Wikivoyage talk:Consensus

When do we use consensus?

We use consensus to determine pretty much everything on this site, and we have avoided majority-rule voting like the plague. But we have not actually noted this (as far as I can tell) in any policy. Would anyone object/add to this reworking of the intro:

Wikivoyage determines virtually everything by consensus. The one exception to this rule is on the votes for deletion page, where nominated articles are presumed guilty until proven (by consensus) innocent. No decisions are made on this site by majority-rule voting. Please remember that the result of any Wikivoyage article will be the consensus view of all the contributors to that article.

Does this look good? Are there any other exceptions in which we use something other than consensus? And should we mention this categorical use of consensus anywhere else? --(WT-en) Peter ^Talk 12:53, 18 September 2007 (EDT)

Alright, I decided to just plunge forward after months of no response. If anyone objects, please comment here. --(WT-en) Peter ^Talk 16:15, 18 January 2008 (EST)

How to build consensus

I've felt for some time that this article had not received enough attention. I've just added a substantial amount of content about (my understanding of) how our consensus practices work, mostly regarding how to build them. Does any of this strike anyone as controversial? --(WT-en) Peter ^Talk 16:15, 18 January 2008 (EST)

The nature of a consensus debate

In striving to reach a consensus on an issue each side of a debate has to either present a reasoned position, or express a reasoned objection to another position. Otherwise, we just get into 'yes' and 'no' style arguments, which are likely to ever progress to understanding and agreement. Is it ever reasonable to head off a consensus by disagreeing without presenting a reason for that disagreement, or a reason for an alternative position? --(WT-en) inas 17:43, 6 December 2009 (EST)

I don't think so. If someone agrees, we can assume that they support whatever the proposed argument was as it was written and for those reasons. If someone opposes, then no assumptions can be made as to why , and an explanation is needed. Nothing should be assumed to be self-evident. If the opposition says only "No", then I think it should be dismissed until it is elaborated upon. (WT-en) ChubbyWimbus 17:56, 6 December 2009 (EST)

Add: Communication of possible changes?

The nutshell for this policy at Wikivoyage:Policies contained "Consensus is difficult to achieve if there is inadequate communication; the community has a right to be informed of the possibility of a change that may affect them." However this is not actually contained in the policy. Should it be added? Nurg (talk) 01:38, 27 January 2013 (UTC)

Yes. This is one of those policies that has always been implicit but nobody thought to make explicit before. -- Alice^✉ 02:22, 27 January 2013 (UTC)

I also support this. Ikan Kekek (talk) 03:10, 27 January 2013 (UTC)

Communication is important. However, I'd hate to see people using this to challenge a successful consensus because they didn't know about it. I think people are obligated to have policy and article pages they care enough about on their watchlists. --Inas (talk) 06:21, 27 January 2013 (UTC)

The content that was removed from the nutshell was: "Consensus is difficult to achieve if there is inadequate communication; the community has a right to be informed of the possibility of a change that may affect them."

Do you think that statement needs beefing up (or watering down) and, if so, how? -- Alice^✉ 07:01, 27 January 2013 (UTC)

Question: I'm having difficulty tracing the history of this policy. A phrase similar to the excised phrase was present right from the first formulation visible here. Does anybody know of an earlier incidence or discussion, please? -- Alice^✉ 04:24, 27 January 2013 (UTC)

I think I may have found an earlier formulation. -- Alice^✉ 07:17, 27 January 2013 (UTC)

Inas, to your point, anybody always has the right to reopen a discussion by posting their opinion in a preexisting one. If they challenge a successful consensus, they can always be pointed to where the previous discussion took place. But I think what this discussion is about is that important policy discussions should probably have links posted in Requests for comment while they are under discussion, and changes in policy should be announced after the fact in the Travellers' Pub or perhaps some new page specifically dedicated to posting and linking to new policies. Ikan Kekek (talk) 09:32, 27 January 2013 (UTC)

Consensus: (from Merriam - Webster online)

a: general agreement : unanimity

b: the judgment arrived at by most of those concerned

Latin, from consentire (see consent)

Without communication there can be no consensus. It is possibly redundant to state that there is a need for communication to achieve consensus, but it can do no harm to make the point clear.

Per Ikan Kekek, a consensus is not necessarily final. Opinions change and circumstances change. An existing ruling based on a past consensus may not be valid at a later date.

A consensus of a small subset of a group is not legitimately the consensus of the group as a whole unless the group was informed of the debate.

If a consensus ruling is challenged, a significant proportion of the group must support the challenge for it to be valid. If more people challenge a "consensus" than those who supported it, it is clear that it may no longer be a consensus.

The matter of whether the group as a whole has the right to be informed of matters which may affect them is a separate issue. I am a proponent of transparency in these matters, and would like to see this right to be informed entrenched as a guiding principle, or failing that, a core community policy. Look at the actions of IB on Wikitravel to see what happens when the right to be informed is suppressed by persons outside of the community. • • • Peter (Southwood) ^(talk): 11:59, 27 January 2013 (UTC)

A very clear statement that I agree with 100%. Please edit our policies to make sure they reflect this. -- Alice^✉ 20:35, 27 January 2013 (UTC)

I'm happy with the communication part. But equal onus on all parties please. If you're interested in policy discussion, then add Wikivoyage:* to your watchlist. If you really care about business and touting, but not about whether userboxes are used or not, then edit your watchlist appropriately. After a group of people have worked to build a common understanding and position on the appropriate discussion pages, don't come back six months later and reject that consensus building, because they didn't specifically seek out your position. --Inas (talk) 04:49, 29 January 2013 (UTC)

That seems a very fair and valid point to make, Inas. But can it be squeezed into the nutshell or should it be in the body of the policy article?

Suggestions of concrete phraseology for both below, please. -- Alice^✉ 05:07, 29 January 2013 (UTC)

I'd say - Most decisions at Wikivoyage are made by consensus, not by majority vote. Finding a consensus can be hard work and should be approached with open communication, developed arguments, and a willingness to understand the various points of view --Inas (talk) 05:16, 29 January 2013 (UTC)

Excellent. Now was that your concrete suggestion for the nutshell for this Consensus policy and visible at Wikivoyage:Policies, or was it your suggestion as a replacement for our current lead paragraph here: Wikivoyage:Consensus, or both ? -- Alice^✉ 05:23, 29 January 2013 (UTC)

My concern is that we never have a situation when a notification of a major issue such as the migration to WMF can be suppressed by anyone, as was done by IB back on WV. I do not suggest that every user should be personally notified of every consensus discussion, just that a user in good standing may notify other users by means of legitimate media, without undue interference or censorship, provided that the notification message is non-partisan. • • • Peter (Southwood) ^(talk): 07:05, 31 January 2013 (UTC)

Bugzilla coordination

Swept in from the pub

I've noticed in two Bugzilla threads yesterday that our methods of consensus are a little opaque to the tech folks—compared to obvious lists of support. To me, the discussion above that we need to change $wgUploadNavigationUrl to Wikivoyage:Upload file demonstrates a clear consensus, given high visibility, the build to that decision, and the fact that no one raised any objections. But in Bugzilla:44572, TTO was asked why the discussion involved just one other person.

Maybe a Bugzilla or Wikivoyage:Tech requests page would make sense here? We could make an effort to add support lists to discussions, after reaching consensus? That would allow everyone to keep better tabs on developments on Bugzilla, and to be involved in that process. It's extra red tape, but it might help things get done more quickly. --Peter ^Talk 18:34, 3 February 2013 (UTC)

If I understand what you're getting at here, I think I agree. I think that in principle, the consensus model we've always had is the best way to address issues. But I also concur that we should definitely explore ways to make it more obvious to the casual reader where the community (and each individual member) stands on any given issue, without having to wade through reams of text. -- AndreCarrotflower (talk) 19:33, 3 February 2013 (UTC)

To be sure, the decisions should be made by consensus as always. The support lists would be "for show." --Peter ^Talk 20:01, 3 February 2013 (UTC)

Suggestion? Perhaps, rather than going that route, you could just "close" discussions with a note at the top saying "This discussion is closed and the request has been submitted to engineering. The result of the discussion was...." That would probably satisfy their need for an "at a glance" without going through a support list that's just for show. Philippe Beaudette, Wikimedia Foundation (talk) 16:19, 4 February 2013 (UTC)

That sounds easier. --Peter ^Talk 06:22, 5 February 2013 (UTC)

The difference, I think, is that discussions are almost never closed here. They're usually dormant for a couple of years before being moved into an archive. Globe-trotter (talk) 09:34, 5 February 2013 (UTC)

We could just put a disclaimerbox at the top of the discussion with the message "We have reached a consensus for this change, and a request has been submitted to engineering." --Peter ^Talk 18:12, 5 February 2013 (UTC)

That sounds good. I don't really see the need to "close" discussions, but I understand why the engineering folk wish to see that their actions are reflecting the community consensus. --Inas (talk) 22:11, 5 February 2013 (UTC)

I don't even think a disclaimerbox/template is necessary. Most wikis usually just end it with an admin posting something like "Closed - The consensus was found to be... . The following action will be taken... . A request has been submitted to engineering here. JamesA ^>talk 11:01, 6 February 2013 (UTC)

Yeah, but it is the "Closed" part that is the problem. We don't close and archive discussions, and often revisit discussions years later when we realise we could have made a better decision. --Inas (talk) 00:23, 7 February 2013 (UTC)

Consensus

We use consensus within the rest of the WM movement. As shown by [1] however it does not require everyone to agree. Travel Doc James (talk · contribs · email) 07:24, 4 October 2013 (UTC)

Whenever possible a compromise that is acceptable by all parties is preferable, When there is a deadlock there should be a way of going forward. The way of breaking the deadlock will work best if previously agreed (by consensus?). How about a two thirds majority of enfranchised voters? That would obviously require a franchise qualification, which could also be contentious. I suggest any user who would qualify for nomination as an admin, just as a starting point. • • • Peter (Southwood) ^(talk): 11:13, 4 October 2013 (UTC)

Yes sounds reasonable. Travel Doc James (talk · contribs · email) 16:07, 4 October 2013 (UTC)

We need to recognise spiraling conversations as deadlocks where there are continuous attempts to try and convince the last one or two people even though unanimity is not required, especially when it's a matter of who can last the longest. It could simply be a difference of opinion which cannot be resolved. If you find more than three back-and-forths with no forward movement, then disengage. This applies to policy and regions discussions. I am undecided whether polls will help, but it may at least diminish some sore feelings (and probably bring up other new ones). -- torty3 (talk) 03:29, 6 October 2013 (UTC)

I think our consensus policy does guide us here. In order for there to be consensus every argument has to be met. So, it's insufficient to just be a hold-out, saying you don't like a proposal without argument. And its insufficient to just repeat your proposal without addressing the rebuttal arguments. If there remain good arguments against a proposal that haven't been addressed, then we haven't got a consensus.

Once we start doing two thirds votes, then we really have to start worrying about stacking. With some discussions only eliciting comment by two or three people here, a vote of two thirds may just be two. --Inas (talk) 03:40, 6 October 2013 (UTC)

Theoretically yes, if there is common sense to recognise there is or is not consensus, but in practice the arguments just keep going on and on in circles. -- torty3 (talk) 03:49, 6 October 2013 (UTC)

And that's when we need to ask someone else to read through and determine what the consensus is, if there is any. --Rschen7754 07:34, 6 October 2013 (UTC)

In reality, someone (preferably an admin, but not always) not involved in the discussion should be closing it, and they should have the ability to use discretion in the event of votestacking. --Rschen7754 06:42, 6 October 2013 (UTC)

I'm a little confused as to how this part of the thread arose and what it is trying to resolve. Texugo (talk) 06:45, 6 October 2013 (UTC)

Me too, but I think it is a good discussion, and the one that we need to have first. --Rschen7754 07:36, 6 October 2013 (UTC)

To add to my past thoughts on the matter, I would encourage all of you to take some time and explore other Wikimedia projects over the next few weeks or so to see how they do things. While the English Wikipedia does get a lot of things done, I would hate for us to blindly import everything from there, because I would hate to blindly import their systemic problems too. --Rschen7754 06:50, 6 October 2013 (UTC)

Another thought: I think that the fundamental issue that we need to resolve is what consensus is. I think that from that, everything else will follow; once we know how we will make decisions, that will greatly influence the topics above. --Rschen7754 07:36, 6 October 2013 (UTC)

Should this discussion be part of the discussion on How to deal with unwanted edits, which is where it started from, and now moved to Wikivoyage_talk:How_to_handle_unwanted_edits#Proposal:_Revise_policy_on_dealing_with_repeated_unwanted_edits or separately, on Wikivoyage Talk:Consensus? (I think Consensus is the proper place) • • • Peter (Southwood) ^(talk): 08:40, 7 October 2013 (UTC)

Will it be noticed on those pages, and have adequate participation? --Rschen7754 17:23, 7 October 2013 (UTC)

This has often been a problem on Wikivoyage. Can you suggest a better way of doing it? • • • Peter (Southwood) ^(talk): 16:25, 8 October 2013 (UTC)

Personally, I think stuff gets lost in the Travellers' pub, and when it is all archived in separate places, it makes things more difficult to find. I would split it into two to three pages, so that edits to sections are more likely to be noticed, and have an automatic archiving bot so that a) we don't spend our valuable time archiving the pub, and b) so that older threads get pushed off the page once they have been inactive for a week or two (or however much time is appropriate). One of these would be a noticeboard for administrators; all the background discussion that caused the kerfluffle of last week should be taking place onwiki anyway, and it provides a place for other Wikimedia users to contact admins.

In short: the infrastructure should be built to serve the needs of the community, not the other way around. We should be spending our time writing a travel guide, not fighting an infrastructure dating back from the Wikitravel days (as early as 2004!) that does not exactly fit our needs today. --Rschen7754 08:50, 9 October 2013 (UTC)

Consensus and decision-making

Swept in from the pub

Regarding the larger issue of trolling, it's clear as crystal to me that what we're witnessing is the ultimate failure of the consensus model of decision-making. Frankly speaking, even dating back to the much simpler days under Evan, in retrospect it was fairly naïve of us to think that setup would remain stable on an indefinite basis. It's way too exploitable by intransigent minorities who, even when we reinterpret "consensus" to mean something different than "unanimity", will still bog down discussions with their relentless unwillingness to compromise, and make the crafting of policy such an unpleasant and seemingly fruitless pursuit that those few remaining people who still bother to participate will inevitably throw up their hands in frustration.

Lest some believe otherwise, by "intransigent minorities" I'm referring to all of us. No one has clean hands, including me - we've all played that role from time to time.

The challenge before us now, as I see it, is a Herculean one. For the sake of Wikivoyage's future - to prevent some lone-wolf admin from taking matters into his own hands, banning transgressors without due process (I admit the thought has crossed my mind a time or two) and generally causing mayhem; or even worse, to prevent the entire community from falling apart in a toxic swill of mutual acrimony - we have to abandon consensus and come up with a new model of decision-making, which will be more streamlined, more structured, less gameable, and, in general, more realistic from a practical standpoint. Not only that, but we have to design and implement this new system using the consensus model that we're stuck with in the interim.

-- AndreCarrotflower (talk) 23:09, 16 November 2013 (UTC)

"we have to abandon consensus and come up with a new model of decision-making, which will be more streamlined, more structured, less gameable, and, in general, more realistic from a practical standpoint." If the consensus model works on other Wikimedia sites (which deal with far more issues than this site does in comparison), then why can't it work here? --Rschen7754 23:11, 16 November 2013 (UTC)

As far as the hacking issue, I'd have to say it's a conspiracy theory; only hashes of the passwords were revealed, not the actual passwords themselves. --Rschen7754 23:32, 16 November 2013 (UTC)

Agree with Rschen on both points above. I think the consensus model works, but that our implementation here needs a few tweaks to deal with the (very common) situation in which a discussion wanders all over the place, ends up with no one sure what's being proposed anymore and what everyone's position is, and results in the issue being dropped out of frustration. Developing and implementing some more structure around achieving consensus would resolve that (see w:WP:CLOSE for some ideas that might be relevant here), but given how many things are currently outstanding (user ban procedure, main page, etc, etc) it seems like it might make sense to let things calm a bit so that we can focus on what is likely to be a very important change to the culture around here. -- Ryan • (talk) • 23:49, 16 November 2013 (UTC)

I don't think we can compare the WP consensus model and ours. We use the same word, but it means something quite different in practice. We have multiple layers of consensus in the space of protecting the wiki and its users by combating disruption. We have to have consensus on the policy, and then consensus on the implementation. And consensus more often than not equates to unanimity.

We obviously have a wide range of views at the degree of tolerance that should be down to people who choose to not participate in wiki norms, and can be hugely time consuming at the expense of the guide. I'm not sure that problem is going to go away by rewriting our policies in this area, because the same people are going to be annoyed when the first user gets banned that they would have tolerated. It's difficult, but we have to address it. --Inas (talk) 04:57, 17 November 2013 (UTC)

Consensus is a Wikimedia principle, however. We cannot throw out the consensus process entirely, because we are on Wikimedia. --Rschen7754 05:05, 17 November 2013 (UTC)

I'm not wanting to get into the semantics too much, but it should be clear that WP interprets consensus quite differently to what we do here, and that there is a fair bit of scope for individuals to act without consensus on WP as long as they are acting in line with a broadly supported policy. I'd be surprised if we ever go further than that. --Inas (talk) 05:10, 17 November 2013 (UTC)

Well, m:Consensus has a bit on how Wikimedia interprets consensus... --Rschen7754 05:12, 17 November 2013 (UTC)

Digging around for another project, I also found b:en:Wikibooks:Decision making. --Rschen7754 05:20, 17 November 2013 (UTC)

I'm afraid my cynicism remains unchecked. Yes, you pick a word that in most contexts means universal agreement and adapt it to wiki. Then, when you realise that won't really work on any wiki, you set about trying to redefine the word, rather than redefining the concept of what is being achieved.

In any event, this won't be limiting. We're not going stray further than WP has. --Inas (talk) 05:24, 17 November 2013 (UTC)

With respect for all, I nevertheless think that arguing over the definition of "consensus" is a waste of time, except that if it means "unanimity" to anyone, that is untenable. Requiring unanimity is what destroyed the League of Nations and led to World War II - not an example we can afford to follow. Ikan Kekek (talk) 06:12, 17 November 2013 (UTC)

Well, if we can't decide when consensus has been reached, having any further discussions on anything is pointless. --Rschen7754 06:25, 17 November 2013 (UTC)

I guess you're right. I just hate this type of disagreement about definitions, though. Ikan Kekek (talk) 08:16, 17 November 2013 (UTC)

(reindent) With respect to Ikan, Inas is absolutely correct: the lack of clarity of the meaning of "consensus" as employed by our community is fundamental both to the argument I'm trying to make and the problems we have with decision-making. Though Wikivoyage is the only wiki on which I'm an administrator or otherwise extremely active, I'm vaguely familiar with the decision-making model that Wikipedia uses, and the mistake I made in my original post that was corrected by Rschen was that because it is different than ours - more structured and efficient - it doesn't qualify as "consensus".

So, in reading my original post, phrases like "we have to abandon consensus" should be read as "we have to abandon 'consensus' as it's been defined on Wikivoyage thus far". But I stand by the main thrust of my argument, and given what Rschen said about consensus being the standard model that all WMF sites use for decision-making, the "Herculean challenge before us now" is reframed as a question of how to redefine consensus as used on Wikivoyage.

As Inas said, it's absolutely vitally necessary that we have this discussion even though it's no fun. It's frankly possible that the changes we'll have to make will be disagreeable enough to some of our current editors that they'll follow Peter, Jan and Alexander out the door. But at least we'll have to show for our efforts a functional, if smaller, community that can deal with trolling in a meaningful way. Burying our head in the sand, and avoiding solving this problem because it's no fun and gives us a headache, will probably continue to drive away our contributors in frustration, only with no upside.

-- AndreCarrotflower (talk) 15:53, 17 November 2013 (UTC)

I think a lot of it is that the community has changed, both in terms of size and in terms of viewpoints represented. What may have worked in the Wikitravel days may not work so well now. --Rschen7754 19:47, 17 November 2013 (UTC)

Once we determine what the ultimate question is, then we can determine what the ultimate answer means :-) In my opinion, our definition of consensus is at least as satisfactory as the floundering definition in terms of what it is not at meta. My view of how we define it here, is that every well formed argument in opposition has been satisfactorily addressed with a supported argument. That means that the Support, Oppose lists we see on WP, and that a couple of users have tried to introduce here are completely meaningless. It means that ten people saying we can't change this because this is the way we have always done, or it may lead us down a bad path doesn't stop a consensus. However, one person with strong reasoning that hasn't been countered can. This is how our consensus differs from WP, which really does turn into a vote very quickly at times, with its support/oppose lists. --Inas (talk) 22:03, 17 November 2013 (UTC)

But who determines what is "strong reasoning" when both sides are completely entrenched in the matter? And en.wikipedia does not do most things by voting, even though it may look like it on the surface; someone uninvolved does read through and evaluate the consensus. It usually aligns with the result of the "vote", but not always. --Rschen7754 22:15, 17 November 2013 (UTC)

As a counter-point to arguments against "support" / "oppose" votes, I think our VFD page is where we most often, and most easily, come to an actionable consensus, and a process similar to that one (which is not a majority-rules vote) would be valuable in cases where discussions deadlock. We've all been a part of a discussion where we feel our valid points aren't being addressed, and also been a part of a discussion where someone repeatedly claims they are being ignored while we feel that their points have been addressed appropriately, and the vast majority of the time these discussions result in no action and generate frustration all around. In those cases I think a "support" / "oppose" system similiar to Wikipedia's would be extremely valuable so that we can actually come to a clear resolution, rather than the status quo in which no one is clear whether it is OK to take action or not. -- Ryan • (talk) • 22:23, 17 November 2013 (UTC)

BTW, I'm not trying to say which way is best, just to give us a definition of what we have, because using the word consensus doesn't define it. Because, reading through 1000 votes, and deciding an outcome based, in theory, on arguments but in nearly every case aligning with a majority vote, isn't a consensus by any reasonable definition of one. Its a semantic fiction - like the revealing of the common law. It's a mythology that's evolved to sound inclusive, when it is actually clear that when there is disagreement based on well reasoned arguments, not all outcomes can win favour. --Inas (talk) 22:32, 17 November 2013 (UTC)

Wikivoyage:Consensus/Draft

There has been discussion in a number of places lately about our consensus policy that seem to fall into three categories:

1. Misunderstandings that equate consensus with unanimity (example).
2. Frustration that so many of our discussions end with no clear consensus and thus no action (example).
3. Inability to determine if/when consensus has been achieved (example).

I've started a draft that makes several updates to the current policy at Wikivoyage:Consensus/Draft. The current Wikivoyage:Consensus text is still present, but I've made additions and re-organizations that can hopefully serve as a starting point to address some of the above issues. The changes can be broken down into three groups, each of which could probably be implemented independently if desired:

1. A new "What consensus is not" section, making clear that consensus is not unanimity.
2. A new "Determining consensus" section that proposes having an admin (preferably one who wasn't involved in the discussion) declare the outcome of a consensus-building discussion when the result isn't clear, much in the same way an admin reviews a VFD nomination and then closes it.
3. Probably the most likely part to be controversial, a proposal to create more formal discussions (similar to what we use for VFD) in cases where a free-form discussion fails to result in a clear consensus. A structured discussion would only be necessary in cases where the free-form discussion does not result in a clear consensus, in which case moving to a more structured discussion would allow us to find some resolution using the same process that has always worked for us with VFDs.

The first bullet point is more of a clarification than a change and is something that we might be able to agree on quickly, but the other two bullet points would change how we've done things and will probably be more controversial. I think it would be a change for the better, but hopefully others can suggest improvements or alternatives that will address some of the frustration that comes from having so many discussions in limbo - personally I find it hugely frustrating to have discussions side-tracked or end up unclear as to the level of support, and would like to see us put a process in place to allow a definitive conclusion. -- Ryan • (talk) • 02:25, 13 January 2014 (UTC)

I appreciate your work and like it, but I also think that it should be made explicit that a lack of change due to a lack of consensus to change is a perfectly legitimate outcome, and, thus, that the declared outcome may not be conclusive for all time. Ikan Kekek (talk) 02:34, 13 January 2014 (UTC)

The third point may be controversial, but it is well-practiced elsewhere (see point #6 on m:Founding principles). Agree with the above. --Rschen7754 02:39, 13 January 2014 (UTC)

I really like the changes you've made, Ryan; this definitely points us in the right direction. Your 3rd bullet may be controversial to some, but I do believe it will be necessary (and can think of past discussions where it would have been) in some instances. I may try and have a play with Plunge Forward as well, just to clear up the degree to which we're happy for people to plunge. --Nick ^talk 02:45, 13 January 2014 (UTC)

Ikan and Rschen: I'm not entirely sure I understand what is required - Wikivoyage:Consensus/Draft#Determining consensus notes Results of the consensus-building discussion may be to implement the proposal, not to implement the proposal, or if support is split it may be "no consensus". Could one (or both) of you update the draft in a way that would address your concerns? -- Ryan • (talk) • 04:38, 13 January 2014 (UTC)

I personally have no issues with it. --Rschen7754 05:02, 13 January 2014 (UTC)

Sorry, I mis-read "agree with the above" as seconding Ikan's concerns. -- Ryan • (talk) • 05:09, 13 January 2014 (UTC)

I encourage attempts to clarify this area of policy, but perhaps specific examples of discussions that have been left unjustifiably unresolved would help highlight the need for some of the more radical provisions of the proposal. I remain unclear on what the point of bolded votes (e.g., Support, Oppose) is when the text explicitly says it's not the number of votes that matters but the quality of arguments. If we truly want to focus on the quality, then prohibiting simple bolded recommendations seems like it would go farther toward facilitating a resolution. My other concern is with assigning admins the task of determining consensus. It hasn't worked all that well at VfD and I'm reluctant to expand that role's remit when "ability to determine consensus" has not been commonly used as a criterion in admin nominations. Powers ^(talk) 18:39, 13 January 2014 (UTC)

A few examples of discussions that have gone on and on without reaching consensus that I remember:

• Wikivoyage_talk:Travellers'_pub#Title_case_.22Travellers.27_Pub.22
• Talk:United_States_of_America#Page_banner
• Wikivoyage talk:Image policy - Several on here
• Wikivoyage_talk:Dynamic_maps_Expedition#Are_we_ready_to_start_deploying_dynamic_maps_across_the_site.3F

Unfortunately, in many cases, the current form of 'consensus' is not consensual at all - many discussions become simple to and fro arguments where consensus is decided simply by who gives up first. I believe that using a simple Support and Oppose could be very useful indeed for cutting through some of the trickiest matters that we discuss on the site when normal, more verbose, arguments have been exhausted. Whilst using admins to determine consensus may be slightly controversial, I think it's preferable to creating a "Consulate" (read: ArbCom) and creating another layer of bureaucracy that WV does not require and cannot maintain at present. We could always have system of appeals where another admin or trusted editor could be consulted if necessary? --Nick ^talk 19:00, 13 January 2014 (UTC)

Ryan, I'm sorry I don't have time to update the draft now, but I do support the thrust of what you and Nick are saying, so carry on. Any changed language that gets consensus support is very unlikely to be deeply troubling to me. Ikan Kekek (talk) 01:05, 14 January 2014 (UTC)

I think considerably more discussion is needed on points #2 and #3 above since they represent a change to how we've done things for more than ten years, but as #1 is a clarification rather than a change ("what consensus is / is not"), would there be any objection to implementing that now and continuing discussion about the other points? -- Ryan • (talk) • 22:40, 14 January 2014 (UTC)

I'd be happy to see that implemented now, yes. --Nick ^talk 22:44, 14 January 2014 (UTC)

Most definitely; implement it now. Ikan Kekek (talk) 04:07, 15 January 2014 (UTC)

Otherwise, we might not have consensus on part 2 and 3. :P --Rschen7754 04:09, 15 January 2014 (UTC)

I support all 3 points, but the first indeed seems uncontroversial. Let's get that one out of the way first. JuliasTravels (talk) 11:19, 15 January 2014 (UTC)

I've updated the policy page per point #1. It would be good to better understand the concerns of Ikan and Powers, and hopefully others can participate in this discussion, so that we can find a way forward on the other points. It might be useful to actually try to put this new process into practice in a couple of discussions to see how it would work out, which would give us a better idea of what tweaks might be needed - Talk:United States of America#Page banner might be a good test case, and I'm sure others are involved in discussions that seem to be languishing and might be good candidates. Thoughts? -- Ryan • (talk) • 03:52, 16 January 2014 (UTC)

I would be happy to clarify my concerns if you could clarify what isn't clear. =) Powers ^(talk) 15:54, 16 January 2014 (UTC)

I'm not sure I still have concerns about the draft, since it was explained above. Ikan Kekek (talk) 01:53, 17 January 2014 (UTC)

Powers - Nick responded above to your original request for links to discussions that have languished, and responded re: explicit statements of support or opposition, and I basically agree with Nick. To cite a specific example of why clear indications of opinion would be helpful, I read your comment here as supporting the new banner, and only realized you opposed it later; a clearer indication of opinion would have resolved that misunderstanding immediately. Re: your statement that having admins declare consensus hasn't worked on VFDs, I would argue that the fact that VFDs always eventually get resolved, and usually do so in a very timely manner, is a strong argument that the process DOES work. That said, while I think that we have a broad agreement that our status quo for determining consensus needs updating, this is just one proposal for doing so, so if there are alternative ideas that would address points #2 and #3 above we should make sure they are discussed. -- Ryan • (talk) • 05:37, 17 January 2014 (UTC)

I would agree with Powers that we don't want empty votes, but I think the primary point Ryan is making is rather for argumentation to be more clear. Would it help to explicitly include some kind of requirement for explanation/rationale with any opinion? That way the discussion is structured enough for clarity and would help if all else fails to move further along a less consensual road, but would still allow to weigh arguments? JuliasTravels (talk) 10:30, 17 January 2014 (UTC)

As somebody who does not have time to follow all the long discussions, I think that Point 3 is an important consideration. It is much easier to quickly get the general feel for how the discussion is going if people either use Support and Oppose possibly optionally qualified with weakly / strongly or the for and against arguments are put in separate sub-headings. If I am thinking of supporting a change, I like to be able to quickly identify the main arguments against.

When the discussion veers off a tangent, these related discussions must be quickly transferred to a separate sub-section. For instance in the audio files discussion the initial proposal was for phrasebooks and article titles. The discussion then veered off to itineraries and nothing was decided. AlasdairW (talk) 14:53, 17 January 2014 (UTC)

If I may, Ryan, that is indeed a perfect example. Boiling down my comments in that discussion to "Support" or "Oppose" oversimplifies my opinion. I apologize if I wasn't clear in my initial comment, but I think my statements convey a nuance of opinion that "Support" and "Oppose" cannot. Powers ^(talk) 18:16, 20 January 2014 (UTC)

It is helpful for a good glance, but whoever closes a discussion should be reading through its entirety anyway... --Rschen7754 03:23, 29 January 2014 (UTC)

(edit conflict - summary of the following is basically "agree with Rschen7754") LtPowers - if we were using a "support" and "oppose" system and you commented without stating either it would at least be clear that you neither supported nor opposed the banner. Similarly, if you actually do oppose the change, an "oppose" comment would make clear that you had concerns that you wanted to see addressed. In either case, you would still be free to convey the nuance of your opinion, but it would be much clearer to others what that opinion was (as I noted earlier, I misunderstood your initial comment completely). -- Ryan • (talk) • 03:32, 29 January 2014 (UTC)

Ironically this discussion, which was at least partly inspired as a potential solution to the problem of so many discussions dying without resolution, has been mostly idle. As the person who started this thread the onus is on me to ensure it moves forward, but at the moment a) while we seem to have some level of agreement, I'm not quite sure how best to modify the existing proposal in a way that would address concerns raised and allow something to be implemented, b) I'm hoping that there will be some additional feedback and alternative suggestions to address points #2 and #3 from the start of this thread, c) I'm holding out some hope that this will be one of those discussions that goes idle for a while and then is later revisited when the topic at hand becomes more urgent. I'd very much like to see some movement to make consensus building a less painful process, so hopefully either others can jump in to help re-invigorate this process, or else I'll do my best to revive it once I've got some tangible ideas on how we might move forward. -- Ryan • (talk) • 03:22, 29 January 2014 (UTC)

Perhaps we need to take a look at why consensus-building has become so "painful". Certainly when I first joined WT, most issues that were raised lacked much disagreement. We all seemed to be largely on the same page, so consensus was easy to determine. Why has that changed? If we can get back to that kind of state, the need for formalized voting procedures (which have traditionally been strongly disfavored here) might become moot.

Absent that, I would be willing to support the option to go to a more formalized discussion structure (as we did in the case of changing "Get out" to "Go next") as long as it's made clear that it's a last resort, only in the most urgent and intractable cases. I'm not sure I'd support requiring an admin to close a discussion, as we have not been choosing our admins based on their abilities to determine consensus. If the result of the discussion isn't clear to most of the participants, then it should be a "no consensus" result, not left to the whim of a random admin.

However, I do sense a slight bias in the entire concept behind this proposal. The proposal is predicated on the idea that "things don't get done" is bad (e.g.: "Frustration that so many of our discussions end with no clear consensus and thus no action"). Sometimes "no action" is good, or at least neutral, and it's a valid result. I am concerned that implementing new procedures that make it easier for idea-proposers to get their ideas implemented will bias the site toward change and away from tradition, rather than striking a healthy balance between the two.

Of course, stagnation is also bad. But I do not believe we are in a state of stagnation; the changes in the last 15 months have been significant and wide-ranging. -- Powers ^(talk) 15:06, 29 January 2014 (UTC)

"If we can get back to that kind of state, the need for formalized voting procedures (which have traditionally been strongly disfavored here) might become moot." reads to me as "If we can get back to the state of always keeping the Wikitravel status quo and never changing anything, the need for formalized voting procedures will become moot." Seems that anyone who is trying to suggest anything else is no longer welcome to contribute to discussions on this site. --Rschen7754 19:03, 29 January 2014 (UTC)

Powers - re: why consensus building is so painful, my opinion is that it was far easier to reach a clear consensus (either support or oppose) when there were only a handful of participants in most discussions. In those smaller discussions there was more likely to be unanimity, and when there wasn't the people in the minority were usually willing to yield when no compromise was found and they didn't feel too strongly on a subject. Additionally, in cases where the consensus was unclear Evan usually played the benevolent dictator role and made a call one way or another. Today a big decision will involve dozens of people offering differing opinions, discussions easily become side-tracked, and in the end no one feels empowered to declare a consensus unless there is near-unanimity (and consensus has never meant unanimity here).

Regarding this proposal biasing us towards making more changes, that might happen, but to me the primary goal should be not to make change happen more quickly, but to create a structure in which we can reach an actionable result in most discussions, even if that result is "do nothing", rather than the current frustrating process of never-ending discussion in all cases that aren't crystal clear one way or the other - the goal is to allow us to get to a "go/no-go" decision instead of a state in which discussion continues forever until everyone is convinced to change their opinion or simply yields out of frustration. Using a more structured discussion format in cases where an open-ended discussion failed to result in a clear consensus and empowering admins to judge the result of that structured discussion (just as we do with VFDs) is one possible solution, but it would also be great to hear any other suggestions that would allow us to come to a resolution to non-unanimous discussions. -- Ryan • (talk) • 19:36, 29 January 2014 (UTC)

Rschen, that's bordering on a willful misinterpretation of my words. The state I was referring to was one where everyone in the community seemed to be on the same page, whether that involved necessary changes, or useful innovations, or maintaining the status quo. There was nothing in that statement you quoted that could remotely be implied to advocate stagnation, not without deliberately ignoring the actual history of this community.

Ryan, perhaps it wasn't clear, but I did attempt to describe a modification to your proposal that might find a stronger consensus, but you seem to have only repeated your call for options different from your proposal. Are you looking for either outright acclamation or rejection, rather than modification? Powers ^(talk) 01:57, 30 January 2014 (UTC)

Sorry if I'm being dense, but you said "Certainly when I first joined WT, most issues that were raised lacked much disagreement... If we can get back to that kind of state, the need for formalized voting procedures... might become moot." You then said you could support the current proposal only if it was used as a last resort. Were you not asking for other options that would take us back to a state more like what existed in the early days of WT? -- Ryan • (talk) • 03:18, 30 January 2014 (UTC)

I was suggesting that perhaps an analysis of what had changed would produce a better idea of how to change our consensus policy. But I also suggested that your third provision might be acceptable if we were careful to restrict its use, which I thought seemed like a good compromise. Maybe I'm missing something. Powers ^(talk) 01:48, 31 January 2014 (UTC)

I'm sorry, but I'm not going to keep quiet with my dissenting opinions just so that we can "be largely on the same page" and "can get back to that kind of state" so that "the need for formalized voting procedures... might become moot". I don't think there is another way to interpret that; you want everyone to have the same opinion, wondering "Why has that changed?" and from other comments I've seen, you are "concerned that implementing new procedures that make it easier for idea-proposers to get their ideas implemented will bias the site toward change and away from tradition." --Rschen7754 03:28, 30 January 2014 (UTC)

Do you disagree that we should strike a balance between change and stability? Powers ^(talk) 01:48, 31 January 2014 (UTC)

I do not disagree, but I feel that you are pushing for more than a "balance". --Rschen7754 04:54, 31 January 2014 (UTC)

I think actually bolding opposition could work to protect those who don't support a change, because it disallows their concerns from being swept under the rug, ignored, intentionally or unintentionally misconstrued, or any other tactics that can be used to deal with users that are "difficult" for someone trying to push an agenda/idea. I believe somewhere above it was stated early on to be used as a last resort rather than to simply jump right into a Support/Oppose vote system. I think that is the best sollution. ChubbyWimbus (talk) 03:49, 31 January 2014 (UTC)

If we accept that "consensus is not unanimity" then it's possible to have an outcome of "we plan to do/not do X because of Y, although some of us have reservations due to Z" - much like Supreme Court decisions (where unanimity is rare) often have multiple dissenting viewpoints placed on record with the full legal reasoning for each by individual judges.

Certainly, we do sometimes deadlock on a question (this fine mess, for instance) but we do need to save the no consensus close for cases where there is equal or comparable support for two mutually-exclusive options. Requiring near unanimity is merely deadlocking everything except the most non-controversial proposals, which is infinitely far from "a balance between change and stability".

As for the "we never disagreed on WT in the good old days" bit? Some of us came here from WP, others from WT. Each has its own utterly-incompatible ways. Any success we may have in attracting people from other WMF projects, our most likely source of new users at this point, is only going to make the "but that's the way we've always done it" at (whatever project) that much more of an inevitable collision. K7L (talk) 15:19, 2 February 2014 (UTC)

Yes, and I've certainly said my piece to other English Wikipedians at w:en:User:Rschen7754/You represent the English Wikipedia! However, there's a certain minimum that all Wikimedia sites are expected to adhere to, and that is where most of my concerns have arisen. --Rschen7754 21:16, 2 February 2014 (UTC)

The WT community was small, so admittedly there was a lot of practice that was not policy and some of that was intentional in order to encourage a case-by-case approach, but with a larger, growing, and more diverse community, it's very difficult to maintain. Although I have admittedly sometimes rolled my eyes at long discussions here about non-written rules that were customary, based on "common sense" and never questioned on WT, it is in fact more of a burden to give a "this is the way we do it" argument with only common practice to back us than it is to say "This is policy" in the long run (unfortunately holes in policy are great ways for trouble-makers to stir things up, as well, in a community run by common practice). Also, I think for anyone frustrated with what may seem like an obsession here with making everything into policy, I do think it's worth mentioning that we DID in fact have a TON of these same types of discussions on WT. While there may have been more unspoken understandings there, discussions there were certainly not always pleasant, people there too felt their views were being ignored at times (myself included), and I think these situations occurred more frequently than people remember. Plus, a larger number of policy discussions should be seen as somewhat inevitable with a new site, and the frustration that these discussions bring is something everyone feels at times, regardless of their origins. Hopefully it is evident that regardless of how anyone feels about the former WT community vs our WV community, everyone is still acting in "good faith" (or almost everyone).

In response to the specific topic at hand, I don't think we can quantify consensus nor do I wish to. It is always a matter of weighing arguments and making sure the topic has been given due consideration. If a group of people support a proposal, it should not be considered "consensus" unless and until the concerns of the opposition have been addressed. The legitimacy of the concerns also factor in, but if they are legitimate, we should never say "Oh well" and plunge forward before addressing them, no matter how much support something has. Likewise, unexplained opposition carries no weight in consensus. I applaud cases I have seen where a discussion participant notifies other users who may not have seen a discussion but are assumed to have concerns especially when the notifier is in favor. It would be too easy to quietly gain a consensus and stick one's tongue out at known opposers for not noticing. ChubbyWimbus (talk) 07:25, 3 February 2014 (UTC)

Well, but we can't make everyone happy. --Rschen7754 09:39, 3 February 2014 (UTC)

We never do! But we should give everyone the courtesy of having their concerns addressed if they are serious and sensible. There can still be consensus to plunge forward after addressing the issue, even if the opposition doesn't like it, but failing to address people's concerns or opposition negates the entire purpose of having a discussion and is not the kind of consensus we should be striving for. ChubbyWimbus (talk) 10:24, 3 February 2014 (UTC)

I actually think that in most discussions, we're all not thát far from the same page really. We all stand behind the basic goals. It was easy to agree on skeleton policies: plain vandalism is a blockable offence, the traveller comes first, we like consensus as a basic rule. The further we work things out, the more we get into details about what's the best way to get to that goal, and opinions will vary more. Sometimes I get the feeling we're reading things a bit too black and white - in both directions. I'm not reading that Powers is rallying for stagnation, or for unanimous consensus before any change. It seems fair enough to implement the idea of moving to a more formal discussion and (yet a step further) calling an outcome as a last resort, and keeping free-form discussion as the first and preferred way. I too like a better overview of arguments for and against though, and the bolded supports and opposes might help. I see no harm in allowing people to do that also in free-form discussions, as long as it's plain and clear that they are not votes, just clearly stated opinions and at that point no part of a more formal discussion yet. JuliasTravels (talk) 10:52, 3 February 2014 (UTC)