Wikipedia talk:Categories, lists, and navigation templates/Archive 4

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

• Guidelines can be changed without prior discussion; only major and/or other questionable changes need to be on the talk page first;
• Re. "categories Cannot include alternative names for the same item + Cannot be organized into sections on a single page." - no major change, just extending pros-and-cons lists with factional, undoubted info, please proceed with these change proposals! These are typically changes that don't need to re-gear the guideline to "proposed"; if you have doubt whether these changes are acceptable, list them on this talk page: if nobody objects (and I don't see why anyone would object), proceed with the change.

--Francis Schonken 10:42, 3 October 2005 (UTC)

Per previous objections, ie [1], I have removed the requirement that Article Series Boxes be subject to a single linear ordering, a requirement violated by many existing navigational templates. I have also toned down other parts of that section to reflect what I believe is common practice. Dragons flight 13:02, 3 October 2005 (UTC)

Thanks for the updates! Changing expressions from "almost always" to "often" and the like, and updating the guideline to current practice is none of it "major change" IMHO.

The misconception (please read guidelines re. the writing of guidelines, or -at least- the text of the guideline boxes) is that a guideline does not have to go back to "proposed" for any of these changes. It doesn't even have to go back to "proposed" for a major update: the major update proposal would be discussed on this talk page, and implemented when there's consensus. During the process of trying to reach consensus over a new proposal, the guideline remains what it is: a guideline.

If you want to attract more people to have a look at this guideline, or to new proposals on the talk page, again you don't put up the "proposed" tag - there are other techniques for that, presently, for instance, listing it on

wikipedia:current surveys

. If you want to draw still more attention to a competing guideline proposal you've worked out, you put that proposal on a separate page in the "wikipedia:" namespace, add the "proposed" tag, and link it from any "talk:" or "wikipedia:" namespace page you like. Again, the old guideline remains guideline throughout such process, until a new consensus is established.

--Francis Schonken 13:41, 3 October 2005 (UTC)

Actually, I didn't mark it as proposed because of the changes I was going to make. I marked it as proposed because there is little evidence that this page has an established consensus as a guideline. As indicated in that edit summary, there has been so little interest in this page, that I think it fundementally fails the "Consult widely" expectation of how to create policy. Aside from your recent change to the labeling of this page, it has never been indicated as a guideline. In other words, I'm not yet convinced any version of this has achieved consensus. I wouldn't be opposed to bringing people in from the Pump and/or Current surveys to discuss it and show me there is consensus however. Dragons flight 14:10, 3 October 2005 (UTC)

Thanks for asking. Takes us back to 2004. The procedure then was

wikipedia:policy thinktank

(that page is a redirect now, so the diffs I mention come from the "history" of that redirect page):

• [2] 31/08/2004: Benc introduces
  Wikipedia:When to create a category
  (what the name of the guideline proposal was at that time) in thinktank;
• [3] 03/10/2004 (funny, exactly a year ago), the guideline, in the mean while renamed to its present name, is removed from thinktank, commentary: Removing "categ,lists,series boxes" - recently promoted to "Policies and guidelines"

Then, a few weeks later "policy thinktank" is changed to the present guideline development system, all guideline proposals in the thinktank becoming "proposed". Nobody apparently still thought about "tagging" this effective guideline, that is listed on wikipedia:policies and guidelines for about a year now.

I still can give more details about the "consensus" proceedings at the time. Result was: there was consensus a year ago. I don't see "fundamental" changes to the guideline since that moment.

Again, present or new disagreements are handled differently. Even if many wikipedians would like the guideline to be something fundamentally different (which, all in all, I don't think to be the case): as long as nobody has an alternate proposal that has consensus, the existing form of the guideline "is" consensus. Having "no guideline" in the mean while is also not something that has whatever kind of "consensus". And the guideline development system is flexible enough to allow updating to current practice and other types of established consensus, etc...

Hope I gave enough detail on this one.

--Francis Schonken 15:22, 3 October 2005 (UTC)

PS: if you ask for the *exact* date when this was first labeled a guideline:

• [4] De-thinktanked (25/09/2004) - commentary: De-think tanked. This is now in play as the deletion guidelines at TfD, and so has pretty much graduated.
• [5] A year ago, inserted in the general "policies and guidelines" category (which was the only available category at the time) - commentary I suppose either "Thinkthank", either "Policies and guidelines" category
• [6] - 27/12/2004 moved from general "policies and guidelines" category to "semi-policy" category (no longer exists - was precursor to "guideline" category)

Maybe interesting to follow the history further, as far as "establishing" this was labeled a guideline once, and within "consensus" I think I was complete. --Francis Schonken 16:00, 3 October 2005 (UTC)

For what it is worth, I was the one who orignally marked it as "proposed" several months ago and went looking for discussion both here and at the pump, however very little discussion occurred (even though no one at that time objected to calling it "proposed"). Little seems to have changed since then and there certainly hasn't been any new showing of consensus. However, I will acknowledge this has been around a long time and most people don't seem too bothered by it, so I guess calling it a guideline is not unreasonable. Would be nice if there was something in the archives to show how people had come to support it, though. Dragons flight 02:59, 4 October 2005 (UTC)

See first half of

Wikipedia talk:Categories, lists, and series boxes/Archive 1

- all contributors to that talk supported it to be guideline (or whatever it was called in those days); discussions are only about the content of the project, nobody casting a doubt on its inclusion in the body of policies & guidelines.

So no, the popular misconception #2 appears to be that there needs to be a discussion about inclusion in the body of guidelines. If everybody accepts the inclusion so natural that a discussion about that is redundant, there needs not be an "archived discussion" about that. If it were the other way around, of course, the most "basic" wikipedia policies and guidelines don't have such archived inclusion discussion. In short: let's get back to the content discussion of this guideline, where and if needed. I think that also covers what D's f wants to say. --Francis Schonken 09:21, 4 October 2005 (UTC)

See: Wikipedia:Village pump (policy)#Lists and references - are references a distinguishing mark between "lists" and "categories"? --Francis Schonken 07:24, 14 October 2005 (UTC)

Here's that discussion, retrieved from Wikipedia:Village pump (policy)/Archive#Lists and references 16:20, 1 November 2005 (UTC):

Lists and references

This topic grew from a conversation between Willmcw and myself here and here.

My original idea is that lists should cite sources like any other page, for example Greek dialects lists these dialects: while experts have developed differring systems for grouping the ancient dialects, that page gives three lists, each with a reference to the expert making this particular listing.

So, when we speak about stand-alone lists, usually in the format [[list of ....]], surely the thing is not different and there should be references also, for example List of compositions by Franz Schubert has an external reference to the Deutsch catalogue (which is the "official" list of compositions by Schubert).

A specific case is when inclusion in a list can be a matter of debate: for example

Now

1. "List of gay, lesbian or bisexual people" (LGLBP), it is considerably more extensive than "List of gay, lesbian or bisexual composers"
2. Due to the nature of LGLBP, miscellaneous editors frequently add spurious material to it. As a principal maintainer of the list, I greatly appreciate the simple criteria of biography inclusion. More importantly, I think that that criteria is the best NPOV system for properly evaluating such claims. If we went to having footnotes for each entry then the maintainers of the list would have to evaluate each reference about subjects who may be totally unknown.
3. It is much better for the editors of respective biographies to evaluate the assertions as they know the subjects and sources best. However, if there is an external reference of high respectability then it might be used as an additional source. There have been some cases where editors of the biographies have resisted mention of a subject's orientation despite evidence where a fallback reference would be appropriate for establishing at least a "debated" orientation.
4. In the case of the LGLBP, relying on internal sourcing has proven very effective.
5. That choice was the result of a long and thorough discussion.

Although I see the point of this justification, I think the system is not as flawless as presented by Willmcw:

1. Erik Satie has been on the list for months, although in the Satie article he was not insinuated to be gay for more than a few weeks a long time ago, so apparently the LGLBP listing was not checked thoroughly for coherence with the wikipedia article, nor for coherence with the "gay composer" list (not listing Satie).
2. There is a risk of circularity of references, when accepting a "categorisation", introduced in an article without external reference, to be "reference" for inclusion in the list. The example discussed with Willmcw is
   wikipedia:categorisation of people
   guideline that says not to categorise with a sensitive category, unless the article reflects it (which was not the case in DD's case) - so external references are lacking, and might even be untracable in this case.
3. Although such "sensitive" topics lists would probably always require some kind of monitoring, my idea is that external references directly in the list would somehow make that less cumbersome (but, I'm really not that experienced in maintaining such lists).

--Francis Schonken 07:36, 13 October 2005 (UTC)

Francis, with all due respect you do not seem to understand the criteria we're using on List of gay, lesbian or bisexual composers. It has no footnotes, other than a couple of old links leftover. New additions to the article are sourced internally, that is the biographies of the subjects must have a reference to their orientation, but the list itself does not need references. Yes, the list needs someone to go over over it and check for conformity to the criteria, a project that no one has adopted. I don't ubnderstand if your complaint is with criteria of the list, the adherence to the criteria (in either case you should bring this matter to the article talk page) or with the general idea of using internal sourcing for lists, which your points don't seem to directly address. -Willmcw 08:10, 13 October 2005 (UTC)

Well, there's no "complaint" of mine. Just trying to figure out what's best for wikipedia. Personally I favour the "external references", but I also understand the reasons to do otherwise.

Would it be possible to provide some link to where the present approach for lists was discussed and finally decided. I would have no problem to adapt

Further, the point with instances like Erik Satie and Dreuxilla Divine is that the editors of the biographies can best evaluate the sources and decide whether an assertion of a particular orientation is verifiable, etc. In the case of Satie, the editors decided that the evidence was that he was heterosexual to asexual and thus that name should not appear on the list. In the case of Divine, the editors decided to add the LGB person category. Should they document it better? Sure. But those chores should be handled by the biographers, not the list editors. -Willmcw 08:17, 13 October 2005 (UTC)

Re. Erik Satie:

• The main problem I mentioned was the considerable time delay between homosexuality claims being removed from the Erik Satie article, and Erik Satie being removed from LGLBP (the "main" list) - I estimate this time delay about half a year, but with some checking of diffs I could give you a more exact number. There's no "reproach" in that, just comparing this system of "internal" references, with a more straight application of "external references", also applied to lists.
• So, I compared that to List of gay, lesbian or bisexual composers, old system with footnotes containing external references: that list appeared more "immune" to the flaw of listing someone with a doubtful assignation.
• I also pointed out that the present system without direct external references apparently does not cross-check lists (in the Satie case: LGLBO and List of gay, lesbian or bisexual composers).

In general I suppose I think this division of "chores" between "list editors" and "article editors" more prone to getting off-track somewhere in the process, than direct external references for every list. But again, that is only my personal view, based primarily on the Erik Satie example: a single example is maybe not enough to decide either way - for that reason I'd like others to join and tell what their experiences are, which might go in another direction than my personal experience. --Francis Schonken 09:43, 13 October 2005 (UTC)

"[T]he present system" is Wikipedia, which means volunteer editors who eschew hard tasks. Everytime I sit down to cross check the list my eyes glaze over. (OK, maybe I just need to have my optical prescription corrected). But seriously, the problem of maintenance occurs no matter what "system" is used. The internal sourcing system makes sense for lists. Whether it be "List of Texans" or "List of left-handed persons", the list editors should be able to rely upon the WP biographies to have the correct info without independently checking and sourcing the facts themselves. To do otherwise would require a serious duplication of effort. -Willmcw 10:03, 13 October 2005 (UTC)

In my experience in attempting to ensure NPOV and factual accuracy of several lists, I have seen that lists have to have certain components, without which these lists evolve into endless edit-wars, vandalism, POV pushing and other problems. To avoid rtes pitfalls, lists need the following:

1. An unambigous and narrow criteria for inclusion, e. g. three lists better than one as in:
   List of people who have said that they are gods
   . If a person fits into two lists, list him/her in two lists.
2. When defining a criteria, consensus must be reached before progressing into adding list items, and care should be taken not to dwelve into original research when defining that criteria
3. The criteria should be clearly stated at the top of the page;
4. Each entry requires one or more external (or WP) references. Entries without references are not allowed;

Another possibility is to evaluate the use of Categories rather than lists. Categories and subcategories can be created to accomodate most of the content of these lists. ≈ jossi fresco ≈ 16:09, 13 October 2005 (UTC)

Re. 1, I don't see a problem to have long lists too. There's, for instance

Wikipedia:Naming conventions (long lists)

: even if there are a lot of LGBT sublists, I think the question for a general list is justifiable as much, and I don't see a problem with it.

Re. 2 & 3, I think this doesn't acknowledge the difference between lists and categories as explained in

wikipedia:categories, lists, and series boxes: as much as I defend categories to have strict definitions (and even these are sometimes rewritten afterwards, see discussions going on at category talk:terrorists

), in the same measure I defend lists to be a more open approach. With external references for lists the validity/credibility of the inclusion depends on the validity/credibility of the external source, and wikipedians don't have to break their head about it.

Re. 4, of course I agree with that, and this is also one of the "natural" differences between cats and lists (while cats lack these individual direct external references completely, category inclusion solely depends on content of the respective wikipedia articles).

In general I think Categories and Lists each have their own specificity, and can be used alongside one another, each in their own right, where lists can definitely be broader than categories. --Francis Schonken 07:17, 14 October 2005 (UTC)

There are contradictory statements in two guidline articles, concerning the question whether pages should belong to the parent categories of the categories they define. For example, should

Well, the general rule is to place an article as deep as possible inside the category-tree. So in this case, Musical notation should be placed inside Category:Musical notation, if I'm correctly informed. Cheers! Peter S. 01:10, 31 October 2005 (UTC)

That is obvious. My question was if it should be also placed in Category:Musicology and the other parent categories of Category:Musical notation (the category it defines). Karol 01:52, 31 October 2005 (UTC)

Should it be placed inside Category:Musicology, Category:Musical terminology or Category:Notation? No, I wouldn't do this, because Category:Musical_notation is a deeper category for all of them. Peter S. 14:01, 31 October 2005 (UTC)

What about the exception stated

here about pages defining the categories they're in (the Ohio example)? Karol

14:46, 31 October 2005 (UTC)

Hm, good point. I would actually disagree with the rule explained with that Ohio example. Maybe somebody else could shed some light on that? Peter S. 14:54, 31 October 2005 (UTC)

The Ohio example makes sense to me from both a practical and "theoretcial" perspective, although I'm having some difficulty articulating the latter. From a practical point of view, the reader doesn't need the category "musicology" to find the musicaology article, but does want to use category:musicology to find a list of articles within musicology. The theoretical reason relates to the use of categories as a "tree" or diagram of concepts. You don't want to find the concept in in its own tree. Using the {{catmore tag on the category page is also a good practice. --Sjsilverman 00:42, 24 December 2005 (UTC)

The way it's demonstrated with Ohio and Category:Ohio makes the most sense to me. Just think of it this way: Suppose there was no Category:Musical notation, only an article Musical notation. Would Musical notation be in Category:Musicology? Clearly it would. Now suppose that an editor looks at Category:Musicology and sees that many of the articles there are about musical notation, so decides to create Category:Musical notation. Which makes more sense: removing Musical notation from Category:Musicology because Category:Musical notation is now a sub-category? Or keeping Musical notation in Category:Musicology because "Musical notation" has just proved such a significant sub-topic of "Musicology" as to merit its own sub-category? To my mind, it's clearly the latter. -- Antaeus Feldspar 06:01, 12 February 2006 (UTC)

The section about lists read:

Example of a list. The policy on lists is generally fairly permissive. While they need to meet the same standards of noteworthiness of any other Wikipedia article, these standards are, as always, fairly lax, and lists are generally permitted despite perceived redundancies with categories or boxes.

I would argue that the statements about "standards of noteworthiness" being "fairly lax" is contradictory to

(You may also be interested in the proposed guideline being drafted WP:LISTS ≈ jossi ≈ t • @ 17:03, 6 January 2006 (UTC))

I adapted the guideline accordingly. This guideline should not make statements about whether other guidelines are "lax" or "permissive": apparently this alleged laxeness is a long lost & obsolete contention. --Francis Schonken 10:28, 23 January 2006 (UTC)

I've noticed that many people don't understand how the categorization is supposed to work in respect to a "defining article" -- that is, an article which defines a category, as Ohio does for Category:Ohio. I'm thinking of creating Template:Defining article, which would just consist of a commented-out explanation of why the article really should be in both the category it defines and that category's parent categories, too. This would hopefully keep people from removing the parent cats as "redundant". Thoughts on this? -- Antaeus Feldspar 05:35, 12 February 2006 (UTC)

OK, created. -- Antaeus Feldspar 02:51, 16 March 2006 (UTC)

If not indicating that the template should be used with a "subst:" it is quite useless IMHO. Even then I'm not convinced it would be of much real use, but don't see why we shouldn't give it a try. --Francis Schonken 09:32, 16 March 2006 (UTC)

Good point; I'll add a note about "subst:"ing to the usage instructions. Is there something that can be done on the template itself to mark it as one that should always be subst'ed (so that, for example, a bot could make the subst if the user didn't know or forgot?) -- Antaeus Feldspar 19:29, 16 March 2006 (UTC)

I don't know, I didn't think so. I only know we had something similar with

wikipedia:footnotes#Helping editors unfamiliar with this system of footnotes

({{subst:footnotes}}) too (that is: write it somewhere in a guideline, and try that guideline to be the access point for using the template, but that's of course still far from the "hard coded" solution you're asking for) – technically it shouldn't be impossible though, signatures (...applying the four tildes) work as an "automatic" subst:.

Just thought of a simple solution for finding out where the template is used non-subst: http://en.wikipedia.org/wiki/Special:Whatlinkshere/Template:Defining_article – any wikipedia article mentioned in that list used the template non-subst. --Francis Schonken 11:07, 17 March 2006 (UTC)

The only second thought I still have regarding your solution is that it might hamper usual wiki proceedings (notably: writing articles in a NPOV approach) if articles are somehow "frozen" to serve as a category definition. That's of course something only hampering in the case of sensitive categories: an article will mention as many different definitions of a concept as NPOV requires, broad definitions as well as narrow; the category definition will have to choose one (usually rather a strict one, one that excludes vague border zones, as indicated in

wikipedia:categorization of people#Clearly define the category, in order to avoid inclusion disputes), but yeah, the "categorisation of people" guideline effectively indicates that the proper way to handle this is not to rely on a multiple-definition article for the category definition, but to write down a workable category definition on the category page. --Francis Schonken

10:40, 17 March 2006 (UTC)

Well, I think the problem is that we haven't hit on a really clear wording for the concept that "defining article" or "article that defines the category" is our current term for. It isn't really the definition of the category -- that, as you point out, is what the category page is for. It's instead "the article which is on the same topic as the category" -- I mean, you don't define Ohio, but you have a category Category:Ohio and you have an article Ohio and it's clear that the relationship the article has to the category is the relationship we're looking for. If we can find a short, simple way to express that relationship, to replace the problematic "article which defines the category", it would help things. -- Antaeus Feldspar 17:01, 17 March 2006 (UTC)

I don't see the point of a {{

WP:SUBST, check it out. Omniplex

21:09, 22 March 2006 (UTC)

• IMHO bad idea (I mean: if anything I think this guideline should discourage confusing lists-in-category-namespace cross-overs, apart from the usual navigational templates per
  Wikipedia:Template messages/Compact tables of contents#To_use_on_Category_pages
  )
• Bad example (rather a "subst:" type navigational template is what I see; indeed many navigational templates contain some sort of "list"; but that is not how "list" is used on this page:
  Wikipedia:Categories, lists, and series boxes#Lists
  addresses lists in main namespace per the intro of that section)
• I can't imagine there are good examples, so I'm not going to look for things that as far as I can imagine are non-existent. This is a reply to Omniplex' edit summary "[...] please replace the example by something better if available" [7]
• "Combination of advantages" as contended by Omniplex: really, I can't see any. I see a combination of disadvantages, plus the major disadvantage: confusing.

--Francis Schonken 16:47, 16 March 2006 (UTC)

The lead section offers: These methods should not be considered to be in competition with each other. Rather, they are most effective when used in synergy, each one complementing the other.

For a simple list and category combining them is ideal, it allows to annotate relevant articles in the category. It also simplifies maintenance, watch both aspects of the same set (list and short category) in one place.

It's of course no good idea for large sets of articles, in that case the list can be easily split into a separate article with links from / to the category.

Omniplex  18:19, 16 March 2006 (UTC)

• The status of this wannabe-guideline is apparently less than clear as documented above.
• It had a clear
  point
  discouraging categories in favour of lists, and you're removing simple facts. Mixed with removals of stuff I didn't add claiming that it was an unilateral modification by me.
• If that's your
  own
  article pleaase move it to your userspace.

Omniplex  18:47, 16 March 2006 (UTC)

There is presently no consensus that the category/list "synergy" is to be represented differently on the guideline page than it was before, so your changes are still unilateral. Your talk above didn't convince me otherwise: putting lists on category pages still a bad idea IMHO. You have no more than 50% approval of your proposed intrusive changes. That is not consensus. Will revert. Discuss all your intrusive changes on the talk page first, and please try to establish consensus, rather than making attempts at igniting revert warring.

Obviously, that's why I picked your version after you moved the example here as stable point for the reversion.

Your next modification reinserted "unsightly" in the series box disadvantages. That was a dupe, we can't have essentially the same disadvantage twice, "ugly" (in #2) covers "unsightly" (in #3), in fact #2 "ugly" originally was "unsightly", if you like that adjective better than "ugly" replace it in #2. The new #4 added by me (users simply remove boxes if they don't like them) covers "complaints" in the original #3, so that's also a dupe (admittedly you didn't reinsert that). Finaly #3 only discussing

POV, "unsightly", and potential complaints is too long compared with other points in the enumerations of (dis)advantages. Omniplex 

19:25, 16 March 2006 (UTC)

--Francis Schonken 19:04, 16 March 2006 (UTC)

Well, I see User:Francis Schonken trying to discourage categories as allegedly too complex for ordinary users favouring lists. However categories are not very difficult to use, they are supported by the software, and many users help to cleanup categories when needed. For a list it's far less clear that it's maintained for a longer period of time.

Comparing "my" reverted version with the stable version discussed in the previous chapter shows that this was in fact your version. Comparing that with your reversion results in the following changes:

1. (rm) Each article can belong to multiple categories.
2. (rm) Categories are easy to use and essential like site maps on Web pages.
3. (rm) Each category can belong to one or more parent categories.
4. (rm) Each category can contain multiple subcategories.
5. (rm) Category talk pages allow to discuss sets of related articles and subcatecories.
6. (rm) Category pages can contain ordinary content like an introduction and/or links to the main articles.

Completely beyond me what should be wrong with these facts. Removed disadvantages reinserted by you:

1. Newbies would need some time to understand how to add an item, how to link new categories into existing schemes, why, when and how to avoid circular category shemes, how to handle piped link assisted sorting, the importance of lean category definitions for sensitive categories and how to address POV concerns (see also Wikipedia:NPOV tutorial#Categories).
2. Although categories can include alternative names for the same item, the technique to achieve that (by categorising a redirect page) is not very user-friendly for newbies, and generally not encouraged.
3. Categories can't be served from the cache and thus may have a negative effect on wikipedia's response time.

Debatable, I tend to agree with the 2nd point, although it's rather esoteric. The 3rd point needs some evidence, my reason to remove it was (bogus "server load" disadvantage removed - compared with article series templates this is ridiculous). Quoted from the edit history. For the 1st point I consider categories as simple, especially in comparison with templates. It fits that you did not remove a disadvantage of categories added by me.

Last but not least you added the bogus Category:Wikipedia style guidelines again, although that's a child of Category:Wikipedia guidelines with its own "look and feel" (= header template and {{Style}}), it's pointless to have a guideline also in style-guidelines, or vice versa.

Omniplex  20:37, 16 March 2006 (UTC)

1. (rm) Each article can belong to multiple categories.

   • Included in point one of the advantages list: "Auto-linking. Create a link to as many categories as you like on an article page, and a corresponding link to that article will be visible on the category page."

2. (rm) Categories are easy to use and essential like site maps on Web pages.

   • Easy to use: contradicting with "Newbies would need some time understand how to add an item, how to link new categories into existing schemes, why, when and how to avoid circular category shemes, how to handle piped link assisted sorting, the importance of lean category definitions for sensitive categories and how to address POV concerns (see also Wikipedia:NPOV tutorial#Categorisation)." and with "Although categories can include alternative names for the same item, the technique to achieve that (...) is not very user-friendly for newbies, and generally not encouraged." And then I didn't even mention a host of other things that are distincly less easy in category namespace (like moving categories; differing naming conventions;...) and things a newbie can have trouble with (e.g. checking whether appropriate categories are available when you're writing an article, instead of leaving a redlink category at the bottom of an article,...)
   • "Site map" analogy is not remotely near to how categories work.

3. (rm) Each category can belong to one or more parent categories.

   • not a difference with the other two grouping techniques discussed on the page (so, no specific advantage when comparing): lists can be part of as many "(lists of) lists" as you like; each navigational template can be inserted in as many umbrella nav templates as you like, etc...
   • as such this feature is implied when talking in point three of the category advantages list about "Multi-directional navigation. Categories are organized within Wikipedia into a web of knowledge (...)"

4. (rm) Each category can contain multiple subcategories.

   • not a difference with the other two grouping techniques discussed on the page (so, no specific advantage when comparing): lists can have links to as many lists as you like; each navigational template can group as many other other nav templates as you like. (Not so long ago a vote was proposed trying to limit that, but hopelessly failed, Wikipedia:Village pump (policy)#WP:AUM rejected)
   • as such this feature is implied when talking in point three of the category advantages list about "Multi-directional navigation. Categories are organized within Wikipedia into a web of knowledge (...)"

5. (rm) Category talk pages allow to discuss sets of related articles and subcatecories.

   • not a difference with the other two grouping techniques discussed on the page (so, no specific advantage when comparing): the same discussions are possible (and occur) on list talk pages and template talk pages. Need examples? In fact sets of related articles and subcategories are also discussed on WikiProject and project namespace pages & talk pages, which is even more preferable.

6. (rm) Category pages can contain ordinary content like an introduction and/or links to the main articles.

   • Not an advantage compared to lists where the same proceedings are possible. Templates are disadvantaged in this sense, that's why I elaborated in the templates disadvantages list: "(...) For this reason article series boxes need to be self-evident, while they can't contain much text for definitions or explanations." --Francis Schonken 20:51, 16 March 2006 (UTC)

Hello, I made a couple minor edits to improve clarity in the lists section. I added the part about why not having auto-linking is a disadvantage, and changed the example list to something more neutral and suitable as an example. DavidBailey 11:43, 1 May 2006 (UTC)

Francis, no one is censoring. I made an edit to make the instructions clearer. Examples should be things that would be clearly understood by the reader without adding unnecessary social or political statements. Please refrain from making statements in Wikipedia "how to" pages. DavidBailey 17:01, 5 May 2006 (UTC)

Not making any statements apart from this one:

If you have another list with numbered footnotes, and that over-all is not problematic, I don't see why it couldn't be used.

Please note the content of these warning templates: {{

So now I'm going to replace the list example by what was there before you removed it the first time. But I'm open to alternative proposals... please let's try to find a compromise here. But until then I place the most successful list I know of as an example. --Francis Schonken 17:34, 5 May 2006 (UTC)

Thank you for being patient with me. I just find some folks insert articles related to personal political or social views for no other reason than they want to "raise awareness" of them. If you have a valid reason for this particular list, such as the structure of the list, I have no concerns. I agree that the structure is a good demonstration of an annotated list. DavidBailey 03:22, 6 May 2006 (UTC)

There's a request dating from March for a third opinion on a dispute on this page on

Would be willing to provide a fourth opinion. Please leave me a note on my talk page in case you still need help. Socafan 12:59, 14 July 2006 (UTC)

There is a new discussion going on here. AdamBiswanger1 04:08, 15 July 2006 (UTC)

At

DGG 20:49, 8 September 2006 (UTC)

Research library is currently a stub article;

Individual research libraries are currently listed in List of libraries.

Where is the problem?

It is always possible to mark research libraries in the list of libraries by adding <ref>[[research library]]</ref> after those entries that are research library, and write <references/> somewhere at the bottom of that list page (see

wikipedia:footnotes for the technique with references) --Francis Schonken

20:57, 8 September 2006 (UTC)

The second is actually adding a minor modification to the rules on how sub-categories relate to parent categories. The general rule has been described as "a sub-category, and the main article of that sub-category, should both belong to the same set of parent categories." However, it was pointed out to me some time ago that this doesn't reflect current practice when it comes to sub-categories named after people. Arthur Conan Doyle is a member of Category:Portsmouth F.C. players. He also has his own category, Category:Arthur Conan Doyle. Is someone who is interested in Portsmouth F.C. players likely to be interested in all the contents of Category:Arthur Conan Doyle? It doesn't seem a very logical connection, and yet the one is a subcategory of the other. I'd like to put something in to mark this as a major exception to the general rule. -- Antaeus Feldspar 16:57, 8 October 2006 (UTC)

I'd take that to Wikipedia:Categorization. I don't think the intent of the CLS guideline is to give the finer details about Categorization: *comparison* with other grouping systems is the focus point of the CLS guideline. None of the subheader topics and additions you propose has a parallel in the other two grouping systems (Lists, Series boxes). --Francis Schonken 18:29, 8 October 2006 (UTC)

A discussion is going on

I've created the above proposal in response to what appears to be a flood of dubious list construction. Please review and comment. Mangoe 20:08, 28 March 2007 (UTC)

I have taken the merge tag off of the the article. The merge tag was suggested by

I concur with the removal of the merge tag. That page is just too wordy. I forgot I even wrote it. When this page naturally grows to include the points on that page, that page should simply be redirected. The Transhumanist 00:53, 11 September 2007 (UTC)

See

On the project page the following statement is made:

An article should usually not be in both a category and its subcategory ....//more text//...

Exceptions should also be considered when the article subject has a relevance to the parent category that is not :expressed by the subcategory's definition. For instance, if Category:Guillotined French Revolution figures was the only :subcategory of Category:French Revolution figures, it would not make sense to remove major figures of the French :Revolution solely because of the means of their death.

Could this be clarified? I simply don't understand the last part it would not make sense to remove major figures of the French Revolution solely because of the means of their death.

Would be helpful, please. --213.238.246.251 10:00, 21 May 2007 (UTC)

Is this actually true? I actually find cat pages are a great hiding place for spam as few editors seem to have them watchlisted. Paxse 14:09, 24 July 2007 (UTC)

Its definately not true. Although indeed only Wikipedia articles can appear as a member of a category, many categories have descriptions above the members and those descriptions most definately can contain external links. --Cpt. Morgan (Reinoutr) 15:12, 24 July 2007 (UTC)

There is a deletion discussion concerning

Look in Category:Economics and you won't find many articles there. Because of how they are used on Wikipedia, categories tend to be narrowly focused even when they do contain many articles. Topic lists allow editors to list a larger number of related articles that may be scattered through many categories and under several layers of the category structure. It can be handy for a casual reader wanting to browse related topics. —Mike 05:25, 22 August 2007 (UTC)

That guidance directly contradicts WP:Template namespace#Usage. One or the other needs to change.LeadSongDog come howl 18:25, 7 July 2009 (UTC)

I've edited the article to remove some of the more dubious justifications. My explanations:

• "First, editors defend their systems of choice vigorously, so forcing confrontations between them in deletion discussions wastes the time and effort of contributors who would be better utilized by allowing them to develop their respective systems." - This is an argument that could apply to any deletion nomination whatsoever (how often do you see an extremist inclusionist on AfD saying "why aren't you people working on articles?"). To say that something should not be nominated for deleton because its supporters are zealous is even more bizarre; how many trolls will zealously defend their hoax or original research articles?
• "These pages often have links that their counterparts do not have - simply deleting such pages wastes those links." - This invites the argument, why not just merge the links from one into the other? This isn't terribly difficult to accomplish in either direction, as long as there aren't redlinks involved, and is easy to automate. It's not a coherent argument against deletion.
• "Third, deleting list pages just because they are redundant to categories may disrupt browsing by users who prefer the list system." I disagree with this one but not quite as strongly - I believe that duplicated information in lists and categories is difficult to keep in sync, and that ideally one should be automatically derived from the other in real time (presentation should be separate from content). However, the software support for this does not currently exist.

I'm also concerned about the tone, which in a number of places seemed accusatory and overemotional ("they are doing Wikipedia a disservice", "deleting link lists is a pointless waste", ).

I've also removed "Categories impose additional load on the servers, because they can't be served from the cache", because this is a misleading statement. Categories can very well be served from the cache, and do not require a query for each request - they only need to be regenerated when an article is modified to remove or add that category tag, which can be done at edit time rather than request time (whether or not it is implemented this way, I haven't checked). I removed the bit about invisible links too, because that's a hack that we really shouldn't be using at all. I removed "Categories are not operational on most mirror sites" because well, that seems dubious - a lot of them run Mediawiki software and fully support categories.

The whole thing has a pro-list slant - the "lists" section is loaded with "advantages" and its "disadvantages" are all straw men that are rebutted ("this seems like a disadvantage, but really it's not"). Article size guidelines are not intended to apply to lists and linkspam is not solved by categories (which have an editable wikitext portion in which they can be placed). I've added some of the very real disadvantages of lists - which emerge directly out of the advantages of categories and are the reason the category system was implemented. Dcoetzee 20:34, 14 January 2008 (UTC)

I've edited the phrases you were concerned with above (those with "disservice" and "pointless") to remove the undesired tone. Let me know if you think it needs more work. I look forward to your further suggestions. The Transhumanist 03:16, 1 February 2008 (UTC)

Dcoetzee, in reply to "These pages often have links that their counterparts do not have - simply deleting such pages wastes those links", you wrote: "This invites the argument, why not just merge the links from one into the other? This isn't terribly difficult to accomplish in either direction, as long as there aren't redlinks involved, and is easy to automate. It's not a coherent argument against deletion."

It is a coherent argument, because the links are simply getting deleted. Many list deletion debates result in deletion in favor of categories, with no merging taking place. A great many links, and by extension the effort it took to compile them, are being wasted. Your argument only works if there is a required procedure to merge those links that is adhered to. There isn't. The "categories handle this better" camp are wasting a lot of list-builders' work. The Transhumanist 19:26, 11 February 2008 (UTC)

A less obvious result of a list deletion, even if its links get merged into a category, is the effect it has on the list builders of that list. They may stop trying to develop a list for that subject, because proceeding would now entail building the list again from scratch before they can move forward, and adding enhancements that individual categories don't support (structure, section leads, annotations, pictures, redlinks, etc.) to protect it from being deleted again (and even those aren't a guarantee). They may just move on to working on other lists, leaving gaps in the overall list navigation system. The Transhumanist 19:44, 11 February 2008 (UTC)

Retaining explanations in lead section

A continuous problem at AfD is the category-vs-list conflict. One purpose of this guideline is to end that conflict. The lead included a good explanation of the conflict, but it was recently removed. I've restored the version of the lead that includes Dcoetzee's refinements above. The Transhumanist 03:01, 1 February 2008 (UTC)

This is a proposal to adjust the title to

Series boxes are a subtype of navigational templates. But navigational templates are covered by this guideline as well. Therefore, it makes sense to replace "series boxes" in the title with its parent "navigational templates" so that it more accurately describes what the guideline covers.

The Transhumanist 20:01, 11 February 2008 (UTC)

• Support - I agree, that's more applicable now. Torc2 (talk) 20:03, 11 February 2008 (UTC)
• TH's proposed name is certainly much better than the current one, but how about we make it even more generic: Wikipedia:Categories, lists, and templates? I think this guideline can also apply to other templates, such as infoboxes (e.g., do we list a company's top executives in the infobox, or in an embedded list in the article?) UnitedStatesian (talk) 20:39, 11 February 2008 (UTC)
  • I like UnitedStatesian's idea, but how about differentiating that from non-list templates with the title "Wikipedia:Categories, lists, and list templates"? "List templates" can then be defined in the guideline to include all the various types of templates used to display lists. The Transhumanist 23:01, 11 February 2008 (UTC)
    • I like it. Perfect. UnitedStatesian (talk) 02:05, 12 February 2008 (UTC)
    • What are they actually called though? I thought
      WP:NAVBOX were the standard terminology. Torc2 (talk
      ) 02:42, 12 February 2008 (UTC)
      • "Templates with lists in them". There is no official name that includes all of them. Infoboxes aren't generally considered navigational templates, but they often have lists in them (like the casts of movies). Just an observation. The Transhumanist 18:56, 12 February 2008 (UTC)
        • I really prefer "navigational boxes" over "list templates". The navigational features in an infobox can be called out in the article without having to appear in the title. In any case, it's doubtful an infobox will compete with a category or list. Torc2 (talk) 19:07, 12 February 2008 (UTC)
          • Reflecting on the above points, "Navigational templates" would be best, because if a person wants to know more about templates, the term will actually work in searches. "Boxes" doesn't turn up much of relevance. And "
            Wikipedia:Navigational boxes" is a deprecated page. The Transhumanist (talk
            ) 18:15, 14 February 2008 (UTC)
    • Hearing no further comment on this for to days, I am going to make the move TH proposed. UnitedStatesian (talk) 04:55, 24 February 2008 (UTC)

I understand the desire to change the title, but content needs to be brought in sync for such a title to make sense. It seems odd that the title is now "Categories, lists, and navigational templates", but the main sections remain

I agree completely that the page and title need to be further sync'ed, but I think the way to do this is to continue to move forward, rather than going back. I have made some further changes (including updateing the sections), and am sure more will be forthcoming (from me and other editors) to continue the process. UnitedStatesian (talk) 06:14, 27 February 2008 (UTC)

Thanks--your updates are making the page easier to use and understand. Libcub (talk) 07:13, 27 February 2008 (UTC)

References

I have boldly edited the section listing unreferencability as a disadvantage. Since by definition a category item is simply a 'handle' to data elsewhere, there is no need to reference it in the category, nor would that be appropriate even if it were possible. The references belong in the articles. (Of course it might be appropriate to reference header text over a cat list.) Halfmast (talk) 05:31, 1 April 2008 (UTC)

What about using objects? They are not so usefull in Wikipedia as they are in Wikihowto.... moa3333 86.122.95.211 (talk) 14:30, 24 July 2008 (UTC)

Recently I've come across several navigation boxes (both right-side and bottom) which unceremoniously employ images which are not always relevant and increase the template's size to ridiculous proportions. Here's an example (so far I have reduced the size, but the edit may be reverted - we'll see). [[Template:Palestinian Arab villages depopulated during the 1948 Palestine War |Here]]'s another example which I'm discussing right now (although it's a much more complicated issue).

In any case, I'm concerned that there's no guideline on Wikipedia (at least as far as I'm aware) regulating images and image size in navigation boxes and infoboxes. If there's consensus here that such a guideline is needed (as part of this guideline or another one, or as a separate page), I'll be happy to prepare a draft.

-- Ynhockey ^(Talk) 10:28, 29 September 2008 (UTC)

A guideline would be very helpful. The Arabs template has a syntax error, and in any case seems unlikely to be useful. Have you made any progress on this issue? --DThomsen8 (talk) 12:41, 30 April 2009 (UTC)

In light of the discussion at

When somebody has a problem with a category they don't like, they come to category-for-deletion

Comments? I'm going to repost this around on the several TALK pages which deal with this matter. SBHarris 01:25, 27 November 2008 (UTC)

A few questions that came up as I was reviewing the supposed advantages of lists in an AfD...

• "Lists can include invisible links to discussion pages, so that clicking on "related changes" will include those." (currently #10). I've never seen a list that does this - is this in fact common practice? How does the utility argument override the general restriction on cross-namespace links and self-reference?
• "Lists are included in searches of Wikipedia." Categories are also included in searches - how is this unique to lists?

Zetawoof_(ζ) 10:17, 17 December 2008 (UTC)

I've suggested standardizing template naming, at Wikipedia talk:Template namespace#Template naming conventions. If you're frustrated with typing template names and constantly guessing at the right capitalization and spacing, please chime in. —Michael Z.  17:49, 10 January 2009 (UTC)

See discussions on

The last point on

I would think

WP:PERF is probably the better view on this issue. Considerations of performance should not be a major consideration in deciding whether or not to create or apply an individual template. Good Ol’factory ^(talk)

02:40, 29 April 2009 (UTC)

I agree too, and took that last point out. UnitedStatesian (talk) 18:44, 29 April 2009 (UTC)

There should be links to articles which use navigation templates in a paragraph that starts out saying Examples of navigation templates. Merely mentioning names like Narnia don't really do the job. --DThomsen8 (talk) 12:45, 30 April 2009 (UTC)

This editing guideline is rather clear in specifying that "the 'category camp' should not delete or dismantle Wikipedia's lists, and the 'list camp' shouldn't tear down Wikipedia's category system—doing so wastes valuable resources. Instead, each should be used to update the other." Something that I think no one anticipated is the use of arguments from the "category camp" that a category should be deleted in favor of creation of a corresponding list (for example, see here for a claim endorsing deletion on basis that "this would make a better list") as well as equal and opposite arguments for deleting a list in preference for a category. As stated in the close of Wikipedia:Deletion review/Log/2009 April 28, "CLN is pretty clear about lists and categories not being competitive". We should add some text here to specify this more clearly. I propose:

"The existence of a list or navbox (or the claim that a list or navbox would be a better method of presenting the information) is not a valid argument for deleting a category. Nor should the existence or creation of a category or navbox be accepted as an argument for deletion of any list. Any category, list or navbox proposed for deletion needs to be independently evaluated on its own merits based on their respective objective standards for retention."

Any discussion? Alansohn (talk) 15:55, 4 May 2009 (UTC)

I don't really understand. I think it is a quite reasonable line of argument that some categories are better off as lists and that some lists might be better off as categories. And of course, sometimes it is useful to have both a category and a list. older ≠ wiser 16:08, 4 May 2009 (UTC)

I might agree to that change if you modify it to add "the CLN guideline should also not be accepted as a reason to keep any list or category, which should be independently evaluated on its own merits..."

But even then, probably not. Just keep it the way it is currently worded. Per Bkonrad, sometimes one is preferrable over the other, sometimes both are. --Kbdank71 16:49, 4 May 2009 (UTC)

I agree with Bkonrad and Kbdank71 above. I do agree with Alansohn that the mere existence of a list is not a valid argument for deleting a category, because they are not mutually exclusive—information may validly be presented in multiple formats and forms of organization (and this is the main point of WP:CLN). But I strongly disagree that a list or navbox would be a better method is not a valid argument for deleting a category. Such arguments have been made—and accepted—for years, BTW, probably since around the implementation of categories in 2004. There are (at least) two basic reasons for having different deletion criteria for lists and categories, and for preferring lists in certain instances over categories. One, we can't allow all possible verifiable categories because they could (and have in some cases) flood articles to the point where trivial ones buried the nontrivial, and the article text itself could be dwarfed by category tags. By contrast, including an article subject in a list only affects that article by adding the list to "what links here." So it could render categories (and articles) largely nonfunctional to make a category for every fact an article could contain, but we can clearly be far more tolerant of lists. Two, the functional limitations of categories make them ill-suited for certain facts or classifications that require annotation or explanation (or exceedingly long category names) to be coherent. Lists can provide this.

Bottom line, "a list exists or could be made" is not a valid argument for deletion of a category, but it does not follow from this that "certain information is nonfunctional as a category but not as a list" is not a valid argument for deletion. That lists and categories are not in competition does not mean they are equally suited to every task. Postdlf (talk) 19:17, 4 May 2009 (UTC)

• I think we're actually all agreeing in many aspects of these issues, but using different terms. Categories shouldn't be deleted solely because a corresponding list exists (which is what I've said above). And despite how I've been frequently mischaracterized, often it seems deliberately so, I agree that the existence of a list does not require the existence of a category. Categories should be deleted because they don't work, not because a list is somehow "better". Both lists and categories have their own separate retention criteria that are not equivalent. While no one seems to be able to define what "definingness" means, I think we are in agreement that "defining" is narrower than (the far-better defined term) "notable". The argument "makes a better list" is often framed in terms of design features (can't provide additional information, can't have red links, etc.) that apply to all categories without any relevance as to why a particular category should be deleted and not have the entire category system dismantled. While we often disagree on its particular implementation, there are plenty of mechanisms, most notably in
  WP:OC, that specify scenarios where categories would be deleted while corresponding lists exist or could be created. Instead of arguing to delete a category because a list would be "better", which conflicts with what WP:CLN already says as confirmed at the recent DRV for Knuckleball pitchers, the argument should be that there is some inherent issue (WP:OC, etc.) that makes the category itself non-functional. If we can end the category vs. list issue and finally see someone specify how to determine "definingness" in some non-arbitrary manner, we might actually find far more common ground than exists now. Alansohn (talk
  ) 03:01, 5 May 2009 (UTC)

• Ultimately then your complaints just target how people word their comments at CFD. That a category "would make a better list" is in substance an argument that the particular classification a category attempts is undermined by the structural limitations of the category system so as to render that category too incoherent or nonfunctional. That a list could deal with it effectively is of course not necessary (nor sufficient) for the category to be deleted, but a discussion framed in those terms is intended to emphasize that particular category's flaws by by comparing how a list's text would accommodate that subject. "A list could handle this" is also often offered as a consolation, so it isn't perceived as a zero-sum choice of keeping it as a category or not documenting it in any form on Wikipedia. And of course the complaint that particular classifications don't function within the structural limitations of categories does not mean that all classifications are equally nonfunctional as categories. That's why none of this conflicts with or is at all relevant to WP:CLN concerns, because it doesn't stem from a general preference for lists over categories or vice versa, such that lists and categories would be mutually exclusive.
• On what "defining" means in the category context, that's a separate issue, but I've typically thought of it as the facts that you'd expect in an article's intro paragraphs. I think "notable" usually just means anything worthy of note in an article. Postdlf (talk) 03:56, 5 May 2009 (UTC)

• It's not a wording issue. Every category can almost always be surpassed by a list, an aspect of the design of Wikipedia.
  List of Presidents of the United States provides ordinal number, images, dates, party affiliation, vice presidents and multiple sources, all things that could never be added to the category under any circumstances. Using the "a list would be better" argument, we'd get rid of this category. I'd be hardpressed to even conceive of a single category that is better than some corresponding list, where the list couldn't be improved with minimal effort to surpass the benefits of the category. The bottom line is that lists can do many things that categories can never do, not just the types of things already listed for the POTUS list, but can be sortable, use color coding and other presentation methods, have red links to articles not yet created and can be divided into sections. The ultimate problem with the "a list would be better" argument is that it is completely vacuous. As all lists are inherently "better" by design, it speaks nothing about why that particular category should be deleted. As CLN emphasizes, lists AND categories are designed to co-exist and ideally would exist for all combinations. That there are certain topics that would make good lists but bad categories is best met by arguing that it makes a bad category, not that it would be a "better" list. Whether the definition of "defining" is a separate issue or not, the fact that no one, you included, has ever offered an explanation of what would make a category defining only further undermines the legitimacy of the "not defining" argument. It's rather hard to discern "I know it when I see it" from ILIKEIT / IHATEIT. Alansohn (talk
  ) 15:07, 5 May 2009 (UTC)

• Please re-read my above comments, as it seems you continue to misunderstand. The point is never simply that a list is better to document the classification or fact under discussion in a CFD. The point is always that the particular classification does not work as a category. Highlighting how a list can accomplish what is necessary for that classification to be presented coherently or meaningfully is not the same thing as blandly saying that a list simply handles it better. It is instead another way of explaining why that classification does not work as a category.
• I appreciate that you find list-category comparisons distracting, but there's no reason why it should kill discussion. If someone posts a CFD comment that a category should be deleted because "a list can do X," replying that "lists and categories should coexist" would be completely non-responsive. A productive counter-response would instead be "this category does not need to do X to function," as that would actually address why list capabilities were raised in the first place in relation to the category. Postdlf (talk) 15:44, 5 May 2009 (UTC)

• If the sole justification for deletion is "a list would be better", the reply that lists and categories should co-exist is intended to elicit a reason to delete the category not to promote an alternative option. CLN emphasizes that they are not an either/or option. No category can do any of the things a list can do. Without an explanation of why the category doesn't or can't work as a category, the statement that "a list would be better" adds nothing to the conversation. Any thoughts on how to define "defining" would also be helpful. Alansohn (talk) 19:34, 5 May 2009 (UTC)

• I'd be happy to discuss "defining" with you elsewhere (wherever the proper guideline is, WP:OCAT, or WP:CAT?). True, any opague statement such as "a list would be better" or "lists and categories should co-exist" may add little to a CFD discussion if it isn't elaborated. But people do use shorthand in CFD (or any XFD) because they presume their statement is understood in context, they're short on time, etc.; "a list would be better" in a CFD clearly means "this category has functioning/inclusion problems that a list can deal with." Once you've responded to "a list would be better" with "your comment is contrary to policy because WP:CLN mandates lists and categories co-exist," you've missed their point, and probably just pissed them off, and lost the chance to actually have a meaningful discussion. If you don't understand their point or simply wish to challenge it, you need to take more time to get at why they're really saying it, and how it applies to the category at hand, instead of jumping on them for what they're probably not saying—I challenge you to find a single Wikipedian who believes that lists and categories should never co-exist. Ask CFD commenters to explain themselves further, or as I suggested, respond by addressing how the category limitations in that context are not a problem for that category. Otherwise you're just attacking a complete straw man argument that no one is making and failing to advance the discussion. Postdlf (talk) 20:27, 5 May 2009 (UTC)

• Talk about straw men. The existence of a list doesn't require the deletion or continued existence of a category. The existence or possible creation of a list has no relevance to the deletion of a category. No vote on deletion of a category should ever be based on insisting that "it would make a better list" or that it must exist because the list already exists. Eliminating all of these evasions and misinterpretations of policy would allow a real dialog to take place in the same way that making the effort after all these years to define what makes a category "defining" would short-circuit much wasted time and effort. Alansohn (talk) 22:42, 5 May 2009 (UTC)

• I'm not sure what you're responding to here, as none of this was said by me. Postdlf (talk) 00:54, 6 May 2009 (UTC)

For whatever it's worth, I agree with postdlf's post (timestamp: 19:17, 4 May 2009). And their subsequent explanations. This is how it's been at CfD for quite a long time (years).

That said, if this isn't clear to the reader here at

WP:CLN, thne perhaps WP:CLN should be clarified in this respect. - jc37

21:36, 5 May 2009 (UTC)

It should be clarified, which is why I made the suggestion. If this is what's gone one for years, there are probably hundreds to thousands of CfDs that were closed based on an incorrect interpretation of policy. Community consensus at DRV -- from every single person who did not participate in the original CfD -- was that your close violated CLN. User:Kbdank71, seems to have the same misperception, voting to "Endorse Jc37's close that this would make a better list" (here). The wording change that I am proposing would only help ensure that further votes for deletion based on the premise that "it makes a better list" are eliminated but that any further improper closes of this nature based on such votes are averted. Alansohn (talk) 22:42, 5 May 2009 (UTC)

I don't want to assume here, so let me just ask. Do you actually know the difference between policy and an editing guideline? And you should probably rephrase your comment "Community consensus at DRV" as it is misleading. "Consensus at this DRV" is more accurate.

And for the record, Postdlf is correct. Had I even the slightest inkling that you would be open to dialog, I would have a) wrote more, and b) responded to you at the DRV. But I've seen you insist (paraphrasing here, don't expect a diff) over and over and over again that the existence of a list cannot possibly mean we can delete a category (because of the CLN "policy"), and I've seen people try and try and try to discuss things with you, all for naught, that I've given up even trying with you; it's easier and less stressful to say what I have to say and walk away. --Kbdank71 23:59, 5 May 2009 (UTC)

If you can find a more current DRV that contradicts this position, I look forward to seeing it. Alansohn (talk) 05:00, 6 May 2009 (UTC)

To clarify where this discussion is at, Alansohn dislikes CFD comments that compare the capabilities and limitations of categories and lists. It seems he dislikes such comments because he thinks that they are based upon the premise that categories should always be deleted in favor of lists. But no one actually believes that premise; no one wants to delete a category without thinking that category is inherently flawed regardless of whether a parallel list does or doesn't exist. Instead, as I have explained above, those category-list comparisons may be helpful to explain category flaws, and to show that when a category is deleted there are alternative means for presenting that information that don't pose the same problems. So ultimately, as expressed here, Alansohn's dislike is with a rhetorical device, not a premise. Or at a minimum, Alansohn has not clarified what premise he is opposed to, beyond the one that no one actually believes. Postdlf (talk) 00:54, 6 May 2009 (UTC)

I've stated the proposed change, so I don't think I could be clearer as to my premise, which I have proposed to reflect community consensus rejecting the use of "would make a better list" as an argument for deletion. I would have assumed that no one actually believes this to be a valid argument, but Kbdank71 actually voted at DRV to "Endorse Jc37's close that this would make a better list" (here). I'll repeat that I don't believe that anyone believes that all categories should be deleted in favor of lists, but the argument that any particular category should be deleted solely because it "would make a better list" has been soundly rejected. I'll give you the diff you're looking for: The existence of a corresponding list is never a valid reason to delete a category; The decision to retain any category should solely be evaluated based on its ability to aid navigation as a category, within the criteria established for "definingness" (a term still left undefined). Alansohn (talk) 05:00, 6 May 2009 (UTC)

As several have indicated here -- your proposition that any particular category should be deleted solely because it "would make a better list" has been soundly rejected. is unsound. That may have been the case in that particular DRV -- but to extrapolate that particular into a universal requires a stronger consensus. Saying a category may be better as a list or vice versa is a valid argument and so far as I can recall has been used with some frequency. older ≠ wiser 13:26, 6 May 2009 (UTC)

I still haven't seen Alansohn explicate the argument he wishes banned, to show that he understands what is meant by it beyond repeating the same five simple words, "would make a better list." Those words in the abstract are obviously not an argument, but they can imply one (see my explanation re: shorthand comments in XFD discussions, supra). And Alansohn has not yet shown why that implied argument is not valid because he has not yet identified what argument he thinks it implies. Perhaps if you were to state it as a syllogism? Postdlf (talk) 17:55, 6 May 2009 (UTC)

The problem with shorthands used in XfD arguments is not that there is an underlying argument that is unstated, but that there is none. As each and every existing category would make a better list, offering the argument at CfD that a particular category should be deleted because it "would make a better list" is an argument for dismantling the entire category system, not for deleting that one category. Pretending that "would make a better list" is a serious argument, without specifying any other explanation, only perpetuates CfD as an ILIKEIT / IHATEIT game. The utter inability to define what "defining" means after all these years only further demonstrates the problems here. Alansohn (talk) 05:04, 7 May 2009 (UTC)

It is not the case that each and every existing category would make a better list and I don't believe anyone other than you has suggested such a straw man or the slippery slope logic that leads to dismantling of the category system. older ≠ wiser 10:55, 7 May 2009 (UTC)

Now we're getting somewhere...though I've already responded to this point above. No one (maybe I'm assuming too much good faith here) believes that in all cases where a list would be better than a category, that the category should be deleted (nor does anyone believe that lists are always better than categories), so that is never the argument that is being made. You should give people a little more credit for thinking things through, as clearly no one wants to dismantle the whole category system. I have explained above what argument is (at least usually) made when people compare a given category with a given list; I have yet to see you actually address the substance of such an argument, and it seems you still deny such an argument even exists, though I have laid it out step-by-step for you. Postdlf (talk) 14:32, 7 May 2009 (UTC)

Almost every title beginning Wikipedia:Navigation.. (or Wikipedia:Navigational...) redirects to

I'm a bit confused by your explanation of the problem:

• On the
  Wikipedia:Navigation templates
  .
• If I search for incoming redirects to
  Wikipedia:Navigation
  .

What am I missing? or, What is it that you want to change exactly? :) -- Quiddity (talk) 19:03, 2 July 2009 (UTC)

• I think there is a problem with this page, and that it rests on these two related facts:
  1. Its title is a list of mechanisms rather than a description of what makes it appropriate to treat them, and only them, in one place; this suggests a page that can't do its job bcz its editors don't know what its job really is and therefore can't make a sound decision about whether the specified mechanisms are adequate for doing the unidentified job well.
  2. There ought to be page where the principles how we intend users to navigate, and how we expect editors to help them do so, and it should be easy to find. E.g., it should not rely on the user knowing that three concepts that are either cryptic or uselessly ambiguous are major means for navigation. It appears there isn't. (I learned that not by consulting this page but by typing "wp:navigatio" in the WP-search box of the Google tool bar, and looking at the pages that it offers as guesses about what else i intend to type; i expect you can get similar results using the WP search page). And IMO this page not only fails to solve the problem; i suggest that its current state (and the Rdr of
     Portal:Contents
     as well) interferes with mobilizing the effort needed to solve it.

I don't intend to change anything, bcz i'm a far better brainstormer than implementer, but i had hoped to inspire some effective implementers to rename this page (e.g. to

WP:Navigation

) and to expand it to

offer an overall presentation of why facilitating navigation by users and by editors is important,

include discussions of the other mechanisms currently in use, and

relate both the previously and newly included mechanisms back to the overall goals.

--Jerzy•t 05:13, 3 July 2009 (UTC)

Ah, that makes it clearer what you're suggesting :)

For context: this page started life as a summarised list of the pros and cons of lists, categories, and navboxes, to prevent repetitive and endless arguments in project discussions and VFDs. (2004 sample diff: "Wikipedia offers three ways to create groupings of articles. Each has its own advantages and disadvantages, and each is appropriate in different circumstances.")

You are suggesting that we retitle it, and also, slightly repurpose it into a comprehensive list of navigational methods. (Currently, it only lists inter-topic navigation methods ("... groups of articles."), and basically still serves its original purpose.)

So, looking around our mess of help pages, I can only find

Help:Wikipedia: The Missing Manual/Building a Stronger Encyclopedia/Getting Readers to the Right Article: Naming, Redirects, and Disambiguation. Perhaps one of those is a better target for a repurposing, or perhaps a new page is needed (I shudder to say it!). (or maybe i missed a better help page that does already list everything. I tried to search for Category list navigation template redirect disambiguation hatnote set index article). Perhaps that will stimulate more thoughts&suggestions :) -- Quiddity (talk

) 07:16, 3 July 2009 (UTC)

Are there any bots or other tools that can help keep these different representations synchronized? Consider Aviation accidents and incidents, {{Lists of aviation accidents and incidents}} and Category:Aviation accidents and incidents in the United States and so forth. As articles are added to (or deleted from) the C, L, and N on an ad-hoc basis, they drift apart. When editors change one, the others should ideally be changed accordingly, whether this is automatic, semiautomatic, or manual. Any suggestions?

This certainly shouldn't be done mechanically. Lists can include redlinks for articles not yet created. Lists can also properly include entries that arguably fit within the list, but for which there may not be a clear consensus among reliable sources, provided that the list explains why some sources believe it belongs. This ability of a list to explain and provide a direct source allows lists to have more flexible inclusion criteria than categories, which should be limited to clear cases. The principle that most people will probably agree governs this, however, is that there should never be an artificially forced consistency between various articles that would override the consensus of editors working on each individual article. The editors of a list may decide that an article subject should be included, while the editors of that article may separately decide that a parallel category does not apply. The only process that should follow from that is the editors of one should talk to the other and see if there's a way to resolve the contradiction, if there is in fact one. Postdlf (talk) 18:38, 16 July 2009 (UTC)

A suggest-bot sort of thing?LeadSongDog come howl 19:48, 16 July 2009 (UTC)