Wikipedia talk:Relevance/Archive 3

Problems to solve

I (WikiLen) am not claiming any proposal successfully solves these yet. To solve:

1. Rule-creep hassles problem: A relevancy "policy" that just links to other places will feed fuel to some editors, especially new ones such as myself, leaving an opening for: "Hey, there is a policy. It is scattered about. Let's move it all into one place — cool..." ;)
2. Bogus problem: Editors need a small set of sentences/phrases, located in policy/guidelines, that they can then quote to wake up confused editors — editors using bogus claims for relevancy. Saves the experienced editor from needing to explain how it is bogus.
3. New-user problem: New users need to be educated about relevancy. First, they naturally want to know about it and second, being so educated has efficiency benefits for the rest of us.

I personally, find the sentiment, "no problems to solve", has a certain elegance to it like WP:Ignore all rules. However, when I look carefully I find "yes", there are problems to solve, certainly minor and mundane but real nevertheless. —WikiLen 01:13, 18 July 2007 (UTC)

That Wikipedia may be getting along fine without a Relevancy guideline, provides no clue as to how things would be with such a guideline. Efficiencies could be worse or better depending on what's in it. However, doesn't the risk it would make things worse loom far larger than the possibility things would improve? I think "yes" — perhaps Father Goose and I have been too ambitious in our scope. What would a simple guideline say that succinctly addresses the above three rather ordinary "problems to solve." And, by the way, I know my version REL4 (archived here) has not succeeded at this — didn't clearly grasp the "problems to solve" until now. —WikiLen 04:52, 18 July 2007 (UTC)

I think it's better not to have a "relevance" policy page. Few or no problems to solve, and the various proposed draft pages don't do much to solve them anyway; whether something is sufficiently relevant just has to be argued out on talk pages.

If there is to be such a policy, the current sections "keep article focused" and "the subject of an article" are pretty good. The section "establishing relevance" offers questions but not answers, which is not much help, especially not if it's to be a policy.

I strongly disagree with this command: "Always explain the effect that a fact has had on the subject." This suggests that for every fact mentioned in every article, additional space in the article has to be taken up explaining why the fact is relevant.

The section "fundamental information" merely replaces the loosely-defined word "relevance" with the loosely-defined word "fundamental".

I disagree with the following: "Facts that are needed to provide a fundamental description of the subject are always relevant. These facts explain what the subject is, what it does (or did), and what it is notable for." Things that the subject did are not always relevant, e.g. what a celebrity had for breakfast one day. Very famous human subjects may be notable for huge numbers of things; not all of these are relevant to an article about the subject.

I still get the feeling, particularly towards the end, that the draft policy was written with only articles about human subjects in mind, but not nearly as much as in earlier drafts.

Sorry for all the negative comments. I appreciate the work people have put into this and I see that it's an improvement over earlier drafts. --Coppertwig 17:31, 21 July 2007 (UTC)

As long as criticism is specific -- as yours has been -- then it's helpful, and much appreciated.

Arguments over relevance do have to be carried out on talk pages, and this proposal is intended to strengthen that process. One thing to keep in mind is that often, that discussion doesn't take place at all. There is a formal process for deleting articles, where people are notified and discussion must take place first. But entire swaths of content can be deleted without so much as a peep, and after it's deleted, there's usually no sign it was there in the first place.

I don't advocate that every single addition or removal be discussed -- that is senseless, of course. But when the discussions do happen, I want to aid both sides by having a solid framework in place, something that states the basics (in a way that reflects consensus) about what belongs in an article and what doesn't. This guideline by no means trumps discussion -- it promotes it. When disputes arise, each side has to present their rationale for inclusion or deletion. Right now, those discussions typically devolve into WP:ILIKEIT and WP:IDONTLIKEIT, and little is achieved. Other times I've seen WP:TRIVIA invoked to justify wholesale deletion, because its advice simplifies down to "integrate or delete" -- and let's face it, integration takes work. This guideline improves that situation, by explaining what context is necessary for meaningful integration, as well as when and how to move off-topic information.

This guideline also tries to lay down some criteria by which to evaluate whether material "offers a broader understanding of the subject". I think most editors would agree at least with the principle underlying this, even if the criteria may need further tweaking and explanation. I think they are a decent approximation of the internal reasoning editors have been using to date. But I readily accept that they need further criticism, field testing, and rewording to state things clearly and correctly, and to represent the consensus position as well as possible.

I don't think the "questions" should be turned into statements -- relevance comes from context, so what should be stressed is not "whether something is relevant", but "whether something can be made relevant" through the addition of context. The answer to that question must be provided by editors, not the guideline. But the guideline does stress that the question must be answered -- "why is this relevant?" -- and gives specific guidance on how to answer that question. I believe this helps to reduce the nebulosity and subjectivity inherent in evaluating relevance.

I can't eliminate all ambiguity from language, but the specific phrase "needed to provide a fundamental description of the subject" is not quite as open-ended as you claim. I also can't forestall all wikilawyering, although any obvious loopholes or omissions remaining in the present guideline can be identified and amended. I solicit all criticism and assistance people are willing to offer.

"Always explain the effect..." probably states the case too strongly. I'll try to fix that -- or if you're willing to fix it, all the better. Similarly, you're right about "not everything it did" is relevant... I'll ponder how to better word that.

The "Connections" and "Biographies" sections do address two "special cases". They could be removed from the guideline, but a great number of fights over "relevance" center on these types of cases, so I think some additional words on them are worthwhile. It's reasonable to shed additional light on areas where it is most needed.

I can't help but feel like perfection is being demanded of this proposal. It covers far-reaching and important territory, so I grant that it should be evaluated with a fairly conservative eye. If it's dead wrong, okay, torpedo it. But if it gets some things right and some things wrong, help me fix the wrong bits. My is aim to cast as much light as possible on an area of ongoing contention. The more people are willing to aid me in this work, the more clarity Wikipedia can offer.

Thanks very much for your input.--Father Goose 22:56, 21 July 2007 (UTC)

I double-checked the proposal against your "Things that the subject did are not always relevant, e.g. what a celebrity had for breakfast one day" comment. A claim that something like that is relevant ignores the sentence that immediately proceeds "what the subject is or does" -- namely, "needed to provide a fundamental description". Any language in any guideline can be taken out of context and wikilawyered, but I don't think anybody trying to be reasonable would pretend that celebrity breakfasts are part of a fundamental description. (Maybe if the celebrity in question were the Pillsbury Dough Boy...)--Father Goose 02:59, 22 July 2007 (UTC)

Things can be WikiLawyered can also be subject to legitimate misunderstanding, so WikiLawyering should be considered as more an example than a sole concern.

Made these changes:

1. Restored Impact instruction without specific requirement for explanation.[1] This way impact can be apparent from context, or needs to be made apparent, without requiring an explicit explanation — should be fine for an encyclopedia. Does this work? I still feel the term "Impact" is a bit vague.
2. Changed Fundamental information to Definition, since I think this is the intended meaning.[2]
3. Anticipating the term "Definition" to be misinterpreted, I added a warning against dictionary cruft,[3] which has been a problem in some articles.
4. Sorted Establish relevance items by likely sequence with an article.[4] No implication of importance is intended.

Revert whatever does not help. / edg ☺ ★ 03:17, 22 July 2007 (UTC)

"Effect" might be a workable substitute for "impact", although I think "impact" conveys something more like "a measurable effect", which is better.--Father Goose 05:51, 22 July 2007 (UTC)

Now with kung-fu grip

User:Father Goose/Relevance (heavily revised)

I am in edgarde's debt for his incisive Review B. I've trimmed off all the fat I could find (there was a lot), increased its clarity as much as I could, and turned it into a set of instructions instead of a wish list.

The "Impact" section still needs clarification, but I think the rest of it is solid. The particular line "Changing the subject's form or history (in particular, any of its fundamental or distinguishing traits)" says what I want it to say but probably not in a way that makes sense to others. I would appreciate feedback or questions about that line to help me shape it better. Any comments on any other part of the proposal are also welcome.--Father Goose 20:18, 19 July 2007 (UTC)

The updated version of this proposal, which has been extensively trimmed by User:Father Goose, has been restored to the project-page. The work put into this version facilitates the best chance for advancing this proposal. Thanking WikiLen, once again, for his efforts; and inviting further work on that draft to continue, in user-space. Having followed this discussion for some time, I believe that consensus currently is demonstrated for the slimmer version now on the page, which achieves greater clarity than other drafts offered to this date, Newbyguesses - Talk 08:09, 21 July 2007 (UTC)

Newbyguesses, what is your basis for finding "consensus currently is demonstrated for the slimmer version" (named version FG 3.x)? I don't see any. Shouldn't we establish that consensus on this talk page first before changing the project page? —WikiLen 16:22, 21 July 2007 (UTC)

Newbyguesses is (I would guess) being polite here. The consensus on this talk page is against version REL3. REL4 is still new, but there has been no rush of support for that version either.

There is no consensus for either version, but the general gist is if there must be an article on Relevance, parts of FG's version might be suitable. / edg ☺ ★ 04:09, 22 July 2007 (UTC)

Dispite the lack of consensus on what was posted (FG 3.x), I will leave it to others to revert, if struggles for consensus so dictate. A revert by me would correctly be perceived as biased and would not help a move towards consensus. —WikiLen 20:05, 21 July 2007 (UTC)

I feel like the goals for the competing versions were to approach each other, with REL3/4 intended as a test bed for some more ambitious or advanced ideas — good for discusson this Talk page, but not suited to in vivo experimentation. / edg ☺ ★ 04:09, 22 July 2007 (UTC)

Three questions test

The section "Establishing relevance" with its three questions is, IMHO, what is challenging consensus for this version: (WikiLen)

• The three questions are very technical and take effort of focus to use.
• It is not field-tested and needs to be because of its complexity. (It may only be useful for Trivia, if at all.)
• Assumes relevancy disputes are struggles of intellect when most likely they are struggles of emotion. i.e: struggles involving bogus or confused reasoning.

Perhaps "three questions" is a specialized approach that could be adopted by Trivia guidelines. —WikiLen 21:19, 21 July 2007 (UTC)

I agree that this is the part of the proposal that faces the most controversy. It's daring; it's an attempt to codify editorial practice which has not been explicitly stated until now. If it fails to capture that practice correctly, it will need to be changed until it does, and if it can never be brought into sync with editorial practice, it will have to be removed.

Field testing! Indeed, that's exactly what it needs. I'll need help with that. I want to test it against specific cases to see if it breaks, or if it is just too awkward to use to be practical.

Having rational tools at one's disposal in the midst of an emotional struggle is positive. First of all, one must assume good faith -- that editors are acting rationally, and are merely disagreeing. Having specific criteria to discuss gives editors a chance to think through their positions, and stands a chance of producing a rational outcome. Responding to emotional language with rational specifics is also better than saying "you're being emotional".

So the question remains -- are the three questions right, or can they be made right? Let's give them some serious stress-testing and see what emerges. I am not claiming that they provide definitive answers -- this is exactly why I presented them as questions instead. I want to see what kind of discussion they prompt, and to see if they can help bring about greater understanding amongst editors engaged in relevancy disputes.--Father Goose 02:13, 22 July 2007 (UTC)

Refactoring of version REL4 complete

User:WikiLen/Relevance is heavily trimmed, revised and refactored from version REL3. Changes are inspired by input from numerous editors,—thanks!—including:

• Review A — by Edgarde
• Critique of WikiLen's draft — by Father Goose

I split draft REL3 into three parts to make draft REL4:

1. User:WikiLen/Relevance — draft REL4
2. "What claims of relevance are false" — an essay
3. "Indirect relevance is sometimes OK" — an essay

Highlights of this draft:

• Edited to focus on "actionable" items.
• Summarizes the two essays I spun off and provides links to those essays.
• The "questions three" concept designed by Father Goose has been cut (impact, fundamental information, and distinguishing traits). I will address my reasons for dropping this at a later opportunity.
• The section on splitting off to new articles has been cut.

My goal remains to find consensus... —WikiLen 06:06, 20 July 2007 (UTC)

New version for this fork

There now exists a REL4.1 version of this fork. —WikiLen 21:30, 21 July 2007 (UTC)

New version again — with new lead

There now exists a REL4.2 version of this fork. Credit to Edgarde for this excellent lead section. —WikiLen 11:47, 25 July 2007 (UTC)

Proposed demarcation for relevance

Submitted: for a fact to be relevant, it must be one of these things:

• Definition – as in the current article.[5] Easiest to identify, always relevant.
• Details – not defined for now, and not all details are relevant.
• Context – this is included in WikiLen's current version, but only vaguely hinted in the current article.

Can anyone think of examples of relevant information that wouldn't fall clearly into one of these three?

Does this list exclude anything other than patent non sequitur?

This short list is over-inclusive, but perhaps this clearly identifies the hedges we need to trim. It is possible that so much of concern will fall under "Details" that this model may be ineffective. / edg ☺ ★ 05:03, 22 July 2007 (UTC)

Definition isn't quite the right substitute for what had been "fundamental". It forces you to drag in that not-a-dictionary paragraph, which is a tangent. Furthermore, there's plenty of stuff in the lead that isn't definition. "Description" would work better.

I'm not sure context should be singled out as "okay"; earlier discussions between WikiLen and myself brought to light "mini-articles" in Gandhi and Bath school disaster which one could claim were useful for context. In truth, they just strayed too far from the subject. I think context should still hew closely to the "about the subject of the article" rule. And on some level, what isn't a detail?

I'd like to suggest evaluating things through the filter of "broadening one's understanding of the subject". Can you figure out generalizations that can help identify when material does, and doesn't, broaden understanding? I'd say "impact" latches on to that concept, and the other two -- description and distinctiveness -- fill in the gaps it leaves behind. Perhaps description and distinctivness should be folded together. But you see how tricky it is to come up with good criteria, and define them correctly as well?--Father Goose 05:47, 22 July 2007 (UTC)

Nonetheless, I don't mind playing around with the ideas. There are probably better ones hiding somewhere.--Father Goose 05:59, 22 July 2007 (UTC)

One of the reasons I like definition is it requires a real problem to be addressed. The "definition" paragraph is an opportunity to address a real, relevance-oriented problem I've seen in several articles.

I still need to comtemplate the rest of your response. Thanks! / edg ☺ ★ 06:13, 22 July 2007 (UTC)

I'll offer a parting shot thought before I go to sleep: I came up with the "three criteria" by going through a list of trivia (this one), deciding which I would personally keep and which I would toss, then trying to articulate to myself why I would keep or toss each. I then tried to combine my reasons into general criteria that adequately represented the "ins" and the "outs" along with "whys". I refined it further by studying at several other trivia sections and full articles on Wikipedia until I was satisfied it did a good job of separating the chaff from the wheat. The end result was a trio of criteria that were definitely not arbitrary.

"Impact", though not perfectly named or explained, represents an underlying principle reasonably well (how did this affect the subject?). I'm increasingly thinking I should have folded "fundamental" and "distinguishing" traits together, as distinguishing may fall under the notability subcriteron of fundamental anyway. Basic description and how did this affect might not be a bad one-two punch in the long run.--Father Goose 08:04, 22 July 2007 (UTC)

Good that you mention that. I was thinking fundamental and distinguishing were closely related, and may be part of the same thing.

I'm still using the term Definition instead of Basic description because

• It's more specific; a "definition" is more limited in its function, and much non-fundamental, non-distinquishing information is ruled out by that term. Whereas Basic description is more vague and debatable. It's not important that this information actually be a definition; however, defining is the quality of information we might consider fundamental.
• There are "definition"-related issues both brought up in WP:NOT#DICTIONARY and in real problems I've seen in articles, and these may relate to relevance. I'm guessing the issue I'm bringing up doesn't seem as hot to you as it does to me, but it both merits a mention and clarifies the above guideline.

The bang is small, but still it's a buck well-spent, IMO. I'll take a look at that Clinton article later. / edg ☺ ★ 08:22, 22 July 2007 (UTC)

"Defining traits", then. "Definition" is too closely bound with a "dictionary" meaning. Although I haven't witnessed the not-a-dictionary issues that you have encountered, I'm hoping they can be reduced to a sentence somewhere. I'd rather this guideline not become a list of special cases, although maybe, like the notability family, that will be unavoidable.--Father Goose 08:49, 22 July 2007 (UTC)

Version nomenclature

Version numbering is now broken and only makes sense to WikiLen (whose numbering incremented from 3.x to 4.1) & the Father Goose (whose number incremented from 2.2 to "FG 3.x", spanning existing numbers belonging to WikiLen). I had to scan several articles to figure out which version "3.x" referred to in comments above — it wasn't immediately obvious to me (and will not be for new editors joining this discussion) that FG meant "not in linear descent from 3.0".

I propose we say Father Goose's proposal for the current article (currently filed under Wikipedia:Relevance), and WikiLen's proposal for User:WikiLen/Relevance, regardless of where they are located. / edg ☺ ★ 06:13, 22 July 2007 (UTC)

Better still, I hope there will just be Wikipedia:Relevance from now on, with everyone contributing to that single version. The version numbers were useful shorthand for referring to specific major rewrites, but in the forked state, that became unworkable.--Father Goose 06:44, 22 July 2007 (UTC)

On second thought, I'd like to recommend Wikipedia:Relevance be renamed 27B stroke 6.--Father Goose 08:14, 22 July 2007 (UTC)

Definition

I've added these subsections

• Avoid dictionary-style definitions (which became a problem in a few articles, notably Misandry and Sex tourism), which I'm satisfied with
• Scope, which needs clarification (the very term scope is not familiar to some readers) and may need much more work.

I think these subsections contain more-or-less actionable advice and address real problems I have seen. (If you want the dreary case histories, ask me.)

I'm wondering if the existing proposals can be organized into my proposed demarcation, but that may be the fevered dream of a madman. / edg ☺ ★ 06:13, 22 July 2007 (UTC)

Consensus is not happening

Father Goose fork is here or sometimes at the project page.

WikiLen fork is here or sometimes at the project page.

This method of alternating which fork is in the project space to see if one gets traction is not working at this stage to achieve consensus. For example, it appears only the editors favoring the command & control approach for policy are doing the edits at the Father Goose fork. Furthermore, Father Goose's fork is currently suffering from rule creep — the wrong direction to go for consensus. All we are doing is polishing the two forks. There are hard issues to work on that only consensus can solve. —WikiLen 09:11, 22 July 2007 (UTC)

It appears only the editors favoring the command & control approach for policy are doing the edits

This must refer to me. Can please I finish my work before it gets rejected outright on political grounds? / edg ☺ ★ 09:39, 22 July 2007 (UTC)

Hard issues are being ignored

Obvious hard issues:

• What to do about rule creep concerns?
• What to do about guideline-not-needed concerns?
• What to do about concerns that attempts to "establish relevance" won't help edit disputes?

To these I add:

• Should rules that only help resolve relevance in Trivia or In-popular-culture be here or in Trivia policy?
• Should this 'guideline' help defeat bogus-relevancy arguments? —WikiLen 09:11, 22 July 2007 (UTC)

Late additions:

• Should we have a minimalist version that avoids contentious issues? perhaps in the tradition of WP:IAR? —WikiLen 20:05, 22 July 2007 (UTC)
• Is "Relevancy emerges" a paradigm to employ for this guideline? —WikiLen 20:05, 22 July 2007 (UTC)
• Should we have instructions to prevent impassioned editors from hacking scope to create or destroy relevance? —WikiLen 20:35, 22 July 2007 (UTC)

Interesting issues to work on. —WikiLen 09:11, 22 July 2007 (UTC)

What to do about rule creep concerns?
What to do about guideline-not-needed concerns?

The best solution to rule-creep concerns is to not make this guideline. This advice has been given several times by various editors on this Talk page. Since we've abandoned that advice, the article should be concise and actionable. That's what I'm working toward at this time.

What to do about concerns that attempts to "establish relevance" won't help edit disputes?

I don't share your opinion that nothing can help establish relevance. And if disputes are not settled by this guideline, we should request page deletion.

Should rules that only help resolve relevance in Trivia or In-popular-culture be here or in Trivia policy?

It's not an either/or, but a guideline on relevance should speak to trivia, so let's try it here.

Should this 'guideline' help defeat bogus-relevancy arguments?

That would be a sensible goal for a guideline. Can you list particular arguments that are worrisome? I feel that Avoid dictionary-style definitions addresses one such concern (in addition to clarifying scope and heading off a possible misreading). / edg ☺ ★ 09:39, 22 July 2007 (UTC)

To achieve consensus the "hard issues" need to be addressed one at a time. Let's get consensus to address these before we actually start jumping into them. There are so many — probably some I've missed. An order approach is needed. Your replies above would make a good starting off point. But anyway, it looks like we are on Plan B instead for the moment. —WikiLen 16:38, 22 July 2007 (UTC)

Regarding "bogus-relevancy arguments": I agree, "Avoid dictionary-style definitions" does address this concern. See also this section from WikiLen fork and this essay, What claims of relevance are false (linked to in the WikiLen fork). —WikiLen 16:38, 22 July 2007 (UTC)

As I mention below, the issue of Editors Too Impassioned To Have Perspective is not addressed.

I think Important and Compelling argument are arguments for inclusion rather than against, so precautions to zealous editors should probably take a less confusing form (than currently stated in the WikiLen fork, and its spinoff essays). Since these admonitions do not apply to all editors, they may be better served by a precautionary essay than within an ordered and logical (intended) guideline, so the spinoff essays may be the start of something good.

As an aside, the mention of "feng shui" and "vision boards" in your examples is very distracting. Users who don't know what those things are will wonder if they are missing something. Users who do know will consider feng shui both pretentious and cliched as a "random thing used for example". / edg ☺ ★ 16:56, 22 July 2007 (UTC)

Agreed. Needs a better example than "feng shui" and "vision boards." —WikiLen 20:00, 22 July 2007 (UTC)

Regarding "the issue of Editors Too Impassioned To Have Perspective is not addressed": Got it, I think. I have added the "scope" issue to the list of issues. —WikiLen 20:40, 22 July 2007 (UTC)

Late additions

Strive for a minimalist version perhaps in the tradition of WP:IAR?

There exists no "minimalist tradition of WP:IAR" in writing guidelines. You've brought WP:IAR up several times and I don't believe anyone has ever agreed this was anything but counter-productive. A guideline should not ignore itself, or other guidelines. If you want to ignore all rules, go ahead, be free and freaky; but you logically opt out of co-authoring guidelines.
This article has been in revision since May, and no one has come up with a satisfying "minimalist" concept. Unless you are entirely satisfied with no rules ... tear down everything ... entropy is the highest order ... inertia, stalemate, accusation and counteraccusation, these are better than any guideline, you might want to let go of your minimalist ideal.

Is "Relevancy emerges" a paradigm to employ for this guideline?

Every version (other than reverting to empty versions) has contained an equivalent to There are no general rules for establishing relevance in all cases, so often it is determined by consensus on what is likely to be useful to readers. I don't see a need to dwell on this. However, you seem to use this issue to recommend a policy of (in effect) slow edit warring and obstruction. No idea what motivates this, but it seems highly counter-productive. I recall Father Goose calling this tendency "self-immolation".

Honestly, some of your "Hard issues" seem like pointless roadblocks. I don't see the points at all, and I feel like they have been brought up repeatly, and responded to already. / edg ☺ ★ 20:36, 22 July 2007 (UTC)

Need guidelines to prevent impassioned editors from hacking scope to create or destroy relevance?

That question is not a sentence. Are you proposing these are needed, or questioning the need? / edg ☺ ★ 20:41, 22 July 2007 (UTC)

Does Relevancy need instructions to prevent impassioned editors from hacking scope to create or destroy relevance?

If scope is identified as a resource for determining relevance, then yes the guideline not to make an artificial scope is a good idea. I think I have explained this several times on this Talk page, plus added a hidden comment in the article so the reason for this instruction is understood to future editors. / edg ☺ ★ 20:50, 22 July 2007 (UTC)

Edgarde, I find this frustrating. You are in some contextual misunderstanding that I can't figure out. I am looking forward to clarification from you. FYI: I am just trying to list the issues behind the fact that there are two forks. We can debate these issues later. Also, when I list something it does not mean I am expressing a position on it or even think it is an important issue. It just means I am predicting it needs consensus work of some kind. —WikiLen 22:01, 22 July 2007 (UTC)

By the way, there is obvously a misunderstanding with my reference to "minimalist" in the WP:IAR tradition — I will fix it. The tradition I see: short is elegant and effective. My intent by the reference was to keep alive consideration of your "minimal" version. The misunderstanding is my fault — thinking on how to fix it for future readers. —WikiLen 22:01, 22 July 2007 (UTC)

Regarding "responded to already": Yes, I know you have already responded to many of these "hard issues", probably all of them. That I list them does not mean I think your resolution of them is flawed. I would be grateful if you would tell me of any I am missing. This includes issues we have already resolved at this talk page and simply need to be communicated—not lost—in the consensus work that lies ahead. —WikiLen 22:01, 22 July 2007 (UTC)

Time to end the dancing forks

We all know (or should know) neither fork is going to get traction. I recommend the Father Goose fork continue in the project space until it stabilizes and then we revert back to the pre-proposal version. My fork is polished enough for consensus purposes—thanks to help from others—so we would then be ready to work towards consensus. A straw poll at that point might be useful. Or plan B: the Father Goose fork can just continue on until it is rejected and then I could post my fork, until it also gets rejected — no wait, mine is brilliant, if we give it a chance consensus will happen... :) —WikiLen 09:11, 22 July 2007 (UTC)

Great. Let's go with Plan B: work on Goose fork, then seek approval. / edg ☺ ★ 09:39, 22 July 2007 (UTC)

I favor going to consensus mode now — was tying to dramatize a point through humor (failed) about us being too obsessed with thinking a particular version is "the one". I suspect the Father Goose fork is getting less stable. —WikiLen 18:55, 22 July 2007 (UTC)

Yeah sarcasm doesn't work as well as WP:SARCASM says. So, in a straightforward way, what are you really proposing? I dunno what "consensus mode" means in the current context.

I suspect the Father Goose fork is getting less stable.

Dunno about that. The Father may revert a lot of my changes, in which case, it would be more stable than it currently appears. / edg ☺ ★ 19:01, 22 July 2007 (UTC)

I propose we revert back to the pre-proposal version and then discuss the hard issues, one at a time, using guidelines of Wikipedia:Consensus as we find appropriate. I suggest we also have a to-do list of these issues-in-contention, for the top of this page; addressing them one-by-one until each one has achieved consensus or until something pops and we realize we have a consensus. I think it especially important we address issues one at a time — not sure in what order.— Preceding unsigned comment added by WikiLen (talk • contribs)

I've been unsure how to respond to your protests from the last few days, but I'll offer what I can now. What's happened over the last few days could be interpreted as a consensus emerging in favor of my version -- two independent editors have thrown their weight behind it, and as yet, none have done the same for yours.

To characterize this as "consensus is not happening" because your views are not prevailing is not constructive. To be honest, I'm still fairly confused by your views, even though you've stated them several times in different forms. I believe I am not the only one confused by them. As long as we remain confused, we will also remain unconvinced that your ideas represent the correct path on which to proceed.--Father Goose 08:55, 23 July 2007 (UTC)

Misunderstanding. I (WikiLen) don't mean consensus with what is represented on my fork. What I am suggesting is consensus between the 3 editors working on the Father Goose fork (in order of initial participation: Father Goose, Newbyguesses, Edgarde) is not a consensus on relevancy. What is needed is consensus amoung the much larger set of editors that have been participating here which could include me or not. It is that larger consensus I am saying is not happening. You have filtered out opposing voices. This happened because everyone is letting the three of you work together to see what you come up with mostly because the three of you want that chance — nothing more. What you have come up with is a less stable version; one that has rule creep; and one that is moving away from ideas such as this:

The editors at each page decide by discussion among themselves what is or is not worth having at each article. For each topic, there's a different set of criteria. — User:Coppertwig

Regarding the WikiLen fork, I am only interested in receiving comments on it — comments from the larger set of editors that have visited this project. I want these comments to contribute to the final consensus discussions. I don't expect the views expressed in my fork to prevail. I expect our mission to achieve consensus to succeed. —WikiLen 19:13, 24 July 2007 (UTC)

The Father Goose fork has been in the main project space too long for something that didn't belong there in the first place. I refer you to: "Unapproved drafts should not be offered as recommended guidelines." —quoting from Edgarde in earlier discussions. Options:

• OPTION A: Move this work on the Father Goose fork back to the draft workspace and keep on working on it there. When you get done we could put it back into the main project space and this time call for larger set of editors to comment on it.
• OPTION B: Declare dancing forks is not working; bring in the larger set of editors; and start working on consensus.
• OPTION C: Put the WikiLen fork in the main project space and immediately ask for comments from larger set of editors. After three days take it out.

I support any of the three options. —WikiLen 19:13, 24 July 2007 (UTC)

It's my impression Option B is in force right now, at least the way you phrased it. My draft still needs work, both to better embody consensus, and to improve in general, but right now it is the "favored suitor", and it is being used as the baseline for continuing work. Coppertwig's views seem to be the most closely aligned with yours of those editors who have commented so far, but he hasn't come out in support of your draft. He did make some criticisms of my current draft, at least some of which I think I can fix.

Forking, ultimately, is not the right way to "achieve consensus": see Wikipedia:Content forking. That guideline is specific to articles, but the principle is the same here. At this point I'd recommend you focus more on convincing others (maybe even me) of the specific flaws of the current draft at WP:REL. You've made points in the past that I've agreed with. You could also edit the current draft to fix those flaws, although if others revert your changes and no one else reverts them back, you will need to continue trying to convince others.

Edgarde's quote is doubly without teeth: it was made in reference to an earlier draft, and he has since come out in support of the current draft. His statement wasn't "right" in the first place, as the only way to get support for a proposal is to propose it. I'd say his comments from then were short-sighted.

Finally, I'm in the midst of changing, and possibly abandoning, the "three questions" anyway. These changes, which will hopefully be ready this week, might come closer to meeting with your approval. So, our work continues.--Father Goose 20:52, 24 July 2007 (UTC)

[shifting thread left] —WikiLen 23:32, 24 July 2007 (UTC)

Dancing forks ending (not working)

Father Goose, OPTION B it is, but not because "it is the "favored suitor." But because you are willing to have the larger set of editors comment, edit, or revert on it. Without that willingness this would need to go back into draft namespace. By the way, how do you get "favored suitor?" With only four people involved out of a much larger set it seems a delusional stretch to suggest consensus. Furthermore, the consensus people were looking for was between you and I. We have failed at that. I am stopping the dancing forks by saying I will give up claims to "my turn" after your fork gets rejected, because someone has to put a stop to the "it's my turn" nonsense (nonsense because it asks for opposing editors to leave the my-turn fork alone). —WikiLen 23:32, 24 July 2007 (UTC)

Someone needs to contact all the relevant editors. I am probably the most motivated in that regard so I will do it unless someone prefers otherwise — I will give you time to object before moving forward. I think it appropriate to encourage comment on the WikiLen fork also which I will do. I will maintain a NPOV on calling for participation and indicate that we are ending the going back and forth between forks, without any prejudice as to which is the better fork, but with the Father Goose fork in the main namespace. —WikiLen 23:32, 24 July 2007 (UTC)

The version on the project page currently represents substantially work done by Father Goose, over a number of revised drafts. If any of the material added over the weekend could be of use, it may get incorporated. Looking at WikiLen's latest draft, my impression is that is has been improved, and shortened, during its sojourn in userspace. How to amalgamate ideas from the competing drafts, is not wholly clear to me; the work seems to proceed on somewhat disparate lines.

Of the competing versions, my suggestion is that Father Goose's version has the most prospect. This may be because of the refining which has transformed it over a period if time. My reason for not preferring WikiLen's version is that, to me, it flips between general and specific considerations, giving an impression of vagueness. Sorry not to be more specific, my grasp of where all this is going, I have already admitted, is somewhat insecure, also, style considerations are obviously subjective.

As of June 15, when work on this page commenced, there was a flare-up over BOT-tagging. Well, it stopped. WP:ATSIA is now WP:ATS, (with some changes to text), and any previous attemps to merge this page with WP:ATS (some of which I participated in) have been rightly resisted. Fr. Goose in the latest version has identified correctly the point of contact with "trivia", it takes but a phrase - ("Trivia" lists, which should be avoided). Developing a separate guideline on Relevance thus remains a useful goal, which can best proceed from the current, (Fr Goose) version, as I see it. Newbyguesses - Talk 23:56, 24 July 2007 (UTC)

Good to get your opinion... I like your observation that the two forks "seem to proceed on somewhat disparate lines." This speaks to why we failed to merge them into a single proposal. Lastly, I too find "a separate guideline on Relevance ... remains a useful goal." —WikiLen 03:38, 25 July 2007 (UTC)

Well, our two proposals did move toward each other quite a bit over the course of several revisions. You convinced me that "relevance to subtopics" was an error, and I adopted your emphasis on the uncontroversial "about the subject of the article". Your current draft says similar things to mine, but omits certain parts.

If I can figure out an effective replacement for the "three questions" approach, it's conceivable you might eventually come to embrace my version.--Father Goose 05:08, 25 July 2007 (UTC)

Good points. In my call for the larger-set-of-editors to participate I will represent that "our two proposals did move toward each other." —WikiLen 12:00, 25 July 2007 (UTC)

Versions mine

The minimal

What's wrong with this version?

I can tell you what's right:

• simple, easy to follow
• imposes least reading on editors
• I think everything included in this version can be agreed upon.

What's missing:

• answers to several specific questions posed by Father Goose and WikiLen. However, I'm not confident with the solutions offered in either proposal (including the "three questions").

2 basic instructions:

• Only add content about the subject of the article
• Split articles (per WP:SUMMARY) when relevant sub-sections grow to the point of undue weight, or they simply get too long.

Abuse prevention: Avoid making an explicit statement of scope is included to prevent article lead from being bias-gamed with prohibitive scope statements. Subject description should (in conjunction with other Wikipedia rules) allow all relevant content, and prohibit irrelevant content.

There may be room to grow (especially in the oddly controversial field of Establishing relevance). / edg ☺ ★ 12:03, 22 July 2007 (UTC)

I find Edgarde's minimal version deserves serious consideration. Addressing how to do the lead section should be dropped — off-topic. I would keep the "Article length contraints" section in. The "Summary style" section seems to have the wrong heading. Maybe "Length of subtopics" is appropriate. The Rwandan genocide makes a better example than the Internet. I would revise the "Scope" section to be: (WikiLen)

Material added to an article should be about the subject of the article.

Information added to articles on very general subjects should address the entire subject, rather than meandering into related topics for which more specific articles exist (or should exist). Articles on very specific subjects will provide room for far greater detail.

Example

The Rwandan genocide is relevant to History of Africa, an enormous topic. Article length contraints limit the genocide to a single concise sentence in History of Africa. However, it is accorded several paragraphs in History of Rwanda, and a comprehensive treament under Rwandan genocide. In turn, several topics under Rwandan genocide have their own detailed articles.

This "minimal" version has a greater chance for consensus than the Father Goose fork and perhaps slightly less chance than the WikiLen fork (in my biased opinion). This minimalist approach seems bland and missing the mark compared to the WikiLen fork. In particular, the "Relevancy emerges" concept in the WikiLen fork may succeed at addressing the concerns of the "guideline-not-needed" camp; a concern not addressed by this "minimal" version. —WikiLen 15:44, 22 July 2007 (UTC)

Addressing how to do the lead section should be dropped — off-topic.

If this refers to my Scope instructions, this isn't how to do the lead section — it's how to read the lead section. The instruction to not create an explicit scope is to prevent impassioned editors from hacking scope to create or destroy relevance. I think this is needed.

The Rwandan genocide makes a better example than the Internet.

I think this example is too complicated for the article lead. Without the Article size contraints instruction, this progression or Rwanda examples emerges mystically, without explanation. I would recommend refining the Internet example, perhaps by adding one concrete example of something to exclude. My interest is waning, so I will leave that for now to another editor. / edg ☺ ★ 16:15, 22 July 2007 (UTC)

Short review of "the minimal" by Father Goose

It doesn't say much, which isn't necessarily a bad thing, but what little it does say doesn't flow. Too many short paragraphs, ideas presented in the wrong order, and it's still got redundancies -- "overview of the subject/details not directly relevant to the primary topic should be moved" as found in my most recent version is better than "general subjects should address the entire subject/specific subjects will provide room for far greater detail", which is a paraphrase of an earlier draft of mine. Both appear in "the minimal" version, which is unnecessary.

As regards the "long version", I've been watching edgarde's changes over the past day, and I've seen some ideas and edits I like, although I think the proposal's structure has become disjointed. I'll do what I can to blend together the best ideas from my work and edgarde's, as soon as my brain comes back from siesta.--Father Goose 23:12, 22 July 2007 (UTC)

I wasn't highly satisfied with the final product myself, especially the long version. Revise as it pleases. I think I've explained most of what I put in. I don't feel strongly that my changes need to be preserved. / edg ☺ ★ 06:38, 23 July 2007 (UTC)

I've been pondering the various changes you made in the long version. I agree with some of them; others I don't see as an improvement; still others take things in an interesting direction but are still not quite right. Rather than leave the page in an experimental state, I think I'll revert to my most recent version, which is less of a "rough draft" than yours, and take a few days figuring out how to blend together your material and mine.

I think I'm starting to see how to phrase the "three questions" better -- perhaps even to abandon them in place of easier-to-understand underlying principles.--Father Goose 08:22, 23 July 2007 (UTC)

Cool. / edg ☺ ★ 08:39, 23 July 2007 (UTC)

Details arranged by type of information

Basicly the length of this article is tied to the amount of examples included under Establishing relevance.

The big version I came up is basicly The minimal plus 4.5kb of Establishing sorted by type of information.[6]

Guidelines:

• Definition (could be called defining information) — always worth having.
• Detail – decided by editors, with help from relevant WikiProjects. No good overall guideline was found in either Proposal, or talk page.
• Context – the "Three questions" seemed most relevant here, and may provide good limits; otherwise, instruction is to offlink as much as possible.
• plus a few bonus guidelines, under the info type they seem most applicable.

Advantages:

• clear guideline on definition, IMO
• decent suggestions on other information types, and refers to appropriate guidance where available
• flexibility: any good instruction can (consensus pending) be included in this article without needing to cover everything

Disadvantages:

• probably not the last word
• optimised for instruction creep
• irrational nut behind the keyboard errors (i.e. pitfalls) are not addressed
• failure to achieve desired conciseness. See The minimal.
• also less likely to be approved than The minimal.

I also expanded Article length contraints. A minimal version can easily be made with this update by simply deleting the entire Establishing relevance section.

Okay I see people are waking up. I'm quitting for the time being. Other editors are encouraged to edit, revert and villify as needed. / edg ☺ ★ 16:04, 22 July 2007 (UTC)

I think I'd prefer it tho if everyone got a chance to look at the long version before it was drastically revised. Maybe I'm overrating my contributions. / edg ☺ ★ 16:17, 22 July 2007 (UTC)

The moon is made of green cheese

The moon's equatorial circumference is 10,921 km.

*laughs* You should be aware that that specific number was added by yours truly and is of uncertain veracity: Talk:Moon#Equatorial circumference. Wikipedia is an incestuous little place sometimes.

One thing I learned after my first draft was to avoid examples wherever possible. I'll probably tear this one out for that reason.--Father Goose 22:38, 22 July 2007 (UTC)