Wikipedia talk:Consensus

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

Consensus is commonly formed by editing pages on the wiki.

[1] Both charts are functionally the same, so it shouldn't matter which one is used. --NewbyG (talk) 01:47, 8 May 2008 (UTC)

I don't agree. The more recent chart has gone through many compromises, is less complicated and addresses other specific concerns. Why are we sliding back? --Kevin Murray (talk) 16:07, 8 May 2008 (UTC)

Both charts have interesting properties, imho. :-) --Kim Bruning (talk) 16:14, 8 May 2008 (UTC)

I've removed the CCC title from the text at the bottom of the chart. --Kevin Murray (talk) 16:18, 8 May 2008 (UTC)

Will we be working on more charts? --Kim Bruning (talk) 17:19, 8 May 2008 (UTC)

I prefer some of the earlier iterations of the 5 series charts, but I can live with the current compromise. However, I'm happy to work with folks on new ideas, or are you saying more charts? I'm happy to work on charts with the team. When we come to a final agreement on the current chart, I'd like to redraw it in better resolution. Cheers! --Kevin Murray (talk) 17:43, 8 May 2008 (UTC)

Sorry for any confusion. I do think either chart should work, and I put the older chart up just as a test to see if it looked OK (which it did). But the newer chart CCC 5f seems best, and it is back now, with better wording, so that's fine. Is the chart better at 400px, or leave it at 200 or maybe 250px? We have put in a lot of work on these charts, particularly Kevin. I hope we can stick with the one we have now, (with the .svg as a back-up). No doubt there could be further fine-tuning of the graphic aspect, but I think we can regard the months spent on this as having been worth it. --NewbyG (talk) 22:38, 8 May 2008 (UTC)

If anyone is interested in working on more charts, then just put the prototypes up on the talk page and we can all have a look at them. Cheers! --NewbyG (talk) 01:03, 9 May 2008 (UTC)

More flowcharts

• Image:CCC Flowchart 5.jpg
• Image:Consensus new and old.svg

Actually, I typically still use the original flowchart for teaching (and it's also the only one that people can easily translate, since it's in SVG format).

I'm ok with having evolving charts up on the main page, but I'm not quite satisfied with what you've gotten up to in series 5 (sowwy :-/ ) If you're no longer intend on working on new charts for now, then they're not evolving, so I'd like to put the original chart back up. When you start making new charts again, feel free to post them upfront, and we'll all take a look again. :-) Is that ok? --Kim Bruning (talk) 12:09, 10 May 2008 (UTC) even so, this particular flowchart 5 right above/next to my text is interesting for many reasons. I'm going to be using '1', 2, and 5 over time.

Consensus new and old.svg needs some work. The two rectangular boxes bottom left should really be one box.

The other problems with that flowchart, which mean that it is not inherently superior to the 5 series flowchart, have been discussed at length in the previous archive, and this one.

I can live with the svg, but the jpg is a better flowchart.

If the svg format is superior, why is that chart so blurry at under 250px?

Both flowcharts can be improved, if work is going to continue.

The project page does not have to have a flowchart on it; if there is one, it should be one that works, and one that satisfies most of the concerns raised by most editors on the discussion page (or in edit summaries!). There are some problems with both charts, minor ones, but the faults of the svg flowchart have been around for longer, without being fixed. Cheers! --NewbyG (talk) 21:57, 10 May 2008 (UTC)

Some suggestions

Here are some of the problems with the flowchart Consensus new and old.svg, and suggestions for improvement. (For the purposes of this page, not necessarily any lectures or anything outside this page)

If the bottom two boxes of the svg were one box, which read Use the talk page to discuss your ideas with other editors, and find a reasonable compromise, that would be better.

And if the top right box read Think of a reasonable change that might integrate your ideas with theirs, and make an edit, or discuss those ideas with other editors that would be an improvement.

And if the text fit into the boxes. --

CCC Flowchart 5f doesn't have the same problems.NewbyG (talk) 22:47, 10 May 2008 (UTC)

• The svg format is so much easier to edit and translate, that almost all other issues pale by comparison. (check out commons for the many variants and translations available for chart 1. Also note how neither of us are editing the non-svg flowcharts.)
• Hmmm, go ahead, why should the last 2 boxes in chart 1 be merged? (in fact, why not just go ahead and do that? Do you have inkscape installed yet?)
• Note that the "Think of a reasonable change that might integrate your ideas with theirs, and make an edit, or discuss those ideas with other editors" is actually a merged box (it includes the discuss first concept, but hides it) , and might very well lead to the infamous infinite loop (aka. using old deliberative assembly terminology, it allows unbounded filibuster without any form of cloture). I'm not sure that I've quite convinced you that infinite loop/infinite filibuster is a bad thing. Else why do you keep proposing systems that allow it? :-P

I've already stated my issues with 5f, I think. Perhaps we have different priorities, and hence different issues? --Kim Bruning (talk) 23:18, 10 May 2008 (UTC)

• Image:CCC Flowchart 5f.jpg

Well the merged concept is actually something like this : think about a possible edit, discuss it if there seems some good reason to do so, or else go ahead and make the edit. That same merged concept is actually hidden in any of the rectangular boxes in both charts. --

Hey : No chart can cure the world of filibustering! Just because the svg chart has an arrow on it or something aint gonna change one thing in the RW. The svg chart doesn't allow for any discussion except after a revert, and even then there is endless possibility there for filibustering. In fact look it up in the dictionary - Take it to the talk page = opportunity to filibuster. --

Flowchart 5f is a flowchart, not a magic wand it cannot prevent filibustering or something like that happening, but the flowchart does not promote filibustering, any more than new and old.svg does. I do not know what you mean there at all. --

If the svg format is so easy to edit, why not put up a prototype with that extra box removed and maybe the wording tweaked, I haven't seen any variations of the svg image, the only one being edited is the jpg. --

Um, when you are finished discussing, someone is gonna have to make an edit. Does that really need to be said, it seems kinda obvious. --NewbyG (talk) 03:05, 11 May 2008 (UTC)

Image:ConsensoAlternativo.svg

[2]

Image:Vejen til konsensus.svg

[3]

Different views of consensus

Here's something that hasn't been documented yet:

Consensus on the en.wikipedia is not a single coherent thing, since the wiki keeps people apart, and there's very little centralization (there should be none at all, but that's a story for another day).

So 2 people can end up disagreeing on the state of (global) consensus. Much dispute ensues!

--Kim Bruning (talk) 12:19, 10 May 2008 (UTC)

I was thinking about this the other day. I'd like to see a statement in the intro that says what consensus is ideally, and then what it is practically on Wikipedia. That way we always have a goal to strive towards, even though we know we will rarely actually reach it.

"Ideally, consensus decision-making means that when there's a disagreement, editors discuss the topic rationally and are swayed by each others arguments until a unanimous agreement is reached. Although we strive for this ideal, "consensus" in practice..." — Omegatron (talk) 18:50, 10 May 2008 (UTC)

+Ideally, the discussion takes place entirely in the form of edits to the page being discussed. :-) --Kim Bruning (talk) 23:20, 10 May 2008 (UTC) this is the shorter of the two loops on 'chart 1' ... ok, we REALLY need to get our charts act together :-P

I agree wholeheartedly that editing is much smoother and more efficient when people edit the page directly instead of talking about it in endless loops. — Omegatron (talk) 00:35, 13 May 2008 (UTC)

So 2 people, or maybe 200 people can end up disagreeing on the state of (global) consensus. Much dispute ensues! Maybe --NewbyG (talk) 23:43, 12 May 2008 (UTC)

Welcome to my hell life. :-) --Kim Bruning (talk) 23:57, 12 May 2008 (UTC)

I don't believe that even ideally "the discussion takes place entirely in the form of edits". Many things on Wikipedia are too complicated for that. Best regards Rhanyeia♥♫ 19:29, 14 May 2008 (UTC)

Wider vs narrower consensus

Wider consensus is stronger than local consensus. Not because we say so, but because there will always be more people who subscribe to a global consensus. So typically situations will trend towards the global consensus, rather than the local version (though influence does flow both ways). --Kim Bruning (talk) 18:13, 10 May 2008 (UTC) I'm trying to word this clearly in the example on the page, but someone keeps reverting. I get the idea they think I'm saying that wikiprojects determine consensus or so. That's not right ;-)

• Fwiw I agree completely and back your efforts completely to provide clear and unambiguous phrasing on this point. Brava! Eusebeus (talk) 18:44, 10 May 2008 (UTC)

Some questions unresolved from /Archive 5

Discuss first sounds like a good idea to some. We're investigating.

• When does discussion take place?

According to Image:Consensus new and old.svg discussion only takes place if you disagree with a revert, and at no other time. That aint right.

According to Image:CCC Flowchart 5f.jpg, discussion takes place if you disagree with a revert or modification. That aint quite right either.

What happens, that after a modification, the "discussion" may take place in edit summaries. So the text in the box that says "Discuss ideas for improving the page" should be interpreted that way (or adjusted) so as to include making an edit, with an edit summary. --

I am not sure where in consensus new and old.svg that the discussion box should go, at the moment that flowchart has "Discuss" as a label on a path between two boxes that actually represent just one box, and doesn't cover cases where discussion might take place. --NewbyG (talk) 22:27, 10 May 2008 (UTC)

Interesting! Let me think about that. --Kim Bruning (talk) 23:21, 10 May 2008 (UTC)

Discussion should take place whenever there is indications of disagreement. So if you suspect that the edit you are making will be controversial you should discuss before making the edit. On the other end of the spectrum is edits you feel are coming from established you where you only need to discuss if the edits are reverted. Taemyr (talk) 15:39, 12 May 2008 (UTC)

No, you should make the edit, then see what the other people actually bring to the table, it might be something totally different from what you expect! If you discuss first, you can find yourself never getting around to editing. So assume good faith of other editors, and take a WP:BOLD plunge, you are likely to come out ok. --18:02, 12 May 2008 (UTC) (User:Kim Bruning)

Still unresolved, I would say. --NewbyG (talk) 07:26, 14 May 2008 (UTC)

Well, it seems the 5 series so far may not have answered the question. Consensus new and old works by avoiding that question. --NewbyG (talk) 00:29, 15 May 2008 (UTC)

Global consensus cycle

First you read how to edit, then you edit, then you learn, then you write down how you edit, then you read how to edit some more, then you edit... etc...

It's a wider feedback cycle by which we all learn how to use the wiki. It is an undeniable fact that we do create project namespace pages, and it is an undeniable fact that these pages reflect consensus. Therefore it is imperative that consensus must be documented. One editor is apparently denying this fact.

Why? --Kim Bruning (talk) 18:02, 12 May 2008 (UTC)

--we do create project namespace pages, = project namespace pages are created, you can see them every few minutes or so on Recent Changes. --

--these pages reflect consensus, = these pages reflect consensus every day, every minute, every microsecond, and for all eternity, undeniable fiction. --

--consensus must be documented, = ?? every consensus must be documented ?? every document requires consensus ?? HOW TO REACH consensus must be documented ?? Only how to reach consensus may be documented ?? If no document exists, it didn't happen. ?? --

--What? --NewbyG (talk) 23:34, 12 May 2008 (UTC)

Back to documentation mode

Errr, I can't quite figure out what you're saying there?

Wait, reading back I think I skipped a step in my reasoning too ... <pulls a nice big "oops" smile>

Lemme try again.

The saying goes "if it isn't written down, it never happened". It's an important part of any process to write documentation. That's how other people learn from your actions, and also helps you to remember how to do stuff 6 months down the line.

So if only for that reason, documentation is very important.

But in the case of a consensus system, documentation is extra important. Someone needs to keep track of what everyone is agreeing on, and why they're making those agreements, because otherwise, you get lost. Sure, you can search diffs and so forth, but there are so many diffs every day on wikipedia, that you just can't keep track.

Part of the consensus process is therefore to keep track of a summary of what current consensus is. It would be very tricky to maintain a consensus system on the english wikipedia without documenting global consensus.

Does that make a bit more sense?

--Kim Bruning (talk) 23:56, 12 May 2008 (UTC)

--Well up to a point, but no not really, and it has nothing to do with this page. It sounds like compiling Hansard or some record of decisions made by parliament, or maybe a newspaper or something. Not the page where the consensus-reaching process on en:wikipedia is explained. --NewbyG (talk) 00:09, 13 May 2008 (UTC)

This is very hard for me to explain, because it's so very obvious to me. Our policies, guidelines and essays document and serve as a common memory of our consensus decisions. Before you decide to do something, (if you're smart), first you read existing policies, guidelines, and essays, then you figure out what to do that's in line with consensus (WP:WIARM) and do it, then when you're done, you write down what you did, thus contributing back to the policy/guideline/essay pool. Next time anyone does something similar, they'll read that information, figure out what to do, and improve on it once again, and write it down again. The time after that... etc. And that's how long term global consensus is built up over time. --Kim Bruning (talk) 01:30, 13 May 2008 (UTC)

REPLY--Our policies, guidelines and essays document and serve as --[PART OF]-- a common memory of our consensus decisions. Before you decide to do something, (if you're smart), first you read existing policies, guidelines, and essays, then you figure out what to do that's in line with consensus and do it, then when you're done, you write down what you did,

--[for personal use and] thus contributing back to the policy/guideline/essay pool

--[If there is an appropriate policy guideline, essay, or document, or if one should be written].

Next time anyone does something similar, they'll read that information, figure out what to do, and improve on it once again, and write it down again. The time after that... etc. And that's how

--[a set of current and historical pages] [ which document the changing history of "long term global consensus" which is built up over time] builds up over time. --

These ideas seem right-alright, regarding the Documentation section. Kim, do you see there in [...]. where I see these concepts in a slightly different light, as I try to understand it, at this yada-yada point in time? --NewbyG (talk) 22:52, 13 May 2008 (UTC)

Just some ideas. (Global consensus cycle - where is that going to be discussed?) --

Our policy is what we do, before it becomes what we write down.

Our policy is to follow what our policies mean, not necessarily what they say!

If a page is not applied or acted upon, it is not a policy page. (It may be a historical page.)

If a policy page is applied, and is edited, it still has consensus as policy.

If a policy page is applied, but not edited, it still has consensus as policy.

If no-one ever reads a page, that page does not have consensus.

Just some ideas. Does any of that make sense? --NewbyG (talk) 23:14, 13 May 2008 (UTC)

I don't understand this:-- Part of the consensus process is therefore to keep track of a summary of what current consensus is.

-- Um. A summary of what current consensus is is the purpose of each individual policy page. Keeping a track of that is the list of policies and guidelines. --

The consensus page could say that "current policy pages reflect current consensus", which doesn't really need to be said. --

Maybe what you are trying to say here is something else, and is something which we ought to think about including in the Documentation section. I think there is something there, but I can't quite grasp it. Cheers! --NewbyG (talk) 23:33, 13 May 2008 (UTC)

• Wikipedia talk:Consensus/Archive 4

--We do encourage people to maintain documentation on the current consensus, and the documentation is typically fairly well maintained, but do be aware that a wiki can never be 100% reliable, and do check everything you read for yourself, as is true of all other namespaces. --

This was reasonable (17 Jan 2008) up to a point; the accompanying material not promising at all IMO. --NewbyG (talk) 07:21, 14 May 2008 (UTC)

It would be very tricky

It would be very tricky to maintain a consensus system on the english wikipedia without documenting global consensus. Well, yes, but that is not the purpose of any section of this project page, particularly the section on Exceptions. --NewbyG (talk) 00:13, 13 May 2008 (UTC)

No, but here I'm arguing that we need to document. This might be a good thing to place in a section called, say, "documentation". ;-) --Kim Bruning (talk) 01:21, 13 May 2008 (UTC)

Need to document what? On this page we "describe" how consensus is reached. We add links to other pages which document where consensus was reached, and what consensus was reached in reference to particular topics, for instance, Wikipedia:Editing policy represents the current consensus as to best editing practice, and Wikipedia:Office actions states current policy and gives details of how that policy is applied. --

On this page we explain some of the processes which relate to consensus-reaching, not the actual details of what consensus was reached in every actual case. --NewbyG (talk) 01:35, 13 May 2008 (UTC)

:-D . This is the page about consensus, so sometimes things can get confusing that way. We need to document the fact that consensus must be documented. I'm going to sleep, I'm sure I can say that more clearly in the morning. :-) --Kim Bruning (talk) 03:23, 13 May 2008 (UTC)

--We need to document the fact that consensus must be documented. Yes, that's right. Confusing, though, isn't it. (My head hurts) --NewbyG (talk) 03:30, 13 May 2008 (UTC)

Just another edit

[4]. NewbyGuesses (and also Rhanyeia)... it's important to explain *why* things work the way they do. Just removing the explanation won't help at all. Removing things simply because they are too complicated is probably not a great idea either. :-P Can we instead figure out a way to clarify them or make them simpler? --Kim Bruning (talk) 18:05, 12 May 2008 (UTC)

Here's some more detail on what I'm trying to explain: Wikipedia:Requests_for_arbitration/Footnoted_quotes/Workshop#Consensus --Kim Bruning (talk) 18:15, 12 May 2008 (UTC)

[5]If these recent additions should go back up on the page, they should be put up on the discussion page for a look. On the project page, they appeared not to explain or make anything clearer, and I reverted them. --NewbyG (talk) 23:27, 12 May 2008 (UTC)

That's what I just did, right? I provided a diff. :-P So. Now that we're agreed, what do you mean didn't make anything clearer? I'm trying to explain how consensus moves around with people. See the arbcom case thread where hmmm, I explain in more detail. I'm not sure what to do with that information. Also I saw the other sections could be clarified as to how things worked too... Can you give me any ideas?

• Declarations by jimmy wales and the board have policy status. Yes, but WHY? And why should we follow them? It's not because they were carved in clay and handed down off the mountain.

It's because "These have policy status because they're there to protect you or help you, it's probably a bad idea to break laws." And why should you listen to developers? Because "it would be a bad idea to lose all your work due to a known bug/feature.", among other things. Isn't this obvious? No, obviously it's not.

• Office actions act outside the english wikipedia community. Oh wow, so do we have to grin and bear that, if you disagree or want clarification, do you just have to sit on your chair clench your teeth? What do you do if you encounter an office action, and want to DO something? Isn't this supposed to be a consensus system?

Well yes... so "You can talk with the foundation office if you have issues with these.". Is this obvious? Perhaps, perhaps not, but in the context of consensus, it's important

• Consensus on a small scale is not expected to override consensus on a larger scale very quickly. But why? Is this handed down on clay tablets? No. Was it decreed by the great maker of policies? No! So why is this the case? Did we just make it up someday? No!

Well, it is the case because even if you have a number of people working locally, sooner or later more people from the rest of the wiki will take a look, and consensus will start to resemble the main consensus more and more. Is this obvious? Nope. it was even misexplained in a arbcom case!

All of these things should be explained as best we can. So how do we go about that? <scratches head> --Kim Bruning (talk) 23:44, 12 May 2008 (UTC)

Details

• --Going into too much detail is not required. This [6] is going into too much detail. That is why it is not required. --NewbyG (talk) 00:03, 13 May 2008 (UTC)

We need to go into detail somewhere, right? Where do you propose? --Kim Bruning (talk) 01:20, 13 May 2008 (UTC)

There is a link provided to Wikipedia:office actions. Any need for details about that item is satisfied best by following that link. That's easy. --NewbyG (talk) 01:52, 13 May 2008 (UTC)

Well, ok, maybe... how about the other two? --Kim Bruning (talk) 03:21, 13 May 2008 (UTC)

Practice

I do agree with simplifying the page, but at some point we might end up oversimplifying. We do actually have to explain what to do.

If you notice that a particular policy or guideline page is not in line with current consensus, feel free to update it. If you notice that certain things haven't been written down yet, create a new page in the Wikipedia: namespace and write it down.

Right.. once again we might be oversimplifying. This is how you write policy. Where do we document this instead?

Are you systematically removing all references to policy, or is that accidental? --Kim Bruning (talk) 01:40, 13 May 2008 (UTC)

• Please do not make foolish statements or accusations about other editor's intentions merely as some sort of rhetorical flourish. --NewbyG (talk) 02:56, 13 May 2008 (UTC)

  No no, I mean you've actually been removing a lot of stuff that just happens to have to do with how policy is created atm. That could be coincidental, or accidental, or maybe you were planning to start a new page, or maybe the content was already on some other page that I haven't seen... or etc. Basically, what's up? :-) --Kim Bruning (talk) 03:20, 13 May 2008 (UTC)

While cutting down on wording is nice, the actual goal is to explain how to form consensus, of course. So I've put this particular section back for now. (some of the other changes took some effort of will to leave untouched, but were ok afaict :-) ) --Kim Bruning (talk) 01:48, 13 May 2008 (UTC)

--Ah, well, yes then. Eureka, we have dug our way out of that one pot-hole, at least.

We do indeed need to document the fact that consensus must be documented (if we can) !!

And, yes, [7] that section (Documentation) is now under way (see, no-one was "denying" anything!) The text is a little "dodgy" at this time, and there is a long way to go, if the section is to really cover everything, which we do not have to do, but it is a worthwhile avenue to go down. --NewbyG (talk) 06:42, 14 May 2008 (UTC)

Here's the possible new section which was added by Kim Bruning and removed by me:

== Documentation ==

If we find that a particular consensus happens we often document it as a best practice, to save people the time having to discuss the same principles over and over. If you notice that a particular policy or guideline page is not in line with current consensus, feel free to update it, or to join the discussion on the talk page. If you notice that certain things haven't been written down yet, create a new page in the Wikipedia: namespace and write it down.

Documentation is important so that we can learn from history, and the mistakes and smart things others have done. If no documentation is written, everyone must guess at the unwritten rules, and new users will never learn how to be good Wikipedians.

Some problems which I see with adding this section: Wikipedia already has a lot of very long policies and guidelines, so writing more isn't necessarily helping newcomers. It's difficult to become a Wikipedian also because there is so much text to read, and newcomers are probably also spending time reading articles and source material. In order for it to be easier for newcomers to join, simple and concise rules/instructions are better than an almost endless amount of text. This section encourages writing more and more and I think that isn't necessarily good. I also don't see anything in this section which would add something so very important. There's already a section about policies on Wikipedia:Policies_and_guidelines#Sources_of_Wikipedia_policy. Best regards Rhanyeia♥♫ 19:24, 14 May 2008 (UTC)

the need to document best practices

True, but at the same time we need to document best practices, both for new people and for old people, so that we know what we are doing. If we do not document, we cannot learn from our history, and are doomed to repeat it.

The fact that one must document, document, document is ingrained in my psyche (I'm trained as a scientist, and work as a programmer). It is almost impossible for me to understand a world where documentation does not exist.

Certainly I agree that we need short concise information for people starting out, but there are also a lot of finnicky details you don't even WANT to remember, and which we rather write down to look up when needed.

Basically, unless everyone has photographic memory, if we don't document, we get lost.

--Kim Bruning (talk) 20:26, 15 May 2008 (UTC)

While I agree with what Kim is saying, in the context of the discussion, I think that Rhanyeia is not objecting to documentation, but to over documentation. This has been a concern of mine for some time. As our rule sets creep in there complexity and number editors seeking solutions can become mired in the detail and minutia. A good balance shuld be sought. I too thought that Kim's clarifications strayed beyond my ideal, but not enough to oppose the edits. The ideas are sound but the added bulk is a concern. --Kevin Murray (talk) 20:49, 15 May 2008 (UTC)

I think I'm shifting towards the position that we actually need 2 sets of documentation. (Actually, I was sort of shifted towards that position before..).

First there's this simplified documentation (like WP:5P) which just gives you a general overview of how things work, and then you have more specialized documentation that describes all the finicky details.

Both sets are very important. The former to learn by heart to have some basic principles and a bit of a road-map, and the latter to actually use in day to day work. This is sort of similar to what I do when programming, there's usually a very simple language definition and tutorial, where you have to maybe memorise 50-100 words (or less these days even), and then there's this huge phone-book like document that folks don't even bother to print out anymore, that provides details of all kinds of add-on libraries that can handle finicky little details you don't even want to know about (until you actually do ;-) ). On any particular day, you might have a general idea of what to do and what's possible, and then you look up the best way to actually get parts of that done in the big reference book. No one memorizes the big book ever, you don't need to. It typically has a really decent index. :-)

Can we structure our documentation like that too? --Kim Bruning (talk) 21:56, 15 May 2008 (UTC)

That is mostly reasonable. The question is (at the moment) not if the process of documentation is important, but rather "how" it is to be explained. The current draft text has been objected to as not doing this well enough, it is unclear. I think maybe the subject is better being explained on another page, not the consensus page.

WP:CON used to have one sentence about documentation. Maybe one sentence does the job better, as far as this page is concerned. Or else a better draft is needed. --NewbyG (talk) 22:48, 15 May 2008 (UTC)

Urk. If anywhere, then here. I'm admittedly having huge difficulties trying to wrap your head around where you might be coming from. To me it's just so integral to the consensus process that I can't see it as separate. You read how to do stuff, you do stuff, then you write down what stuff worked. That's totally a part of what we do. /me shuts up and tries to think of ways to explain why the sky is blue and why apples fall down off trees... --Kim Bruning (talk) 23:08, 15 May 2008 (UTC) Colours, prisms, gravity, down-ness, light going slow... ARGH!

Kim, the policy pages should be concise and actionable. The why should be discussed in essays etc. The problem is that when the policy pages get cloudy, then actions such as AfD require wiki-lawyers and wiki-judges. You are right - right - right for science, just not for guidance. Cheers! --Kevin Murray (talk) 23:14, 15 May 2008 (UTC)

I'm not disagreeing ;-) But I'm thinking about this at the process level as well. There's many ways to do what we want that ostensibly work, but kill the all-important day-to-day, minute-by-minute feedback cycles of the wikipedia decision making process that keep the place running (which is what it's actually all about in the first place :-P ), and there's many ways to improve the decision cycle, but they make the project namespace a bigger mess than it is already, and even less useful to new people. Now how does one think of an elegant system that achieves both aims? :-) --Kim Bruning (talk) 23:57, 15 May 2008 (UTC)

Kim, I'm with you on this, because solving this problem would solve many problems at WP. Simple direction with a complex why documentation. Talk pages wander way too much and make a poor archive for all but the most dedicated researchers. Essays are prone to opinion and we will likely end up with mountains competing explanations. I look forward to helping you to seek the best solution. --Kevin Murray (talk) 00:03, 16 May 2008 (UTC)

From the nutshell

A --- Consensus is Wikipedia's fundamental model for editorial decision-making.

B --- Policies and guidelines document communal consensus rather than creating it.

A lead section first two paragraphs.

B lead section last two paragraphs.

A,B Sect 1

A Sect 2

A Sect 3

B Sect 4

A,B Sect 5

B Sect 6

B Sect 7

Some focus. --NewbyG (talk) 09:09, 16 May 2008 (UTC)

A comparison of two flowcharts

• Image:Consensus new and old.svg

• Image:CCC Flowchart 5f.jpg

There are only five possible paths through either flowchart. Three of them are the same for both flowcharts, and two of them are different.

Same

• Edit A --> not modifed --> not reverted --> New consensus (both)
• Edit A --> modified --> Editor A agrees --> New consensus (both)
• Edit A --> reverted --> Editor A disagrees --> Talk page (both)

Different

1.

• Edit A --> reverted --> Editor A agrees --> New consensus (new and old)
• Edit A --> reverted --> Editor A agrees --> Previous consensus (flowchart 5f)

2.

• Edit A --> modified --> Editor A disagrees --> Make an edit (new and old)
• Edit A --> modified --> Editor A disagrees --> Talk page (flowchart 5f)

These are basically the only differences between the charts.

In interpreting the charts, and how the chart applies to the editing process, the wording in the charts, as well as the text in the section concerning the flowchart, also apply.

What do they say?

Basically, (new and old) says edit, and keep on editing until there is a revert. Then go to the talk page.

And (flowchart 5f) basically says edit, and keep editing until there is a disagreement, then go to the talk-page.

But that is nothing like the whole story, which is a bit more complicated than that. --

The flowcharts only tell a part of the story. --

Another way of looking at it

Actually, these particular differences between these particular charts are irrelevant to the way I personally see the editing process, because I always agree with every edit, as soon as I see it, so I always take the "agree" paths. This doesn't work for (new and old), because I like to use the discussion page whenever I want to, so I just IAR that step in (new and old) --

I agree immediately with every edit, and usually wait for some time, maybe a couple of hours to see if it is edited again. Then, if I see a valid edit, I make it, or go to the talk page. That's easy and it works for me, no problemo. --

Using this method, I consider that my best editing is done by leaving completely alone those thousands and thousands of good edits made by other contributors, which, if I find them interesting, I will read and re-read and consider over again. The few edits which I make myself, are pretty easy, and work some of the time, not all of the time. --NewbyG (talk) 00:41, 14 May 2008 (UTC)

• Each chart is flawed. We can accurately map the entire decision process, but then people object to a chart which is overly complex. The latest version of Chart 5 became too simplified, thus the objection to having to discuss rather than edit if your edit is opposed. If this is the objection it is easilly fixed since it was addressed in earlier versions of the 5 series charts. Will that resolve the issue? --Kevin Murray (talk) 06:51, 15 May 2008 (UTC)
  • Possibly, depending on what you do :-). And actually you bring forward an interesting project there *eyes glaze over, thinking of that huge flowchart in shimmering technicolor* --Kim Bruning (talk) 00:01, 16 May 2008 (UTC)
    • Since the new format has failed. Why don't we return our discussion to an ammended vewrsion of the original chart. But in the interim I oppose any chart so the old chart doesn't get stuck as erroneous status quo. --Kevin Murray (talk) 00:10, 16 May 2008 (UTC)

--Flowchart 5f was up on the project page most of April, and editing of wikipedia continued pretty well as normal, didn't it. Consensus new and old works. It has been up on the page without causing problems for a long time, even if there is current discussion about discussion being documented in a flowchart. 5g is a prototype. There doesn't have to be any flowchart on the project page, but I don't have any problems with Consensus new and old going back on the page, in whatever section is appropriate. Then we can workshop prototypes on ths page. --NewbyG (talk) 01:00, 16 May 2008 (UTC)

5F was a significant compromise between many people, but wore the restictions of too many cooks spoiling the broth. The older chart was not widely discussed and has many flaws in form and logic I'd prefer that we do not repost a chart until there has been a robust discussion. Maybe we should wait until the page is stable again and then chart the final text rather than trying to back into a poorly conceived flowchart. --Kevin Murray (talk) 03:12, 16 May 2008 (UTC)

I would say that this flowchart represents a very simple process. Two copies of this flowchart hooked up would like quite a bit like consensus new and old. --NewbyG (talk) 04:40, 16 May 2008 (UTC)

I <3 chart 6. I'll now admit to only sort of politely agreeing to some of the other charts, but ... I really really like this one :-D

I still need to scrutinize it carefully and figure how it applies in more detail. But it's certainly very elegant. --Kim Bruning (talk) 15:44, 16 May 2008 (UTC)

Ahh, now there's a Gordian knot unbound. Thumbs up on Chart 6.--Father Goose (talk) 05:36, 17 May 2008 (UTC)

5g reintroduces ye olde indefinite filibuster (aka infinite loop is what I called it earlier I think). I'm pretty sure that Kevin and I don't actually like filibuster, so I'm scratching my head why filibuster keeps coming back? You need some way to break the cycle ("cloture".. very roughly). Traditionally this has been by allowing a BOLD edit. I know rare occasions where people call a poll (which then typically asplode though ;-)), although when used properly, these simply clarify a discussion (as opposed to making a decision). Are there mechanisms known that allow for "cloture", besides those? --Kim Bruning (talk) 15:54, 16 May 2008 (UTC)

Kim, were you talking about 5G or 5H (below)? What you call a filibuster seems inherent in the WP process unless there is intervention. --Kevin Murray (talk) 16:14, 16 May 2008 (UTC)

5H, sorry, yes. And no, like I said, nothing stops you from editing during a discussion. The 3RR further prevents a revert war (which you could call a "filibuster by editing"). --Kim Bruning (talk) 17:08, 16 May 2008 (UTC)

Hmm, I prefer making flowcharts, and then describing them in text, for many reasons. (not the least because that's how I learned to write texts and build other things besides ). Flowcharts allow you to quickly view and asses the logic and flow of your ideas, and they then form the framework upon which you build your description of a situation or solution. --Kim Bruning (talk) 16:02, 16 May 2008 (UTC)

I agree. In construction a lot of poor ideas are eliminated when you draw the plans in multiple dimensions. --Kevin Murray (talk) 16:17, 16 May 2008 (UTC)

The map is not the territory

Wikipedia:Governance reform

Although this proposed policy seems likely to fail, there are some interesting observations, concerning consensus.

Policy pages are typically maintained by a large and diverse community of editors who visit and help maintain the pages. New editors join in and help out. There are situations where people think it is appropriate to filibuster. Currently these situations are slowly being corrected. This process is thriving, with hundreds of policies, guidelines and essays to provide advice to people on every single subject from how to convince the community to change course to how many spaces should follow a full stop. --

A lot of talk about consensus

WT:BLP A lot of talk lately about consensus, but not much in the way of convincing arguments.

-- consensus alone, not one user presenting some holy grail of information, decides. --

-- consensus is not the majority, nor is it your personal will above everyone else. --

-- Policy pages are intended to reflect consensus, but they are not the policy/consensus themselves ("the map is not the terrain"). --

-- You can't claim that you are the only person able to divine consensus. --

-- Hmmm --NewbyG (talk) 05:50, 14 May 2008 (UTC)

You've been following me around. :-) I'd been warned that BLP was a tricky place to go ^^;; --Kim Bruning (talk) 06:07, 14 May 2008 (UTC)

Well, I was wondering why I was getting no replies to my extremely well-considered posts to this page, and then I saw on my watch-list that WT:BLP was hosting a shout-a-thon. I have nothing to say there, but it makes for interesting reading. Very interesting! --NewbyG (talk) 06:50, 14 May 2008 (UTC)

• --Convincing arguments are needed in order to successfully implement changes to currently established project wide practice or to document changes to established project wide practice. Convincing arguments are those that can be expected to sway the larger community.-- [8] --NewbyG (talk) 09:01, 14 May 2008 (UTC)

Hmmm, I just pulled out a section that had some large amount of redundancy... I hope I didn't just stomp that phrasing. Hmmm... now what? <scratches head> --Kim Bruning (talk) 15:46, 16 May 2008 (UTC)

I wouldn't worry too much about what you cut out. I think that the lead section is fairly tight. I'd focus on streamlining and condensing some of the sections, and then come back to make sure that the lead is an accurate summary of the main text. --Kevin Murray (talk) 15:52, 16 May 2008 (UTC)

Edits

Could the following edits be explained please, and can we think about these sentences? [9] What do you mean with "foundation issue #3"? [10] Isn't it so that if the consensus was based on good reasons it doesn't necessarily shift, sometimes it happens and sometimes not, I wouldn't think it's possible to predict that it "will tend to" shift. [11] Do you mean that the talk page message is the "edit summary"? It's not understandable from the text, with the earlier version I didn't even know why it said a "long edit summary". Edit summary means what's written in the edit summary box. One could write a talk page message, wait for two days or longer time, and then edit, and say in the edit summary that there's a message on the talk page. And even if you edit immediately after a talk page message you would still use at least some edit summary. Best regards Rhanyeia♥♫ 17:37, 16 May 2008 (UTC)

• m:Foundation Issues, #3. (Originally, wiki's did not have talk pages, so they are clearly not essential. If talk page consensus was the final decider of content, then that would have been what the foundation issue said :-P )
• In theory, if consensus was perfect, and humans were perfect (and identical), yes it would stay the same. In reality, every human being is unique, and therefore has a slightly different view of the world. The practical upshot is that anytime a new person starts working on a page/joins a conversation/etc... the consensus will always shift in practice.
• A "long edit summary" is an edit summary you want to make that is too long to fit into the edit summary box. What do you do? Well, first write your summary on the talk page (instead of in the summary box) ... that's one of the reasons we have talk pages you see... then save your edit, with "I've put my summary on the talk page" or some such as your edit summary. If you do things in any other order, or with delays in between, you end up confusing people in a myriad of interesting but frustrating ways... Perhaps this kind of detail could go elsewhere though. <scratches head>

--Kim Bruning (talk) 19:29, 16 May 2008 (UTC)

• • Kim, I love you but your explanations are more confusing than your edits. --Kevin Murray (talk) 19:31, 16 May 2008 (UTC)

Urk. Ok, ask me to clarify something specifically... maybe I can word more clearly. --Kim Bruning (talk) 20:54, 16 May 2008 (UTC)

Kim, you answered in double speak. The first answer denied the need for talk pages because we have edit summaries, the third anwser says we need talk pages because edit summaries don't work well. Too much go-juice? --Kevin Murray (talk) 21:00, 16 May 2008 (UTC)

Not enough go juice, rather :-P

The first denies a need for edit summaries and talk pages at all. Wikis can function without them. Of course nowadays we're totally addicted to our shiny summaries and pretty pages, so we sometimes forget that they're not actually essential. (compare: before the 1980's cell phones didn't exist, nowadays a lot of people think they would die if they didn't have one ;-P )

The third answer doesn't say that anything doesn't work well, it just describes a workaround in case things don't work like you want them to. (some people are so addicted to edit summaries, that they make them too long, and the text spills over onto the talk page :-P).

I think you've mostly just assumed some things which I actually didn't say (thank goodness), and then it was the assumptions that collided. :-) --Kim Bruning (talk) 21:47, 16 May 2008 (UTC)

It's kinda funny, if there is anyone more "addicted" to talk pages than myself, it just might be KB! --NewbyG (talk) 22:16, 16 May 2008 (UTC)

Some thirty edits

Other than the flowchart, which is still at an experimental phase, there is a lot of good tightening and streamlining here. (34 intermediate revisions not shown.) [12] (12,462 bytes) --> (9,848 bytes) a more presentable policy page, no policy bloops? Any comments? --NewbyG (talk) 23:41, 16 May 2008 (UTC)

Later Diff. (29 intermediate revisions not shown.) (29 intermediate revisions not shown.) --NewbyG (talk) 02:02, 17 May 2008 (UTC)

Later Diff. (64 intermediate revisions not shown.) --NewbyG (talk) 03:00, 18 May 2008 (UTC)

Consensus within the framework of ... is circular

When consensus is referred to in Wikipedia discussion, it always means 'consensus within the framework of established policy and practice'. Consensus among a limited group of editors can not over-ride community consensus on a wider scale.

This confounds some matters, and some of the matters so confounded are circular.

• When consensus is referred to in Wikipedia discussion, it always means 'consensus within the framework of established policy and practice'

  Big problem. Consensus informs policy, then policy informs consensus? That's circular. One informs the other. So either wikipedia is run by policy (but then there is no mechanism to form it), or it is run by consensus (in which case consensus is the mechanism that forms policy too).

• Consensus among a limited group of editors can not over-ride community consensus on a wider scale.

  Yes it can, if only via WP:IAR. It's just that as more people join in the discussion, the consensus among the limited group will tend to yield for that of the larger group in a natural fashion. group a (10 people) says x , group b (10000 people) says y. Over time, more people from group b join group a, and group a will shift towards x. But this is not always true, and the smaller group might tend to convince everyone who enters. It is not the case that one can walk into say WP:BAG on one's own and demand they change their rules to match the community(and not entirely for lack of trying, let me tell you ;-) )

• I think the concept here is that global consensus tends to be more important than local consensus. So that's what needs to be stated somehow.

--Kim Bruning (talk) 00:09, 17 May 2008 (UTC)

See also: Wikipedia talk:Consensus/Archive 6#Consensus is Global not Local

Wikipedia:Policies and guidelines#Sources of Wikipedia policy Documenting actual practices and seeking consensus that the documentation truly reflects practices is the most effective method for policy documentation.

see above.--NewbyG (talk) 01:05, 17 May 2008 (UTC)

Well, ok, now I saw it at this location and in this context, and whatever the case may be , it's wrong here. When changing process, you first gain consensus, then you update the page describing the process, right?. If consensus must always be within existing documentation of the process (as this suggests), then consensus can never change the process or the documentation. --Kim Bruning (talk) 12:47, 17 May 2008 (UTC)

I think we can merge the talk page stuff into participating in discussions.

As a general comment: One thing to keep in mind is that we do actually need to explain what to do, not just the underlying theory ;-) --Kim Bruning (talk) 13:02, 17 May 2008 (UTC)

Readded section on gaining consensus on talk pages. I have a feeling we're sort of missing the central theme here... "what is consensus" :-P --Kim Bruning (talk) 13:13, 17 May 2008 (UTC)

Sia GRASSETTO

Sia GRASSETTO nella pubblicazione; potete anche usare la pagina di colloquio per discutere i miglioramenti all'articolo ed al consenso della forma riguardo alla pubblicazione della pagina. Nel caso della politica pagina un più alto livello di partecipazione ed il consenso è preveduto. Nei casi in cui il consenso è particolarmente duro da trovare, la partecipazione dei redattori indipendenti o l'aiuto sperimentato nella discussione può essere necessario. Se la pubblicazione della pagina è impedetta vicino pubblichi le guerre, o è interrotto, o il consenso non può essere trovato alla pagina di colloquio con la discussione ordinaria, là è processi di risoluzione di disputa più convenzionale.

This is in Italian, but which section? --

Be Bold in editing; you can also use the talk page to discuss improvements to the article, and to form consensus concerning the editing of the page. In the of policy Standardhigherseiten ein of ordnet Teilnahme and Konsens expectedis. In cases where consensus is particularly hard to find, the involvement of independent editors, or more experienced help in the discussion may be necessary. EditingsEibe of the Seite is impeded by warsedikt, aber is disrupted, aber Konsens Boot Be found man the talk Seite through ordinary Diskussion, there Ar more formal Streit processesentschließung.

Ist Deutsch? --NewbyG (talk) 06:12, 18 May 2008 (UTC)

Gallery of flowcharts

I haven't really followed this discussion, so my apologies if this has been suggested before, but why not just make a gallery of flowcharts at the bottom of the page? "here's some ways consensus has been formed in the past", etc, "click to expand". -- Ned Scott 03:40, 18 May 2008 (UTC)

Good idea, Ned, but to what purpose? (Check the flowcharts in sections above.) --NewbyG (talk) 05:58, 18 May 2008 (UTC)