User talk:Quiddity/Navigational pages RfC

Give thanks

Thanks to Quiddity for setting up this page. --Gavin Collins (talk|contribs) 09:53, 23 August 2010 (UTC)

Discussion

Of the currently suggested solutions, I'd support 1, or 2, as the most beneficial/practical.

For Indexes alone, I'd hesitantly support option 5 (move to wikiproject space).

Possibly there are further alternative solutions? -- Quiddity (talk) 21:18, 26 February 2010 (UTC)

Unrelated

I completely oppose the moving / renaming / refocussing of this RfC from outlines to navigation in general. This is not what was agreed on, and gives a completely undeserved legitimacy to outlines. Please revert. Verbal chat 20:37, 10 July 2010 (UTC)

This is a completely separate RfC [separate from the RfC at User_talk:Karanacs/Outline_RfC_draft. added August 23], that I've been working on for months. Perhaps I should move it to a subpage of my username, to avoid confusion? -- Quiddity (talk) 20:58, 10 July 2010 (UTC)

Ok, that would be a good move. You should probably remove outlines from this RfC (not sure if they are mentioned, I'm very tired) and I don't understand the redirect to here from the outline RfC. If/when this goes ahead, it should be separate (including timing) to the outlines RfC). Best, Verbal chat 21:04, 10 July 2010 (UTC)

It makes the eyes glaze over

Quiddity, you have a tendency to overcomplicate things. The discussion has such a large scope and so many threads that it would take years to iron out the matrix of issues you've presented. It is so convoluted that only die-hard editors would or could wade through it and make any sense of it. Only a handful of individuals on Wikipedia are fluent in the overall navigational system.

You've listed 3 problems applied to around 9 page types = 27 topics of discussion. (I see Categories, Books, and Templates lumped in there too).

You've positioned the RfC for failure from the start. With such a web of problems, you'll scare most editors away. What you'll wind up with is a conversation between 3 to 6 persons, perhaps enough to create a proposal for the community to vote on. Or perhaps a discussion in which you are endlessly explaining things to people or correcting their misconceptions. And because you've included outlines, you'll have to deal with the anti-outline rhetoric of User:Verbal in addition to everything else.  :) Have fun.

Proposals that affect all of the navigation page types are doomed to failure, because you've included far too many factions, each with their own special interests.

Will you develop a proposal to move them all out of article space? Ratification won't happen. Can't happen - there will be far too much opposition.

Will you wind up proposing some big plan that treats each of the 9 file types with respect to the 3 problems for each? Every editor will likely object to some element of such a comprehensive plan.

See the KISS principle.

Even if your reason is to establish a precedent against a particular action, such as moving them to another namespace, by virtue of there being a failed proposal, it won't make any difference, because the convoluted plan will be unlikely to fail on any given point (out of so many), but will instead fail due to general lack of support (because it will be pushing so many angles at once). Therefore, no specific precedents (pertaining to specific file types) will be derivable from the outcome.

My advice is to pick the problem apart, and deal with each facet individually - on separate discussion pages.

Good luck.

The Transhumanist 00:06, 3 August 2010 (UTC)

Aha! You see the trick is, I don't have an agenda!

I'm laying out all the interrelated problems, along with the various solutions that I've seen suggested, and hoping that people will see patterns from their own perspective, and form new ideas. I believe a number of the current solutions are logical, some are feasible, and some should be encouraged. I hope to learn of additional possibilities, and obtain (and spread) fresh understandings. Yes, it's complicated. Good morning :)

Possibly I should add this statement as a note at the top?! -- and, talk about pot kettle tl;dr! heehee! Well explained though. :) -- Quiddity (talk) 07:37, 3 August 2010 (UTC)

There is no question that the matter is complicated. However, a Content namespace, which could include all topics which might qualify as "non-article", might I think be creatable. The various pages within that space could all be titles according to whatever type they are, and they all could, presumably, be alphabetized by the name of the topic (Japan, seafood, earthworms, whatever), so that they would be in proximity to each other in the list of the contents of that namespace. John Carter (talk) 20:13, 12 August 2010 (UTC)

Wikipedia already has a "non-article" namespace, and that is inhabited by categories. The good thing about categories is that they are an aid to navigation by design, and because they are not subject to the notablity guideline,their inclusion in Wikipedia is based purely on subjective criteria, e.g. WP:USEFUL. --Gavin Collins (talk|contribs) 09:52, 23 August 2010 (UTC)

Well, if Quiddity's presentation is overly complex, maybe it's because this is an overly complex problem. Wikipedia articles don't fall 100% into any one type of article: there are, & always will be, a small number which defy simple pigeonholing. That's why we've had a lot of debates over issues like this. -- llywrch (talk) 02:06, 24 August 2010 (UTC)

Navigational lists

I would assert that the types of page that could be called "navigational" are really link farms. I have had a look at the definition of linkfarm at WP:NOT, but the defintion seems puzzling, and I have raised this issue at WT:NOT. Do you think discussing this together with the Problems for discussion is appropriate, or is list inclusion (inclusion/notability) and list exclusion (content/WP:NOT) seperate issues? My view is that they different sides of the same coin, and discussing both together is a good idea. --Gavin Collins (talk|contribs) 12:38, 24 August 2010 (UTC)

I've replied at Wikipedia talk:What Wikipedia is not#Linkfarm regarding some of this.

In general, I believe all these page types fulfill a need, and that we need to determine some way to incorporate them "officially" in our rulebook, partially to prevent editor conflict, and partially so that the readers who use them can continue to do so.

There are really only 2 solutions, as detailed in the "Solutions" section of this RfC: 1) Mark them in some way visually, in the same way we treat disambig pages. (and establish the pages describing them as "guidelines"). 2) Move some or all of them out of mainspace (with many options, depending on the pagetype being considered). I've given my thoughts on those solutions above, at #Discussion. HTH. -- Quiddity (talk) 19:51, 24 August 2010 (UTC)

I will agree with you so long as they aren't in left in mainspace, where they currently fall foul of Wikipedia's content policies or inclusion guidelines. --Gavin Collins (talk|contribs) 22:18, 24 August 2010 (UTC)

"Lists of lists" are the least complicated page-type (imho), so, what exactly would you suggest is the ideal solution, for those pages specifically?

(E.g. Lists of people, Lists of mathematics topics, Lists of hospitals, Lists of languages, Lists of countries and territories)

And, if your ideal solution involves removing them from mainspace, why does that not also apply to disambig pages? (Or are you one of the few editors who has previously suggested that we need a new namespace for disambig?). Thanks :) -- Quiddity (talk) 02:03, 25 August 2010 (UTC)

Lists of lists are based on categorisation, rather than on verifiable definitions, so theoretically it would logical to delete them on the grounds that they are link farms. However, the list owners are unlikely to accept this prognosis, and will fight this tooth and nail. Disambiguation pages are a grey area, and I am not familiar with the ongoing debate, but it seems to me that non-mainspace page would be ideal, because I would argue that the disambiguation is a process of categorisation. --Gavin Collins (talk|contribs) 04:04, 25 August 2010 (UTC)

Disambig:

There is not an ongoing debate about disambig pages. It's just an infrequent "perennial" proposal, to give them their own namespace. It never goes anywhere. The last discussion was at Wikipedia:Village pump (technical)/Archive 33#Disambiguation: namespace in April 2008. The other main discussion is at Wikipedia talk:WikiProject Disambiguation/archive3#Disambiguation name space from 2005/2006. (afaik)

Categories tangent:

There are a few questions I ask all my colleagues and acquaintances, about their interactions with Wikipedia. Among them are "Do you use the navboxes?" and "Do you use the categories?" (if I don't have a browser nearby, I explain what they look like and where they are generally located). The vast majority say "No" to both. Many are surprised when I click [show] in a navbox. — However, most respondents recognize the infobox, when it is pointed out or described.

Readers don't tend to use our categories. I have no hard data to back-up this assertion! (And it's a gross generalization, but I think you'll understand what I mean).

Essence:

1) I would humbly suggest that you are a namespace idealist, and are surrounded by namespace pragmatists.

2) I would set that suggestion alongside a core issue, that is a pattern I've run into a few times: You are arguing on the basis of "what categories should be, if they worked as well as they could". I agree that once the software/interface catches up to that vision, there are many old ideas that will become viable (eg category intersection). But currently, categories are a muddled zone of confusion and incompleteness – hard to browse; hard to search through; unwatchlistable so hard to maintain; and either too exhaustive or too divided (taxonomies vs folksonomies, etc). Hence WP:CLN, and the current practices of prolific list and navbox creation.

All in my opinion, and full of self-doubt, but that's what I'm currently thinking. (With many tangents deleted, for brevity. (!)) HTH. -- Quiddity (talk) 06:12, 25 August 2010 (UTC)

I agree entirely, your analysis is very preceptive. However, this still does not resolve the issue that main-space should be reserved for topics that are based on verifiable content, not self-referencing links. If there are issues with categories, such as a lack of indices, then that is what needs to be addressed. If you are suggesting that content policy should be watered down to accommodate technical problems with MediaWiki, I would say that is an example of the tail wagging the dog. --Gavin Collins (talk|contribs) 07:26, 25 August 2010 (UTC)

Thank you :)

Regarding "should be watered down", I would answer that with a meta-reference to "policy is descriptive, not prescriptive" (which has, of course, been edited out of its place in WP:NOT... (discussion here)). Idiomatically, Cart before the horse.

If we can accommodate disambigs (which we do), then we can accommodate lists of lists, and some or all of the others, possibly using the same method of "visual features to distinguish the page as "not an article"". (On a related note, see this week's Template talk:Disambig#Delete this template...). And nothing is forever, so perhaps it's worth leaving practices in their current imperfect-but-consensus state, until a more perfect solution is actually possible, and then discussing it fully. Policy can accommodate future possibilities of change.

I tried a few experiments with {{category tree all}} last night. Afaik, that's the state-of-the-art for category browsing (and is occasionally disabled for resource overusage (?)). Might be something worth poking at. -- Quiddity (talk) 19:01, 25 August 2010 (UTC)

Gavin, the problem with your desire for clean borders between name-spaces is that our users don't think in these precise groupings. And the less a user knows about a given subject, the less precise their thinking is. For example, this morning I was looking to see if we had an article about "Goddam" as the nickname for the British; now what I did was to start typing "Goddam" into the go/Search box & watch what options I was returned, & the one selected happened to be a disambiguation page which directed me to the article I wanted -- Les goddams. Now I wouldn't have found that page if we followed your rigid handling of name-spaces: how would I know Disambig:Goddam exist, if I were a casual user of Wikipedia? (Hell, I'm one of the most experienced users of Wikipedia, & I'm often surprised at how consensus has decided content should be presented.) We should put the disambiguation pages & the list pages where users expect to find them, which is known as intuitive design. -- llywrch (talk) 19:51, 25 August 2010 (UTC)

Llywrch, the problem with your approach is that you need to think out of the box :p

At the heart of this issue is the key problem with categories, navigational lists and disambiguation pages: they are not searchable. My guess is that the only way to meet the need for searchable navigation pages is a Special page that supports advanced searches, by which I mean it provides the reader with the abiity to search, say, a category (including its sub categories) and obtain a complete list of all of its members in an intuative fashion. I don't understand why this has not been done already (maybe it has, but I am not aware of such a facitlity), because this would be an effective tool that would make navigational lists redundant. Such search factilities are common place in the real world, a good example of which is a Help files that are shipped with most software products. --Gavin Collins (talk|contribs) 10:40, 26 August 2010 (UTC)

No amount of technical advance will make navigational pages redundant

My two cents. They just offer so much more flexibility in terms of arrangment and organization of information, in a "human" form. "Search" is fine if you know what you're looking for, but actually it's missing the point. There are very many ways in which people navigate and searching is just one of them. Similarly for categories, which work in an entirely different way to navigational pages (not in humanly arrangable order; nav pages organize links while categories they actually organize articles; categories are very good for "eyeball search" and the display of all resources on a topic, while lists can highlight key resources; the lack of descriptions possible in a category*...) Personally I can't see the difference between the "index" and "outlines" systems and think the two ought to be brought together somehow.

In my opinion the big divides are (a) 'between "verifiable/objective" and "subjective/conceptual/creative" arrangements. The index and outline systems depend on editorial judgment to form what they feel is a conceptually coherent map of the most important articles in Wikipedia for a reader to introduce themselves to the topic with. However, while timelines serve a purpose as navigational pages in the sense that they can provide a series of links to various articles in an organized manner, that order is predertermined by the dates involved. Similarly something like "List of town in Idaho" or "List of birds of Australia" is objective; the list is useful for navigational purposes, but it's also (in the language of the 2002 "what is an article" discussion), almanac-like and editorial discretion is limited. Further, I would argue such lists are (b) lists of things not lists of articles - whereas the index/outline articles, as well as portals, are lists of articles not of things. This point is perhaps subtler, more subjective and maybe less useful than (a). But I think (a) and (b) taken together get to the heart of the problem. My feeling is that something that clearly i.e. under criteria (a) and (b) is "navigational" not "content", probably belongs outside the mainspace. It's a way of navigating the encyclopedia (like the index, micropedia or outline of knowledge in a physical encyclopedia) not part of the encyclopedia itself. I think Contents: or Index: is a possible namespace, or possibly as subpages or collapsed sections of a Portal: page.

(*By the way, I wrote something about the history of navigation pages, and also the extended means of navigation on other projects such as CZ - including their "definition" system, which comes close to producing a "Micropedia" effect, and which would allow annotation of automatically generated lists of articles such as those found from searches or from category contents/intersections. Do you think I should copy and paste, Quiddity? I know you felt they were fairly useful, and I don't think they're redundant to anything on this page so far.) TheGrappler (talk) 20:24, 27 August 2010 (UTC)

Categories vs. Lists

This is my understanding of how categories and lists differ. Is my thinking realistic? --Gavin Collins (talk|contribs) 14:44, 21 September 2010 (UTC)

Not realistic, but idealistic.

See again my comments to you regarding categories in the thread further above, and also see my reply to TheGrappler at Category talk:Fundamental categories#Gah!. Categories are currently an imperfect implementation, and navigational lists are currently an imperfect implementation. They each do different things well. However, Readers are much more likely to find and use lists.

See further experiments with categorytree at User:Quiddity/sandbox3. It's arguably better than our current default interface for categories, but lists are still much better at some jobs.

(Lists can be structured (categories are always alphabetical), lists can exclude non-core items (categories are always exhaustive listings), lists can include redlinks, lists can be searched, lists usually appear all on one page, etc). Lists only major drawback is that they have to be manually updated.

HTH. -- Quiddity (talk) 19:03, 21 September 2010 (UTC)

Perahps I was not explaining myself quite as clearly as I hoped. The difference between "descriptive" categories and "proscriptive" lists is what I was trying to differentiate between, so I have taken a second shot[1].

I think you are right about categories being a poor form of navigation, but I still hold that lists in article mainspace should not be used or constructed in the same ways as categories. Because lists are normative groups, compared with descriptive nature of categories, they should be attributed in the same way as any statement of opinion - directly to a particular published source, rather than by mixing up sources to construct a synthesis. Again, I would value your opinion. --Gavin Collins (talk|contribs) 22:19, 21 September 2010 (UTC)

Finally, I think I have nailed down my thoughts - see Wikipedia talk:Requests for comment/Inclusion criteria for Lists#Lists & Notability: Lessons learnt from the RFC. --Gavin Collins (talk|contribs) 05:14, 26 September 2010 (UTC)