policy page, by correcting or improving a description of policy in practice, then do it. What is discouraged is the bold editing of policy as an attempt to "change the rules" as a means to changing practice. Policy pages, freely edited, should lag current practice and describe recent practice. New ideas, and new rules should be discussed first. --SmokeyJoe (talk) 06:59, 12 October 2011 (UTC)

Should vs. May

The older but still valid method is to boldly edit the page. Bold editors of policy and guideline pages are strongly encouraged to follow WP:1RR or WP:0RR standards. and Substantive changes should be proposed on the talk page first seem to directly conflict. If multiple methods are allowed we should say that or at least not imply that there is only one course of action for change. If the community view (or established practice) is that the only time we don't start at the talk page is when we're ignoring all rules we should reflect that consistently across the policy pages as well. The simple change I attempted was to change should to may in the second passage, undone with the comment "should" is still correct, with rare exceptions per usualWP:BURO and WP:IAR. It seems to me that saying should is inherently bureaucratic as a requirement, where leaving it as a strongly encouraged option would be sufficent. Thoughts? Crazynas 08:28, 12 October 2011 (UTC)

Are you disagreeing with the statement that substantive changes to policy pages should be proposed on the talk page first? How would replacing "should" with "may" (diff) help? Johnuniq (talk) 10:52, 12 October 2011 (UTC)

"Should" works better than "may". There is a slice of Lie-to-children here. If you need to read this page to know what to do, then you should propose first. If you don't need to read this policy, then it matters little what the precise words are. Other constructs may improve upon "should". --SmokeyJoe (talk) 10:56, 12 October 2011 (UTC)

I see your point. Perhaps the solution is a re-write from "Substantive changes should be proposed on the talk page first...:" to something like "Accordingly, the best practice is to propose substantive changes on the talk page first ..." Butwhatdoiknow (talk) 10:59, 12 October 2011 (UTC)

There are probably three kinds of changes being mixed up here. One is merely cosmetic changes to policy pages (spelling corrections, wording improvements, even quite substantial reorganization) - these shouldn't require prior discussion any more than changes to WP articles would. The second is changes that change the substantial meaning of the policy page (i.e. change the rules, if policy pages are seen as rules) to bring it into line with what is actually accepted best practice. The third is changes that aim to change what is accepted as best practice. It's the third that obviously ought to require prior discussion (you can't say that the community accepts something that it doesn't yet do unless it explicitly says it wishes to start doing it). But the first two types of changes ought to take place just like changes to any other page. Unfortunately in practice they often don't, since certain people are very trigger-happy with the revert button on policy pages, having got i into their heads that these pages contain some kind of semi-sacred text (or at least, that they always describe best practices eloquently, accurately and authoritatively, which is unfortunately not the case). --Kotniski (talk) 11:32, 12 October 2011 (UTC)

As to polishing up the language my feeling is perfect is the enemy of good. If somebody sticks in something better fine but I don't want anything contorted which is exactly and perfectly true and consistent. Dmcq (talk) 12:37, 12 October 2011 (UTC)

Majority vote?

Consensus is not necessarily unanimity. Ideally, it arrives with an absence of objections, but if this proves impossible, a majority decision must be taken. More than a simple majority is generally required for major changes.

This edit introduced a novel policy; it contradicts WP:NOTDEM, and does not seem ever to have been discussed at this page. Its effect, if taken literally, is to replace consensus by majority vote any time a closer chooses to invoke it.

I trust that it is intended more narrowly: that sometimes a decision must be taken, and if it is absolutely necessary to act one way or the other, the majority prevails.

We are not a government; there is very little we must do, that cannot wait for discussion to include as large a proportion of editors as possible. On the assumption that this was what was meant, I shall edit to:

Consensus, on Misplaced Pages, is not necessarily unanimity. Ideally, it arrives with an absence of objections, but often we must settle for as wide an agreement as can be reached. Silence on the disputed issue is sometimes an option, but in those cases where a decision is absolutely necessary, and compromise is impossible, it may be necessary to follow the majority. More than a simple majority is generally required for major changes.

Please note that even this redefines "consensus." Off Misplaced Pages, it does mean "Agreement in opinion; the collective unanimous opinion of a number of persons," as the OED puts in (italics mine). Septentrionalis PMAnderson 19:04, 29 October 2011 (UTC)

If consensus in the real world requires unanimity, then Misplaced Pages redefined consensus a long time before this page caught up.--Kotniski (talk) 20:06, 29 October 2011 (UTC)

Not sure what you're saying here, but I think the proposed change is an improvement. The way I read the new paragraph, it does not imply that unanimity is a prerequisite to consensus. Crazynas 20:10, 29 October 2011 (UTC)

No, sorry, I wasn't objecting to the new paragraph, just pointing out that the original change to this page (way back whenever) that introduced talk of a majority was hardly a novelty - decisions have been called "consensus" without being unanimous for as long as I've been on Misplaced Pages, and probably ever since the total number of editors rose above about five. --Kotniski (talk) 20:37, 29 October 2011 (UTC)

Let me distinguish:

• For us, decisions don't have to be unanimous to be consensus. I agree, and that is in the new text.
• But for us, decisions made solely by majority vote are and ought to be vanishingly rare. They would have to be both situations where we couldn't simply say "no consensus" and leave the matter alone and ones where there was no hope of attracting wider approval by amending the proposal ("not A or B, but A'. ")

The novelty here consisted in suggesting that majority votes were somehow normal.

These are separate points; perhaps they should be separate paragraphs in the policy. Septentrionalis PMAnderson 02:27, 30 October 2011 (UTC)

Looking at my own language, I think one more tweak is warranted. When there is no consensus, sometimes we are silent, but more often we leave things are they are. Septentrionalis PMAnderson 20:05, 31 October 2011 (UTC)

Consensus, on Misplaced Pages, is not necessarily unanimity. Ideally, it arrives with an absence of objections, but often we must settle for as wide an agreement as can be reached. When there is no widespread agreement, consensus-building involves adapting the proposal to bring in dissentients without losing those who accept the proposal.

When no widespread agreement is possible, we call that no consensus. Often this results in no change or in silence on the disputed issue. If there is a case where a decision is absolutely necessary, and compromise is impossible, it may be necessary to follow a simple majority. More than a simple majority is generally required for major changes.

This seems a better and clearer description of what we do. Septentrionalis PMAnderson 20:12, 31 October 2011 (UTC)

Agree. "... and a majority decision must be taken" is a statement of opinion that doesn't belong in this policy. Where an active decision must be taken, WP:Rough consensus (see also Rough consensus) may be relevant. Or WP:DR. --SmokeyJoe (talk) 21:12, 31 October 2011 (UTC)

• The change to the policy was made without consensus, and is also contrary to the spirit of the policy. I think it should be removed completely. By "completely" I mean that we should not replace it with a watered down or modified version. Misplaced Pages does not work by majority votes. To suggest "we work by consensus, but when we can't get it we settle for a majority vote" is nonsense, because it amounts to no more than a roundabout way of saying "we work by majority vote", which is just not true. JamesBWatson (talk) 21:26, 31 October 2011 (UTC)

The draft above included a majority vote because we do, on rare occasions, use one: ArbCom elections, for example. But I have no problem treating those as IAR. Septentrionalis PMAnderson 21:48, 31 October 2011 (UTC)

If there is a case where a decision is absolutely necessary, and compromise is impossible, it may be necessary to follow a simple majority. More than a simple majority is generally required for major changes.

is removed. Thoughts? Septentrionalis PMAnderson 21:51, 31 October 2011 (UTC)

Three thoughts:

• That consensus does not require unanimity has been in this policy for years. If "consensus is not unanimity", then consensus is (sometimes) a (weird kind of) majority (including, importantly, that you "count" the !votes of the whole community when you're determining the majority, so that any one editor easily outvotes any number of vandals).
• The current discussion is prompted by the RFC at WT:V. To avoid any time-wasting allegations about "gaming the system", I strongly suggest making no changes until that RFC is firmly, totally, definitely over with, including plenty of time for whoever feels aggrieved by the outcome to go off and complain in whatever (usually multiple) forums they choose. If this line can be improved, then it can be improved a month from now.
• There are times when "no consensus" is an inappropriate, or even impossible, outcome. We cannot simultaneously delete and not-delete images; we cannot simultaneously elect and not-elect people to the Arbitration Committee, and we cannot simultaneously block and not-block users. Rather than pretending that a true consensus is always possible, we actually pre-define the meaning of "no consensus" in some situations (e.g., "no consensus" at AFD is treated exactly like a full-consensus keep). This policy has to cover all of those situations, not just the text-changing disputes that are clearly on the OP's mind. WhatamIdoing (talk) 22:02, 31 October 2011 (UTC)

  I was prompted to read the section by a quote at the WT:V discussion, but that is not why I object to it; I did not realize, despite citing this policy for years, that Kotniski had added a reference to majoritarianism, and I would think it unwise if nobody had cited it. I would object to a restoration of the text at the head of the page; this entire discussion casts doubt on it ever being consensus - or indeed what Kotniski actually meant to say.

  As for What's examples: those are the sorts of thing that belong in guidelines and other policies; specifying them here makes them too hard to reconsider. Do any of them fall outside the range of no consensus = no change (deletions, moves, blocks) or no consensus = silence? Septentrionalis PMAnderson 22:13, 31 October 2011 (UTC)

  Certainly there are exceptions. If there is no consensus on whether something is a copyright violation, we normally delete it—a change. If there is no consensus behind a block, then the account gets unblocked—a change. RFAs normally pass with 70%, even though that's technically "super-majority" rather than "consensus".
  More importantly, "no change" can be meaningless. Imagine that I create an article. The next day, you add a sourced paragraph to it. Some people say your paragraph is off-topic; others believe it's relevant. The discussion dies out with no consensus. What does "no change" mean in this situation? No change, so your newly added paragraph is reverted? Or no change, so your newly added paragraph is kept? Either of these choices could legitimately be construed as "no change". WhatamIdoing (talk) 01:28, 1 November 2011 (UTC)

My thoughts on this are in my essay: User:ASCIIn2Bme/What "no consensus" really means. ASCIIn2Bme (talk) 22:41, 31 October 2011 (UTC)

Guys, this is really getting away from the spirit of consensus. If you want to make a meaningful revision, I think you want to lean this way:

Consensus is a different concept than a numerical count. It is not unanimity and not majority rule, but rather a process of negotiation and balancing intended to create an outcome that most people will consent to, even if they don't agree with it 100%. Consensus in the natural outcome of a healthy collaborative process, where editors have worked together to craft an article they can all live with.

In some cases consensus is not achievable, usually because editors working on the article are unable to edit collaboratively due to ideological differences or personal frictions. In such cases editors may seek outside comments to break the stalemate, using tools such as RfC's. But these are not intended to be votes or efforts at establishing some 'majority opinion'; they should be considered extensions of the consensus process that bring in new editors to overcome the failure of collaboration on the page. Often, in fact, an RfC will produce a result unrelated to its original request, merely because new editors add fresh perspectives to the dispute.

--Ludwigs2 05:00, 1 November 2011 (UTC)

Whatever we do we must not remove or hide the key information that on Misplaced Pages (in contrast to many cases in the real world) the word "consensus" is often used to describe a situation where there is far from unanimous agreement. One gets the feeling that this page, like so many other policy and guideline pages, is not being written with the intention of conveying to people clearly how we actually do things, but in a weaselly way in order to produce (or eliminate) statements that can be used as "arguments" elsewhere.--Kotniski (talk) 11:03, 1 November 2011 (UTC)

There is likely to be a difference between Wikiepedia's practices and its rules. That is common for institutions and is not necessarily a bad thing because it allows the solution of difficult cases. Definitions or rules are reductive and, as Kotniski notes, are often written with an eye to how they might be misappropriated; this, too, is normal. This page tries to express the practice of consensus on Misplaced Pages without writing in an excess of procedure and that rightly leads to editors using "consensus" and its related policies on the hard cases. I don't think we want to discourage that. --Ring Cinema (talk) 14:12, 1 November 2011 (UTC)

One more thing. There is a sensible rule of thumb for when a reasonable consensus is reached: when a majority of the minority accepts a proposal, that is very good evidence that the minority views are recognized and incorporated. --Ring Cinema (talk) 14:19, 1 November 2011 (UTC)

The issue is what to do under a very specific circumstance: (1) consensus cannot be reached and (2) a decision must be taken. For example: an image has been reported as a suspected copyright violation. The discussion is strongly divided. We cannot both "keep" and "delete" the image. What do we do? Pretend that we have a consensus anyway? Keep the discussion open, possibly for years (a process that puts the WMF in legal jeopardy if it actually is a copyvio), in the hope that a true consensus will magically appear someday?

We all want consensus. Actually, we all want perfect unanimity, because it's simple and easy to interpret. But it is not always possible to achieve this. And sometimes (not most of the time), we actually have to make a decision despite the obvious absence of consensus.

Three days ago, this policy talked about what happens in that (rare) circumstance. Now it doesn't, and we're back to pretending that consensus will always appear, if only you talk long enough (and block irritating people often enough).

Think about the possible copyvio: There's no consensus, with strong arguments and good editors on both sides, and we must ultimately make a decision one way or the other. What would you do? Endlessly extend the discussion? Always keep the disputed image? Always delete the image? Go with the majority? Stick your head in the sand and pretend that choosing to take no action is not itself a decision? WhatamIdoing (talk) 15:36, 1 November 2011 (UTC)

New paragraphs are covered in WP:BURDEN: the burden of establishing consensus rests on those who would add material, so it defaults to no addition, which is no change from the extablished text.

Copyvio (and BLP) fall under the same principle: adding material requires consensus. They also have a touch of IAR, since we are acting under legal constraints.

Nevertheless, I don't think that we should list such cases here: discussing things on two different pages leads to divergence. Is there a list of places where the meaning of "no consensus" is pre-decided? That would be a useful addition, as another subsection. 15:44, 1 November 2011 (UTC)

I don't find it useful to think of an IfD as a case where a decision must be taken; if that were the case, once an image was kept, that would settle it indefinitely. But there are, in practice, three resolutions: delete, keep and stop asking, and no consensus. If there is no consensus, there can easily be, and often are, repeated IfDs until consensus does form. Septentrionalis PMAnderson 15:50, 1 November 2011 (UTC)

Once positions are ossified, there should be an implicit burden on the minority to craft a proposal that their own like-minded editors endorse, as that mitigates the possibilities of stonewalling. If a minority's proposal is accepted by a majority of the majority, that should be good enough, as a practical matter. Similarly, a majority proposal accepted by a majority of the minority should end the discussion as a practical matter. That recognizes that unanimity is not always possible, that sometimes a decision must be taken, that decisions are revisable, that minority stonewalling is bad and that, in the absence of anything else, majorities are not irrelevant. --Ring Cinema (talk) 16:30, 1 November 2011 (UTC)

Thinking about that paragraph, does a brand-new stub really count as "established text"? If you've had something for a few years, then it's easy to claim a "prior consensus" (a claim, BTW, that violates the CCC provisions of this policy), but it's silly to make such a claim for two paragraphs whose age differs in only a matter of minutes or hours.

About the image in my example, the fact is that we don't know whether it's a copyvio: we can't make up our minds. Deciding to choose the default is, itself, a decision.

We have many defaults defined in scattered places. A unified list might be useful. WhatamIdoing (talk) 16:57, 1 November 2011 (UTC)

Certainly we should say something about how the "consensus" principle is applied in different areas. It's quite different with deletion decisions, for example, than with day-to-day article editing decisions, and different again for major policy changes (or even minor policy wording changes that people get emotional about, like this "verifiability not truth" nonsense).--Kotniski (talk) 08:42, 2 November 2011 (UTC)

Consensus seems to offer the best method to establish and ensure neutrality and verifiability

Why do we need to say this in the lead? What is it even supposed to mean? What other methods have been considered? Is this the sole or main raison d'etre of the consensus principle?--Kotniski (talk) 08:44, 2 November 2011 (UTC)

To remind those of us who would prefer to run by 51% that this is not mere philosophy; it's how Misplaced Pages chooses to do things, because - when practiced - it works. Septentrionalis PMAnderson 13:45, 2 November 2011 (UTC)

But why specifically "to ensure neutrality and verifiability"? Surely in so far as it works for those two things, it works just as well for all other things we find desirable? (And again, what other methods are we comparing it with? Who has done the comparison? On what basis do they conclude that consensus "seems best"?)--Kotniski (talk) 13:49, 2 November 2011 (UTC)

Neutrality, because a truly neutral text really needs agreement from rational readers from all the contending sides; finding them is the problem. Verifiability needs the same readers, as a check on what we need to verify.

As for the rest, this is a program; those who want to see what an encyclopedia genuinely produced by majority vote would look like, with all the sockpuppetry and meatpuppetry that implies, are welcome to take a fork. Septentrionalis PMAnderson 19:35, 2 November 2011 (UTC)

Well, majority vote is not the only alternative to what we call "consensus" (assuming there really is meaningfully a "method" corresponding to our use of that word). And my objection to saying "verifiability" and "neutrality" was not that consensus fails to produce these goods, but that consensus is equally successful (or arguably unsuccessful) at producing our other goods (completeness, clarity, usability, etc.), so there seems to be no particular reason to single out these two.--Kotniski (talk) 09:06, 3 November 2011 (UTC)

Completeness and clarity are less valuable to us; we are more willing to tolerate an incomplete or turgid article, on the grounds that it will get fixed before WP:DEADLINE, than OR or bias. Also, they are more likely to be attained by individual work, not requiring consensus. But if you want to say something like including neutrality and verifiability, suggest language. My reason for not saying something myself is that I don't see how to qulaify without verbosity Septentrionalis PMAnderson 13:20, 3 November 2011 (UTC)

I'm not convinced that it's the best method to establishing verifiability. For example, recruiting, training, and retaining subject-matter experts would often be more effective.

I more or less agree with the claim that it produces NPOV compliance. Additionally, it normally produces what we might call good editorial judgment: if everyone agrees that these two articles ought to be merged, then that probably is the best choice. WhatamIdoing (talk) 17:16, 3 November 2011 (UTC)

Yes, although "it" (assuming there is an "it") doesn't necessarily mean "everyone agrees".--Kotniski (talk) 17:53, 3 November 2011 (UTC)

what consensus is

I changed this section to read as follows:

Consensus is a process in which editors seek to reach a mutually acceptable decision through discussion. Sometimes the term is used to refer to an ongoing process (a consensus discussion) and sometimes to the result of a past discussion (a current consensus). In both senses of the term, consensus implies that editors have raised any concerns they think are pertinent and weighed them in the light of the needs of the article and the policies and principles of the encyclopedia. Consensus is the natural process of healthy collaboration, and all editors are expected to make good-faith efforts to engage and respect it.

Consensus is not a numerical process; it is not the same as unanimity or majoritarianism or any other system of 'counting votes'. When things that look like votes happen on wikipedia (as in RfCs, AfD discussions, or other administrative actions), they are actually intended to solicit a broader range of discussion, not merely to tally up agreement. Such discussions are usually closed when they reach a saturation point in which most of the participants share a loose agreement about which policies and principles are pertinent and how they should be applied to the debate. The aim of consensus is to reach understanding, not agreement - it is expected that people can disagree with the outcome but accept it because they understand the principles in play.

Consensus is not always achievable for the same reasons that collaboration sometimes fails - strong ideological convictions, interpersonal frictions, or simple good-faith disagreement can frustrate the natural discussion process. Sometimes these discussions are left as no consensus, which ends the discussion without any action being taken, and sometimes they proceed on to dispute resolution mechanisms in order to seek a solution in a different way.

Ronz reverted with an unclear edit summary. Are there any actual problems with this revision (which I think sums up the concept nicely)? --Ludwigs2 16:14, 3 November 2011 (UTC)

"Actual problems" This is called "poisoning the well." It's deemed disruptive in consensus-building.

The question is, is this an overall improvement? I think not.

Better to keep the "What consensus is" introduction concise.

If details need to be added, do it elsewhere in the policy, but get consensus to do so. --Ronz (talk) 16:22, 3 November 2011 (UTC)

Looking at it briefly, Ludwigs' text seems to be an improvement - it actually attempts to explain things reasonably fully, which is what I would have thought a section titled "What consensus is" ought to be doing. It's not an introduction (that's what the lead is for), this is the heart of the policy, where tricky concepts (as this one undoubtedly is) need to be explained at whatever length is necessary. --Kotniski (talk) 17:48, 3 November 2011 (UTC)

Ronz: I just wanted to know what your 'actual' objections were. The current 'what consensus is' section is largely meaningless and even ungrammatical at points, so that objection can't really be correct. you still haven't said anything specific about the issues you see, and I can't really credit this as a meaningful objection until you do. But let's wait a day and let others chime in.

Kotniski: Ok. give it a more though review when you get a chance, and let us know what you think. --Ludwigs2 17:52, 3 November 2011 (UTC)

"meaningless?" "ungrammatical?" where?

"I can't really credit this as a meaningful objection until you do" Being dismissive of others is disruptive to consensus-building. --Ronz (talk) 18:42, 3 November 2011 (UTC)

I like this text in general. If there is a grammatical error or an unclarity, I do not see it.

I do disagree, however, with "The aim of consensus is to reach understanding, not agreement". No; if someone understands my arguments, and still thinks them evil and mistaken, he is not part of the consensus for them. The aim of consensus is acceptance: that everyone will tolerate the consensus position, without active efforts to overturn it. Septentrionalis PMAnderson 20:38, 3 November 2011 (UTC)

Good distinction. It's acceptance as opposed to agreement, and "understanding" is a very poor word to use because of its multiple meanings. --Ronz (talk) 22:26, 3 November 2011 (UTC)

Interesting. The idea behind that, really, was that the more someone understands your perspective, the less likely they are to think your perspective is stupid/evil. They may still think it's mistaken, but mistaken beliefs are grounds for conversation and compromise, rather than combat. In the philosophical presentation of consensus-decision making sources talk about creating the correct form of discussion, because setting up the right form is essential to productive consensus processes; but that form always involves trying to understand what your opposite is saying in his/her terms.

the theory behind consensus decision making is a little high-toned for use on project, but it would be nice to incorporate some of the principles. --Ludwigs2 23:01, 3 November 2011 (UTC)

Tout comprendre c'est tout pardonner is an apocryphal piece of late Enlightenment optimism; even Mme. de Staël, to whom it is attributed, never went quite that far. To see it after the twentieth century is astounding. Septentrionalis PMAnderson 00:24, 4 November 2011 (UTC)