Jump to content

Wikipedia talk:Overcategorization/Archive 2

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 1Archive 2Archive 3Archive 4Archive 5


CleanUp

I did a general cleanup, including:

  • Added headers for easier reference (in a CfD discussion, for example)
  • Attempted to streamline the header names (since they will be linked to)
  • Re-ordered the entries so that similar concepts were adjacent

Several of the entries could use some general wording cleanup, especially the first one. - jc37 06:57, 22 December 2006 (UTC)

WP:VPP repost

To try and substantiate Jeff's claims that

  1. the community wasn't told
  2. consensus isn't respected

I've posted a new notice to Wikipedia:Village pump (policy). Circeus 14:20, 27 December 2006 (UTC)

I tossed it over in proposals, too. --badlydrawnjeff talk 14:21, 27 December 2006 (UTC)
While it was advertised - you guys didn't get much comment. Only a single dissenter Francis Schonken. Advertising is not the issue here, outside *comment* is at issue. We need more, hands down. If more advertising will suffice to bring us that comment, so be it. Fresheneesz 11:19, 28 December 2006 (UTC)

Category:School massacres overcategorized?

Is Category:School massacres overcategorized? It has been split up into "by country" categories, which generally have one or two articles, and sometimes five or six articles in them. This category used to be helpful in that you could see at a glance the 20-30 school massacre articles, but now they are hidden away in the "by country" categories. If I undid all this, and put the categories up for deletion, would that be the right procedure, or should I go to CFD? Carcharoth 16:46, 27 December 2006 (UTC)

I think you should leave it be. I don't believe it's overcategorized at the moment, and an argument could be made that it's undercategorized. --badlydrawnjeff talk 16:50, 27 December 2006 (UTC)
  • (edconf) See also WT:CFD#Discussion about diffusing categories. I'd say that you're right, if there's only 30 articles or so there's no need to subdivide by country, as this makes the organization less comprehensive. At any rate there is no procedure, so you are welcome to undo all that and put the categories up for deletion, which will then be discussed at a better spot than this talk page. >Radiant< 16:53, 27 December 2006 (UTC)
Why not have both the general category for school massacres, and the country by country categories? It serves both needs. Fresheneesz 11:20, 28 December 2006 (UTC)
Really... the pages in subcategories should be automatically added to the supercategories.... Someone needs to bring that up somewhere - dunno where to suggest that. `Fresheneesz 11:29, 28 December 2006 (UTC)
Perennial proposal. If you think it through, you'll notice why it becomes widely unpractical. For one, it would make such categories as Category:Actors entirely unwieldy if they contain everything in all subcategories (this also touches on performance issues). For another, it would lead to redundancy, in that articles will indicate "This guy is an American actor. He's also American. And he's also an actor". >Radiant< 11:44, 28 December 2006 (UTC)
I've nominated them all for deletion. There are not enough articles to warrant a split on any sort right now. Circeus 14:25, 28 December 2006 (UTC)
So would you rather that "american actors" NOT be american, and NOT be actors? That makes no sense. Performance issues aren't an issue for policy - as has been told to us by the people that run wikipedia. What is *really* at issue is usability. If you have to search through every subcategory of a category to find something, that might take hours. If its all there in the category you're looking for - well... much easier. It wouldn't even be a performance issue if there was a button to press to load a list of all items in every subcategory in the supercategory. Then only the people that really wanted a huge list would get it. Fresheneesz 00:14, 30 December 2006 (UTC)

Guideline-worthy

As far as I can tell, this page describes current practice at CFD and therefore reflects the current consensus of where to draw the line regarding overcategorization. -- Rick Block (talk) 17:21, 27 December 2006 (UTC)

But is it a recent consensus at CfD, or a larger consensus of the community? --badlydrawnjeff talk 19:56, 27 December 2006 (UTC)
I believe this article in general has larger consensus of the Wiki editor community. Note that consensus doesn't mean "unanimous"; there will always be a small number of editors that disagree with any guideline. But most of this guideline seems to reflect the general view of most of the editors who contribute to cfd discussions, from what I can tell. Just my opinion. Dugwiki 20:06, 27 December 2006 (UTC)
There are certainly positive comments on this talk page from people who seldom if ever participate at CfD, such as myself. -GTBacchus(talk) 20:08, 27 December 2006 (UTC)
To be more specific, I think this page describes what has been current practice at CFD for many months reflecting the general consensus of the editors who have any interest in CFD discussions. IMO, this is as broad a consensus as is necessary. Putting this a little more harshly, the opinions of editors who do not participate in CFD discussions, although perfectly welcome, are basically irrelevant. If the "larger community" thinks (to pick an absurdly extreme view) all combinations of every word in every article should be a category but does not defend this position with participation at CFD, then creating a guideline reflecting this view would be nonsensical. The point is this guideline reflects current practice, which is the "community consensus" that matters. -- Rick Block (talk) 21:14, 27 December 2006 (UTC)
I disagree that non CFD editors are irrelevant. Thats like saying that the opinion of business men are the only opinions that matter in laws associated with business tax cuts - simply because business men are the only ones that will use them. Everyones affected, therefore everyone's opinion matters. This is not to say I disagree with this as a guideline - i have no opinion right now. Fresheneesz 21:42, 27 December 2006 (UTC)
Presumably, those who care about categorization issues know how to find CfD and express their ideas there. If there are a lot of people who have opinions about categorization that they're not bothering to express, what are we to do? Assume that silence implies opposition? Consensus among those who bother to speak up is pretty much what we mean by consensus here, right? -GTBacchus(talk) 22:10, 27 December 2006 (UTC)
You've been warned, Fresheneesz, if you "have no opinion right now," your arguments are nothing but pointless(or WP:POINT) wikilawyering. Please stop or decide whether there is something you disagree with the guideline. Circeus 22:20, 27 December 2006 (UTC)
Whoa - I don't see Fresheneesz hurting anyone. Surely we can explain why we ignore unvoiced/absent opinions without accusing people of "violations"? We're just people, having a conversation here. -GTBacchus(talk) 22:26, 27 December 2006 (UTC)
Let's put this another way. Is there any evidence that this guideline does not reflect the broad community consensus we'd all like to think it does (i.e. is there some reason to think the self-selected folks who have expressed interest in CFD are not representative of the community at large)? -- Rick Block (talk) 23:28, 27 December 2006 (UTC)
For all his arguing, Jeff has so far failed to provide anything even remotely close to being "evidence." Circeus
Rick, my concern is that the CfD consensus does not match the belief of the community, a community which continues to add categories that may be considered "overcategorization." I've said a number of times that my protests evaporate if this consensus exists. I can handle being a minority in this situation - it isn't the first and won't be the last - but I won't allow this to be steamrolled because of the actions of one or two people. --badlydrawnjeff talk 03:37, 28 December 2006 (UTC)
Individual users who don't know the ins and outs create all kinds of categories, but I don't think this means anything in particular about the "community consensus". If it did, it seems to me this argument would imply that the "community" should accept the dreck that folks create as articles as well. The proof of the pudding is that categories that deviate get nominated for deletion and reasonably consistently get deleted. If this didn't follow the larger community consensus, this wouldn't consistently happen. I'm not sure I understand what you'd accept as stronger evidence of consensus than months and months of CFD history. -- Rick Block (talk) 04:32, 28 December 2006 (UTC)
I'd accept a number of people coming here from the other non-CfD-related areas and affirming that this should be codified as practice. CfD is very insulated, doesn't get a whole lot of traction the way, say, AfD or DRV does, and even then, we're careful not to hold those decisions as any sort of binding consensus. --badlydrawnjeff talk 04:41, 28 December 2006 (UTC)

Just dropping in from RfC/Village Pump to say that I believe this page accurately depicts existing community norms for categorization, particularly CfD. Kudos to all. Cheers, -- Visviva 04:43, 28 December 2006 (UTC) PS. I'm not a CfD regular, though I do pass through there from time to time.

It takes more than silence to imply consent, it seems. Please add me to the list of non-CfD regulars (I've used it half a dozen times, perhaps) who affirms that this page describes good, common-sense, already-existing practice. -GTBacchus(talk) 05:46, 28 December 2006 (UTC)

Cut to the quick

Okay. This likely does have the support of the CFD regulars, and may or may not have the general support of Wikipedia as a whole.

So. Who doesn't like this? Why? - A Man In Bl♟ck (conspire | past ops) 05:37, 28 December 2006 (UTC)

I don't. It makes our categorization system less useful by suggesting to lessen the categories, when it's a great way to expand the information. The essay is based on the opinion that a wealth of categories is unhelpful, which I strongly disagree with. I take the opposite tack - the fewer the categories, the less helpful they can be due to people looking for specific information. --badlydrawnjeff talk 05:43, 28 December 2006 (UTC)
Sounds fair. Does anyone, to your knowledge, agree with you? (Feel free to speak for other people here, BDJ, I'm just trying to figure out what's going on.) - A Man In Bl♟ck (conspire | past ops) 06:03, 28 December 2006 (UTC)
I provided a link to a mailing list discussion where a similar discussion took place. Radiant falsely believes it takes the opposite tack of my own opinion, but i'm not the only one who thinks that overcategorization isn't a problem, or at least doesn't reflect this. --badlydrawnjeff talk 13:39, 28 December 2006 (UTC)
Where is this link? Forgive me, this talk page is a mess. - A Man In Bl♟ck (conspire | past ops) 05:10, 29 December 2006 (UTC)
here. And I think at some point I'll archive some of it. It's pretty needed by now. Circeus 05:14, 29 December 2006 (UTC)
So far, most users have agreed that it started with a comment indirectly related to the topic at hand (a user lament that articles are being moved into more specific categories, which is exactly what BDJ wants to be allowed to do!) and complains people are reverting the duplication of parent/child duplicate categorization). Ensues a a long technical/philosphical/historical discussion of categories in Wikipedia. Overall, it has to do with WP:SUBCAT, not WP:OC.Circeus 05:26, 29 December 2006 (UTC)
I'm not understanding you, Jeff. Further up the page, when I made the comment "There is a basic decision that we all have to make about categorization. It is whether we see categorization as a system of tagging in which a tag may be trivial or important, or as multiple taxonomies in which we are trying to construct an intelligent map of knowledge. As this is a project to create an encyclopedia I think we are doing the second..." you replied, "I'm with you on the second." I am all in favor of adding more categories, and I think we all agree that the issue is not really about the number of the categories, but the quality of the categories. I'm not understanding what you want as an alternative to this page? How do we prevent the situation when instead of having the categories "American actors", "People from Georgia", "Methodists", and "Political activists" for a hypothetical article, we allow categories to expand until we have "American actresses", "Women from Georgia", "Methodists", and "Political activists", "Left-handed people", "People convicted of speeding", "Members of the Methodist hall of fame", "Political activists arrested in 2001", "Political activists arrested in 2002", "Political activists arrested in 2004", "Left-handed American Methodists", "Political activists from Georgia", "Left-handed political activist actresses from Georgia convicted of speeding", etc...? -- Samuel Wantman 08:35, 28 December 2006 (UTC)
I'm not sure what there is to prevent, honestly. If a category can be useful, why are we worrying about it? --badlydrawnjeff talk 13:39, 28 December 2006 (UTC)
  • I guess I support the idea but not the way it is currently documented. This is overly specific and the first thing which jumps out at me is that we're saying a category of Muppet Show guest stars is a bad idea. Now I can see the value in that category, or at least in the fact that we impart that information in some form. Now the cfd there had a small set of participants and really the outcome was listify, not delete. Sadly, since the category has been deleted, it's going to be bloody hard to quickly recreate. So yes, I get the point this page is trying to address, but I don't think this is the solution. Where is the guidance to the effect that some items are better listified? Where is the guidance on how to convert categories to lists? Where is the process for doing that. Yes, categorisation needs to be well maintained by us to be useful to readers, but that maintenance should not mean that useful information is discarded because it is in the wrong form. Maybe the people writing this and using it in practise could consider amending WP:CFD practise to allow a listify result, which will see editors listify the elements in the same way a merge is done at afd, and maybe there should be some discussion as to whether WP:CFD can be expanded to allow lists to be discussed there rather than at WP:AFD, since many times Afd will ask a list be categorised, and then cfd will reverse that decision. Perhaps a list namespace would now be an idea, after all it does have a featured process, and it does lend itself to possibilities; lists could be transcluded into articles in a similar manner to templates where appropriate. This could perhaps see navigational templates hived out of template namespace and allow us to better lockdown template namespace. Just my thoughts. Hiding Talk 12:22, 28 December 2006 (UTC)
    • Listifying is an excellent alternative. There are several references to doing so on the page, but adding more should not be a problem. As a matter of fact we do have a list of Muppet Show guest stars ordered by date - so we do impart that information. I'm not sure we need a formal process for listifying, it is easy enough with some copy/paste work. >Radiant< 13:47, 28 December 2006 (UTC)
    • You know, I searched for the bloody list before I posted this, but our great search function is of course, non-existant and google seems to be using it's usefulness. Yes, I agree we don't need to formalise the listify process overly, but look at what we've got on lists and categories at the moment: I mean, if we could come up with something better than Wikipedia:Categorization, Wikipedia:Categorization FAQ, which might be better in the help space, Wikipedia:Categorization/Gender, race and sexuality, Wikipedia:Categories, lists, and series boxes, Wikipedia:Categories vs lists and that one we helped draft, um Wikipedia:Naming conventions (categories) as well as this one, I'd be happy. A lot of this stuff was drafted when categories were implemented, and we didn't have a clue how to use them. Look at Wikipedia:Categories, lists, and series boxes; by now we should have worked out roughly when it is better to listify than to categorise, and don't get me started on serial boxes. Do we even call them that anymore? Look, maybe a few dedicated souls might like to form a working party and rewrite what we've got on categories and lists to reflect current practise and thinking. I seriously don't think that stuff has been revisited properly in over two years. We've had new namespaces and an explosion in user categorisation since then, at the very least. And I'm not after references within all the little examples, but big giant letters at the start: I don't know, Categories are useful tools for grouping similar articles, but a swamping of an article in categories is often counter productive. Sometimes a category can be inappropriate: when dealing with information which is better presented in an annotated and referenced form, use a bloody list like this: explain how a list works, or link to good guidance on lists, and then give some examples like The muppet show or discographies or bibliographies, they are better done as lists because we can do publication dates and so on. Anyway, my two cents. Hiding Talk 14:07, 28 December 2006 (UTC)
Jeff, you're basically saying "any factual categorization is good categorization and must be kept," How 'bout you actually look at WP:CFD and see how that work out? Circeus 14:36, 28 December 2006 (UTC)
  • As a new User, my experiences thus far are in sync with what Hiding and Radiant have said. I think that lists are underused and categorization, over-used. People seem to be category-happy. I don't know if it's an ego thing ("this is so important it must be its own category"), or lack of knowledge or understanding about other options. In support of the latter theory, I agree that there is definitely a lack of guidance or policy about lists (or series boxes). I appreciate the Muppets example mentioned above because I've been frustrated by several catergories (or subs) in the Television area that I think should be lists (or even just information included in the primary article). --Vbd 09:24, 15 January 2007 (UTC)

Problems with this guideline

This guideline does not explain why overcategorization is a problem. I think everyone knows that wikipedia isn't a paper encyclopedia, but perhaps have different ideas as to what constitutes useful information on wikipedia. I'd like comment on the hypothetical situation in which Semantic MediaWiki becomes a reality. This would enable people to search for things that *we* may think aren't good criteria - but obviously they do. For example, if I wanted to know if there were any bald effeminate fictional male characters... well this guideline would prevent my ability to find such a thing.

While articles need to be verifiable, categories do not have such restrictions. People who make categories do so with their own hard work - the effort going into destroying people's work is not justified in my mind. Well, i'm glad I finally read the guideline and can make a relevant comment - I'd very much appreciate a (civil) discussion of my points. Fresheneesz 11:07, 28 December 2006 (UTC)

I had to read that several times to make certain I just read what I read. (Please pardon my stunned shock.) Categories very much do have to be verifiable, and in some ways are actually held to a higher standard, since, unlike lists, each entry's inclusion cannot be explained individually. See WP:CLS for at least one set of examples. - jc37 11:19, 28 December 2006 (UTC)
  • The point is you don't have to look at it purely from the side of the categories, but from the articles as well. If an article is in 50 categories, it is entirely unclear from those what it is and what is relevant. Cross-section categories tend to aggravate this problem, as well as being redundant and divergent. Categories aren't intended as semantic metadata, and cannot practically be used as such because they don't do unions and intersections. There is an open feature request for all of this, though. If you really do want to find "bald effeminate fictional male characters", a far more effective way of doing that is using the search engine. That's what its there for. >Radiant< 11:41, 28 December 2006 (UTC)
In terms of astronomy categories, I have seen categories that describe objects that do not exist or categories where the objects are misnamed. See my recent nomination of Category:Local Filament galaxies, for example. Leaving such categories in Wikipedia gives the impression that the objects exist, and it encourages people to create articles for the objects. It is much better to destroy such categories before they cause more problems. Dr. Submillimeter 12:27, 28 December 2006 (UTC)

Why this is a horrible page

Point by point:

  1. Non-defining or trivial characteristic: This is a poor rationale because of the fallacy that "someone's tastes in food, their favorite holiday destination, or the amount of tattoos they have are trivial...is not useful categorization." If done singularly (i.e., only one or two people/things), it may not be useful. If we use something like, say, "People photographed with a bag of Pirate's Booty" (and yes, such a phenomenon did exist), that is useful to people. Trivial is entirely subjective, and has no bearing here - what's trivial in one article is not trivial in another, and we cannot make any sort of useful statement on it here.
  2. Opinion about a question or issue: Patent nonsense. Quote: "holding an opinion is not a defining characteristic." Indefensible, and without any merit - many people are known solely because of their opinions on specific issues, and such categories are useful to find people of a certain persuasion ("People opposed to the Iraq War," "People in support of George W. Bush," "Holocaust deniers").
  3. Subjective inclusion criterion: First, props to whoever went with the absurd categories as opposed to using something that may, in fact, be useful. Subjectivity can still be reliably sourced, and such subjectivity ("Tall basketball players," "Large defensive linemen") is inherently useful as a research tool.
  4. Arbitrary inclusion criterion: Our entire inclusion system is based on arbitrary numbers. Arbitrary numbers and criteria can be based on consensus of editors, and need not be governed by some overbearing guideline.
  5. Intersection by location: More absurdity. "Avoid subcategorizing items by geographical boundary if that boundary does not have any relevant bearing on the items' other characteristics...quarterbacks' careers are not defined by the specific state that they once lived in" is entirely irrelevant. As categories are designed to help sort into useful areas, it's entirely useful to have the ability to sort, to use the quarterback analogy, where quarterbacks are from. Location can also be useful to point out interesting situations, such as (theoretically) "Mormons from Mississippi" or "Conservatives from Massachusetts," where one may believe that a small amount of notable people in such a category would exist.
  6. Intersection by ethnicity, religion, or sexual preference: See above. This is actually more important to have as a categorical structure than anything else listed here - people are interested in knowing who gay performers, Jewish politicians, and black ice hockey players are. Such categorization is entirely useful, whether it be "LGBT literature" or "LGBT quantum physics."
  7. Narrow intersection: Another section created based on absurdity rather than reality. Is "Fictional Black African-American DC animated Superheroes with the power to manipulate electricity" useless? Sure. Are narrow intersections bad? Of course not. "Filmmakers who have not made a war film or a political drama" is "narrow" and "useful" in a way the superheroes one is not, but this "gudieline" would frown on it.
  8. Trivial intersection: See above, the rationale is exactly the same as "Intersection by ethnicity."
  9. Small with no potential for growth: Irrelevant and subjective, ironic given that the rationale for including so many of these line items are based on that standard. "Potential for growth" is boundless for nearly any subject.
  10. Mostly overlapping categories: Again, poor justification. For instance, "Members of the 2004 Boston Red Sox" would be an entirely useful category, as would "Members of the 2003 Red Sox," although there's signficant overlap. Why? Both years are important to the history of the team. There's also no worry of overcategorization for this criteria, as there is a finite end to the amount of years a player can play, and thus a finite end to the amount of team categories that a player can be placed in, especially since each category can be useful.
  11. Award winners and nominees: I'm still disgusted that this is listed. Just plain stupid. Not even worth arguing against, I should probably just remove it.
  12. Published list: This doesn't even fit in with the rest. The CfD result was piss-poor on the example, such lists are inherently encyclopedic if the publisher is notable/well known in the field.

This sucks. Plain and simple. Kill it with fire. --badlydrawnjeff talk 16:46, 28 December 2006 (UTC)

You are right that this kind of information is frequently encyclopedic. However, that doesn't mean that it is an appropriate use of categories, which are a unique and limited (and in many respects problematic) aspect of Mediawiki. In most cases these are far better dealt with as lists, which is why CfDs on such categories frequently end in deletion and/or listification. I honestly don't understand the desire to use categories for these purposes. -- Visviva 17:09, 28 December 2006 (UTC)
Because the search mechanism on Wikipedia is really poor, and doesn't show signs of not being poor anytime soon? Because you can't cross-reference lists the way you can categories? --badlydrawnjeff talk 19:54, 28 December 2006 (UTC)
These are all interesting argumentations. Now how about showing categories fitting them that actually survived deletion (and I don't mean a "no consensus" result)? Until you can do that your saying that the arguments "are subjective" is just as much so, but at least that "subjectivity" is accepted by the community at large.Circeus 18:59, 28 December 2006 (UTC)
Not really, and prior CfDs have no place in this discussion, as consensus can change. --badlydrawnjeff talk 19:54, 28 December 2006 (UTC)
I'd say "no place in this discussion" is a bit strong. They may not be the be-all end-all, but it's certainly not bad for our discussion to be informed by current and past practice at CfD. "He who cannot remember the past...", and all that, right? -GTBacchus(talk) 19:57, 28 December 2006 (UTC)
It has no place because it's not a reflection of the greater community and because we don't do precedent. --badlydrawnjeff talk 05:41, 29 December 2006 (UTC)
There's no potentially useful experience that I'm willing to close my ears to. CfD history is potentially useful experience. You seem to me to be taking a stance that since it's not the only factor to consider, therefore it has no place at all. (Please correct me if that's a wrong impression.) That seems unnecessarily extreme to me - I see a middle ground, where multiple factors are considered, including CfD history. -GTBacchus(talk) 10:38, 30 December 2006 (UTC)
Remember, guidelines are enshrinements of current practice. If you can't reliably demonstrate that the guideline is not a reflection of current practice (and don't even get me started on the mythical "absence of consensus" you attempt toraise), your fine demonstration is nothing but slothful induction: We have provided plenty of occurences of such categories being deleted. How can you claim than that the guideline is inappropriate, hence that it contradicts current practices? And there must be a reason why there so few such categories created in the first place. The obvious one being that people intuitively realize they are irrelevant.
Consensus can most certainly change, but right now, it certainly hasn't changed since the guideline was first proposed, and saying that consensus is irrelevant because it can change? That's a prime example of appeal to consequences, if not a complete denying of the concept of consensus itself! Circeus 22:22, 28 December 2006 (UTC)
Guidelines are not enshrinements of current practice - they are the result of (hopefully) careful discussion, compromise, and consensus. Show me one category that was deleted and I'll show you one category that wasn't. Precedent is *not* valid argument for a guideline. Sure there are plenty of cases that categories are deleted in such a way - therefore there is precedent. But let us not forget that there is HUGE precedent for vandalism, trolling, and personal attacking as well. Does this mean that we should make vandalism something that is supported in a guideline? I think I've made my point clear enough. Fresheneesz 23:58, 29 December 2006 (UTC)
Guidelines are documentations of observed good practice. It's not that every current practice, good or bad, becomes a guideline - that would be absurd, and nobody's claiming that. It's more like this: someone notices that people are doing something smart somewhat consistently, and says "why not write this down, so more people will do it, and so we can more easily explain just what it is we're doing?" That's the model according to which many of our guidelines have come to be. The idea of developing guidelines in discussion and then putting them into practice more closely resembles how legislation works in most societies. We're trying to record good habits, as they develop organically. Do you see how that's different from just "enshrining current practice"? -GTBacchus(talk) 21:40, 30 December 2006 (UTC)
I disagree with your entire premise. Can you actually address my points? --badlydrawnjeff talk 05:41, 29 December 2006 (UTC)
I haven't seen you address my point (i.e. that there is nothing in there even attempting to demonstrate that the guideline doesn't "reflect current practices") either, not here, not in any of the previous comments you've made on this entire page. Oh, and my point right now is that the conclusions you draw from WP:CCC are at best inconsistent, and at worse, nonsensical.Circeus 06:56, 29 December 2006 (UTC)
I have addressed it. --badlydrawnjeff talk 14:14, 29 December 2006 (UTC)
Interesting. I see hypothesis, suggestions and refutations, but not a single word about how things are currently done. Yhat you "have addressed it." is at best untrue.Circeus 14:33, 29 December 2006 (UTC)
Then you need to keep reading back until you see it, I guess. --badlydrawnjeff talk 15:01, 29 December 2006 (UTC)
You have refuted the guidelines, not the reasons why it has become a guideline. i.e.
  1. That it is actionable
    Yes, it is. It doesn't mean we should do it - we could make a guideline banning articles about redheads who appear in pizza commercials, too.
  2. That it is agreed by consensus
    Not sure that's the case.
  3. That consensus can change doesn't means it is changing right now
    No evidence.
  4. That is enshrines current practices
    No evidence.
  5. That it leave room for flexibility (WP:HCP)
    It doesn't.
  6. WP:CFD still works case-by-case, and exceptions might continue to arise.
    More reason not to have this.
  7. It is in agreement of other guidelines and policy. It's parent, Wikipedia:Categorization, states
    • Restraint should be used as categories become less effective the more there are on any given article. (echoed in Wikipedia:Categorization of people:Try to limit the number of categories.)
    • Articles should not usually be in both a category and its subcategory. (this is not universal, but does apply in a vast majority of cases)
    • Categories appear without annotations, so be careful of NPOV when creating or filling categories (pertinent for subjective categories and those by opinions)
    • Consider whether a list or other grouping technique would be more appropriate for trivia [...] such as "dog owners (Wikipedia:Categorization of people)
    Then we don't need this, as it's redundant and goes too far.
All of this you have very careful tiptoed around. Ignoring the elephant in the room won't help you convincing anybody.Circeus 15:47, 29 December 2006 (UTC)
I haven't tiptoed. You've simply decided my opinions aren't worth hearing, since I've only been "wikilawyering" the guideline anyway. --badlydrawnjeff talk 16:08, 29 December 2006 (UTC)

Jeff, I'll try and address your points.

  1. Nobody is saying that this information is not useful to some people. What we are saying is that this information is not well suited as a category. This information can be found in an article or list. The list and article can be linked whenever the fact that that the person photographed with a bag of Pirate's Booty is mentioned in their article. Is this such an important fact that it should be emphasized by having a category at the bottom of the article? What is wrong with just having this linked? And, if it is not mentioned in the article, why would it be listed as a category?
    For sorting purposes. For ease of searching. For further information that wouldn't fit nicely in the article's prose. badlydrawnjeff talk 14:14, 29 December 2006 (UTC)
  2. I don't think this is patent nonesense. It isn't the opinion that makes the person noted, but their actions. They are activists, politicians, philosophers, theorists, writers, etc... I have some wonderful opinions but that doesn't make me notable. Also, consider what would happen if categories like this matured so that virtually any opinion could be noted on every article. Politicians especially would be in hundreds of categories. I suspect that there might a small handful of categories that might be exceptions to this that might be ok, but even in those cases, it might be better to have a list instead of a category. Since this deals with opinions, and opinions are rarely black or white, there should be annotations and citations that explain each person being mentioned. This can happen on a list, and not a category.
    I think you're incorrect on this. Again - it's hard to sort a list the way a category can be sorted, and our search mechanism isn't good. Oftentimes, it is the opinion that makes someone noted, and sorting them based on this is fine. And if it's hundreds of categories, why is this bad? It's useful, it helps the reader... --badlydrawnjeff talk 14:14, 29 December 2006 (UTC)
  3. This is essentially the same as the one before. I think this is much better as a list. Tall basketball players can be listed by their height, which is much better than having them listed alphabetically. Also, consider how many categories like this might be added just to athletic categories.
    As above. --badlydrawnjeff talk 14:14, 29 December 2006 (UTC)
  4. I don't buy this argument at all. The process here is not at all arbitrary. We are all hashing this out in good faith. If this were arbitrary I would discuss this with Radiant, post the results of our discussion, then protect the page, and ban anyone who objected. Arbitrary is essentially the same as POV.
    Then we have a fundamental disagreement on that, and that's okay.
  5. These things might be interesting, but are they encyclopedic? If you pick up a book on guarterbacks do you find tables listing where everyone was born? Where do you draw the line? Should we have categories for people by their High School GPA? by the number of cavities they have? by their shoe size? Is interesting the criteria by which we make these decisions? I don't think interesting is enough. Where would you draw the line? Should we just poll people to see what their level of interest is in the category or try to come up with some criteria like this one. I think this is one of the worst offenders. Many of the professions by locations categories seem to be overcategorized in my eyes. The worst part of these microscopic geographical subcategories is that the larger categories get diffused into them. Then instead of having a good category for film directors we end up with film directors by nationality which is often meaningless (though not always), but it seems you wouldn't mind if these were broken down further to "film directors from Toledo" and similar. It might be interesting to see how many film directors are from San Francisco, but since categories like this are always getting diffused (that is another discussion), having these categories make the larger ones less useful.
    Of course they're encyclopedic. If they weren't, why mention those facts in an article to begin with? As for making the larger categories less useful, we already do that and no one has a problem with it. We have the hierarchical system on purpose. --badlydrawnjeff talk 14:14, 29 December 2006 (UTC)
  6. LGBT quantum physics? What is that? Just because categories have an intersection is not a valid reason to make a category out of it. There are proposals in the works for making any category intersections you want, but they require changes to the Wikimedia software. With our current system, there has to be some constraints. Look at an article like Hank Aaron and calculate how many intersection categories are possible. Do you want all of them? If not all, how do you decide which should stay and which should not? Many people spent weeks working on the guidelines for how to make some of these decisions for ethnicity, sexuality, gender, etc... I don't think the results are perfect, but probably nobody else does either. I do think things have been much better at CFD since the guidelines were created.
    So if we have some good guidelines in other areas, we don't need to govern it here, now do we? And yes, for the record - if a good category can be pulled for intersecting at Hank Aaron, we should be doing it. --badlydrawnjeff talk 14:14, 29 December 2006 (UTC)
  7. Useful to whom? I think there might be wikipedians who find "Fictional Black African-American DC animated Superheroes with the power to manipulate electricity" useful. You think "Filmmakers who have not made a war film or a political drama" is "useful" in a way the superheroes one is not, but I think they are both useless. And there lies the problem. If a category is only useful to a very limited number of people, its presence as a category says that collectively, we think this category is important. This goes back to what I was saying higher on the page about categories being a collective attempt to create multiple taxonomies of knowledge. Categories like this trivialize that effort, and that might be why many CFD regulars like myself are trying to create these guidelines. We could all abandon this effort and just allow any article to be tagged with any category that anyone feels like creating. Is that what you are advocating?
    I don't think it would be a horrible thing in most cases, no. --badlydrawnjeff talk 14:14, 29 December 2006 (UTC)
  8. As you say, see above.
  9. Some categories do have a very small limited number of possible articles. There is no point in having a category when it has nothing that isn't easily found from an article. What is the point of having an eponymous category that only has the single eponymous article as a member? It is almost like we are playing "made you look" with our readers. "Ha, ha! fooled you, nothing here". This seems like a waste of time for everyone involved.
    I'm not really buying that argument, sorry. --badlydrawnjeff talk 14:14, 29 December 2006 (UTC)
  10. We recently removed categories about each All Star team from ball players. It removed about a score of cateogies from Hank Aaron and others. Now if you look at his categories, it is much easier to read than before. This information is returning in the form of articles about each All-Star game that include the rosters from each team. This seems to be just a more efficient way to handle these sort of categories. Instead of having a category for each session of the US house of representatives, there can be a single category with everyone. There can still be lists for this information, and you'll find those lists in the same parent categories, so the only thing lost is a huge amount of category clutter. What is gained is a better way to organize the information.
    I disagree. I don't know if it's better per my argument above. --14:14, 29 December 2006 (UTC)
  11. This is very similar to the last one. Consider the 71 awards and 52 nominations received by Brokeback Mountain and imagine a category for each one. Is this useful?
    Undoubtedly. --badlydrawnjeff talk 14:14, 29 December 2006 (UTC)
  12. There are thousands of these lists. Which do we keep and which do we discard? Many of these are created as a form of advertising or as a way to bolster readership of magazines. What are your criteria for notability for these? --Samuel Wantman 07:59, 29 December 2006 (UTC)
    If the publication is notable, perhaps? --badlydrawnjeff talk 14:14, 29 December 2006 (UTC)

Jeff, why wouldn't lists be better than categories for a lot of these? Hiding Talk 14:28, 29 December 2006 (UTC)

Lists may be better when our on-wiki search is better. Right now, we're constrained by the technology. --badlydrawnjeff talk 15:01, 29 December 2006 (UTC)
Which is why they should (as a rule) be linked from the articles that are included in them, either in the body of the text or in the "See also" section. -- Visviva 17:11, 29 December 2006 (UTC)
Radiant says taht there are more than 4 billion possible intersections for Hank Aaron. However, since these intersections must be done manually, only ones people think are useful will be made. Therefore, 4 billion intersections *will not* be made, but perhaps a hundred or so could be made by people who think they are useful. Destroying people's work is not what we should be doing at wikipedia. Fresheneesz 00:19, 30 December 2006 (UTC)
With that same logic one could also say that only useful articles would ever be created, and thus we shouldn't have article guidelines or a deletion process. The useful categories that might be caught in the fire of this guideline are very much a minority and don't offer enough benefit for the chaos / damage(whatever you want to call it) caused by the current mess. You are asking us to sacrifice quality in one area for a very small benefit that could be done via lists and a little more effort. "Easy" is not always a good reason. -- Ned Scott 04:03, 30 December 2006 (UTC)

All I can conclude from this discussion is that Badlydrawnjeff would like our categorization system to be a tagging system, similar to how flickr.com works, with minimal editorial control and few constraints. I can imagine a system of user tagging running in parallel with the categorization system we have already set up. That might not be a bad thing. The user tags could have very minimal constraints to keep out vandalism such as having everything tagged with "penis". I would support adding something like this to the Wikimedia software. I do not think that this is appropriate for what exists now, and I think it would be very destructive to our current effort to move our categorization system in that direction. As this seems to be a difference in basic philosophy about what the categorization stystem should be, I don't think we can be in agreement. The reality of Wikipedia is that it changes very very slowly. The type of change Jeff is proposing is a very big change, and I see very little chance of it happening, short of a software change like the one I described, because there is a very long standing consensus that categories should be limited to important defining characteristics of each article. -- Samuel Wantman 05:13, 30 December 2006 (UTC)

  • I'm not sure I'd class them as "tags" per se, although that's an interesting way of looking at it, and isn't improper. In any case, I don't think I'm proposing any sort of change, or any big change at all, or really making a proposal outside of ending the creeping changes to how we run categories. --badlydrawnjeff talk 16:48, 30 December 2006 (UTC)
    • This is not a "creeping change", the same arguments have gotten categories deleted for months, if not years.Circeus 16:57, 30 December 2006 (UTC)
      • So we've always had a guideline on types of "overcategorization?" --badlydrawnjeff talk 16:58, 30 December 2006 (UTC)
        • No, this is a new page with guidelines on "overcategorization". We started with adding categories about 2 and 1/2 years ago, and slowly worked on categorization guidelines. They have become fairly stable over the last year, but there are still unresolved issues, mostly dealing with category duplication in parents and siblings. Over time, patterns emerged for which types of categories always get deleted and which types always get kept. For about a year there has been discussion off and on about creating some sort of record about these "precedents" so that we don't have to repeat the same discussions over and over. There has been defacto guidelines if you look at the contents of the discussions over time. There is an accumulation of knowledge and experience, and people often reference similar categories that were kept or deleted. Some of these patterns go back to the beginning of categoriztion, and some are still being refined. Radiant! recently took the initiative to create the page. This seems like par for the course for the way guidelines get created. -- Samuel Wantman 01:27, 31 December 2006 (UTC)

Narrow intersection

RE: Fictional Black African-American DC animated Superheroes with the power to manipulate electricity I think it's a real howler--I followed the link because I couldn't believe it was real--but, I might have nominated the cat at CfD for the redundant Black African-American!

Maybe submillimeter could supply a different example for this section--something just as crazy but otherwise correct? Also, remember to (re)place it in the parenthetical too: "(e.g. "Fictional Black African-American DC animated Superheroes with the power to change shape")"

Keep up the good work! --RCEberwein | Talk 17:00, 29 December 2006 (UTC)
If I see a better narrow intersection category, I will insert it into the guideline. (I think other users would do this, too.) Dr. Submillimeter 23:14, 29 December 2006 (UTC)

I've started a mediation request between myself and BDJ, since we're the one mainly arguing each side, and can't seem to come to anything even close to an agreement. I don't know if it is relevant to add anybody else (except possibly User:Radiant!), but I think it,s fit people know about it.Circeus 21:58, 29 December 2006 (UTC)

Awards and nominations

In looking over this section, and searching for good precedents, I notice that this topic may not have been clearly decided. Most minor and not-well-known award categories do get deleted. But there is more than just the Academy Award as an exception, most notably the Golden Globes. The section says this:

In general, the winners of all but the most internationally well-known awards should be put in a list rather than a category. It may nevertheless be useful to note the awards in the article. If an award doesn't have an article, it certainly doesn't need a category (and not every award that has an article needs a category). (The sole exception to this is the Academy Awards, which is currently under debate.) Nominees should not be categorized.

I suspect that in not too much time virtually every award of note will have an article written about it, and these categories will start to proliferate. If there are to be distinctions about which remain and which don't, this paragraph does not make those distinctions. I'm not certain that this has been hashed out yet. If it has, can anyone suggest a CfD discussion that makes this point well, and the result was delete? If it hasn't, we should probably discuss this more and nominate a test case for deletion. How do people feel about having these categories for the Golden Globes? If the Golden Globe categories were deleted, how can we justify the Academy Award categories? If the Golden Globe categories are to be kept, where is the line to be drawn? -- Samuel Wantman 04:49, 30 December 2006 (UTC)

Honestly I don't have a problem with categories/sub-categories for winners of industry-wide awards. The Nobel Prize, The Pulitzer, the Oscar, even The Newberry Award. I am more concerned about nominess and cantidates than actual winners. Though obviously we should clarify that "industry-wide" doesn't include: The Springfield High School 12th annual Motion picture awards given by Mrs. Jane Doe's 5th grade class (the Springies : ) - jc37 10:40, 30 December 2006 (UTC)

  • I've thought for some time that the Academy Award nominee categories are ridiculous and should be lists (compare List of Best Supporting Actor nominees to Category:Best Supporting Actor Academy Award nominees). Now that tables can be sorted by various columns, I think the argument that such lists should be categories is even less tenable. Even the category for the winners (for the Academy Awards at least) contains far less information than an equivalent list. I've seen categories redirected to lists (one of the Star Trek episode lists?), which allows users to navigate to the list by clicking on a "category" link, but I really don't understand the seemingly very strong preference for categories over wikilinks in the article itself to much richer lists. -- Rick Block (talk) 15:07, 2 January 2007 (UTC)
Well, I seem to recall several semi-recent CfDs resulting in "Keep the award, remove the nominees". So this shouldn't be a problem (Any one who can find the examples would be greatly appreciated : ) - jc37 11:49, 4 January 2007 (UTC)
  • "Nominees of X" is generally bad, and frankly, I think even the categories for "winner of X award" just add unnecessary clutter. Especially with, for instance, prolific actors, prestigious scientists, and so on, they will have lots of awards. Lists are much better -- the person can have the list of their awards, and the awards can have the list of their award-winners. The only two awards that everybody in the world has heard of are the Academy and the Nobel (right?) and I wouldn't do a winner of X award for anything else. --lquilter 14:40, 10 January 2007 (UTC)
  • What are people's thoughts on Category:Presidential candidates? Should there be exceptions the the no candidates rule? -- Prove It (talk) 00:22, 16 January 2007 (UTC)
Maybe, however there needs to be two rules. 1)Only candidates who have officially announced their intention to run so we don't have a category filled with people who might run based solely on media speculation. 2)Separate by country. There is no reason to have the candidates for say the US presidential election with the candidates for the Ecuadorian presidential election. Koweja 00:43, 16 January 2007 (UTC)
  • In general I'm not too happy with cats for people who ran (or even intended to run but didn't) for a political function but failed to get elected, but it'd be reasonable to make an exception for presidents. Just not for the literally dozens of candidates for some function in some town. >Radiant< 10:01, 16 January 2007 (UTC)

A poll

I think this discussion has broken down, largely because (what seems to me, at least) BDJ desires a categorization system at odds with current CFD and categorization practice, and historical CFD and categorization practice, and that he hasn't had a lot of luck convincing anyone his system is sufficiently better to justify changing that practice.

However, there exists the possibility that he's one of many supporters of such a system, and that the others have (for example) been driven off by the heated atmosphere, or that no real consensus supports current practice.

I think a poll might be best, in this case, to establish where people actually stand. - A Man In Bl♟ck (conspire | past ops) 06:41, 30 December 2006 (UTC)

This would appear to be a reasonable way to gauge whether consensus exists, since that seems to be an open question here. No objection here, provided that we remind ourselves that this is only to gauge existing consensus, not to determine or shape it, per Wikipedia:Dispute resolution and WP:DDV. -- Visviva 07:34, 30 December 2006 (UTC)
That sounds like a good idea. Fresheneesz 10:26, 30 December 2006 (UTC)
It all depends on how the poll is formed. If you're suggesting a straw poll based merely on the question of BDJ's perspective of Categories, and whether it has consensus of any kind, then I think I'd agree to such a straw poll. But in fairness, please allow User:Badlydrawnjeff to write the description of his perspective. - jc37 10:31, 30 December 2006 (UTC)
What purpose would this serve? We know that guidelines are not "enacted" by voting on them (see WP:POL). We know this is a description of current practice (once more, see WP:POL). And we already know where people stand, because they have said so on this talk page, which is hardly excessively long, nor particularly heated in atmosphere. So what's the point? >Radiant< 11:05, 30 December 2006 (UTC)
I very much disagree with that. Of course guidelines are not "enacted" by voting, but a straw poll is not a binding vote.. As you well know. However, we do *not* know where people stand, because not everyone comments on this talk page. Polls are useful because they are very easy to participate in - you don't need to follow a debate half a year long to figure it out. You obviously believe that there is no controversy, and that everyone except me and jeff agree wholeheartedly with this page as a guideline. However, some may not see so clearly as you - a poll will help clarify. There is no harm in a well constructed poll. Fresheneesz 22:11, 30 December 2006 (UTC)
Saying "there's no harm in a well-constructed poll" is something of a tautology, isn't it. I mean, if it's constructed so well as to cause no harm, then there's no harm in it, but... there's no content to that statement. Any poll that presents a dichotomy tends to reinforce that dichotomy. Reinforcing a dichotomy undermines worlds of possibilities of compromise, synthesis, reinvisioning, etc. It also creates an "underdog" and attracts those who champion underdogs for the sake of it, rather than on the case's merits. That can be enough to create an appearance of controversy where there wouldn't naturally be any. So, one tries to write a poll that avoids dichotomous thinking or presenting of "sides" to take.... continuing down this road, the best poll is indistinguishable from a discussion, so why not cut out the the middleman and just discuss?
Fresheneesz, you've been making the point that we need more input. I won't disagree. Is a poll a way to generate input? Only a certain kind of input. At some point, there is always selection bias, in any discussion, any election, any decision: the decision is biased towards those who choose to take an interest and participate. Polls are biased towards those who !vote in polls.
I've seen situations that call for polls, but I don't think this is one. It's not even clear to me what would be asked in a poll here. To address the problem of information being scattered all over the talk page, one solution is creative refactoring - you could arrange the core of each discernable argument into some kind of summary, and add that to this page instead of a poll. People could edit it, and we might learn some stuff about each others' minds without the disadvantages built into surveys and polls. There's a million ways to get things done - polling is a high-friction way around here, so why not explore alternatives? -GTBacchus(talk) 23:32, 30 December 2006 (UTC)
  • Once more, "Fresheneesz has shown by his comments that he fundamentally misunderstands how Wikipedia treats policy, and how it is created. He has stated that guidelines need not be reflections of common practice, and that Wikipedia resolves discussions through the use of voting". That is not a personal attack, it is a statement of fact. In the past, you have made several incorrect assumptions about how Wikipedia works. The penultimate authority on Wikipedia has pointed this out to you (as have others), and yet you stubbornly cling to your false notions anyway. That is not a constructive attitude. >Radiant< 23:37, 30 December 2006 (UTC)
    I agree it's not a personal attack, but it may be nearing its point of diminishing returns. -GTBacchus(talk) 23:46, 30 December 2006 (UTC)
Fact or not - it is, once again, entirely irrelevant to this argument. Stating the opinion of less than 10 people, and constrewing it as fact, is not a constructive way to continue this discussion radiant. To GTBacchus - while you're right that polls are biased towards those that vote in polls. But my thinking is that those people who vote in polls are also those that discuss in depth - plus a whole lot more who don't neccessarily want to discuss in depth. Fresheneesz 06:07, 31 December 2006 (UTC)
  • If you consider the outcome of an ArbCom case to be "the opinion of less than 10 people" that underlines the fact that you are seriously mistaken about how Wikipedia works. >Radiant< 15:50, 31 December 2006 (UTC)
Fresheneesz, the comment about polls being biased towards people who participate in polls was by far the least important part of my message. I'd hate to think you took that to be the content I was trying to communicate.
For what it's worth, I do agree that polls tend to draw those who prefer a shallow level of participation. If people who are willing to invest in a deeper participation have more voice in decisions... that seems right to me. If polls lower the bar so that people who can't be bothered to discuss have more influence, that's another reason to avoid them. Wikipedia is not a "pigerocracy" - rule by the lazy.
Like I said though, that's the least significant point I was making. I'm much more interested in your response to my question above, "why not explore alternatives?" Especially at this stage, a poll would be premature.
The guideline tag may be premature, too - I can't see why people rush to tag pages. This has been around a month - we ought to initiate a custom that a WP namespace page has to sit on the servers for three months and have 40 unique talk page contributors before a guideline tag can go on it. And no, I'm not suggesting that should be a "rule"; I'm suggesting we grow out of the fetishization of rules. -GTBacchus(talk) 08:40, 31 December 2006 (UTC)

Fresh and Radiant, I'm not sure either of you has anything new or particularly useful to add to a discussion of how to construct a poll, at this point. Could you both please voluntarily refrain from commenting further?

Such a poll would be simply on the subject of BDJ's category tagging ideas. - A Man In Bl♟ck (conspire | past ops) 23:35, 2 January 2007 (UTC)

Agreed. I also agree with GTBacchus - this page was tagged too quickly. Fresheneesz 02:32, 6 January 2007 (UTC)

Another attempt

User:A Man In Black said above:

  • "Okay. [...] So. Who doesn't like this? Why?"

Other than User:Badlydrawnjeff, I don't think we've seen any other direct responses to this. So I'd like to voice the question again. Anyone else out there who's been possibly just reading and not commenting, who perhaps disagrees even in part? - jc37 10:49, 30 December 2006 (UTC)

Different angle

I've been thinking about this a little bit. I initially found the idea very appealing, and everything on the page seemed like good sense to me. Badlydrawnjeff has raised some arguments that I think are worth considering, whether or not I agree with him in the final analysis.

When I think about the idea that any "potentially useful" category is a good category, the image that comes to my mind is that of an article with 60 categories at the bottom. It sometimes happens with user pages, and we know what it looks like on the screen. I think it looks quite bad, and is bad for readers, because the categories are all just listed together in that box, with no sub-organization in their presentation.

The human mind has a hard time dealing with any information that exceeds a certain threshhold without being arranged in some way. I think it's important to keep the list of categories brief, so a reader may scan through them easily, or else to come up with a completely different way to present categories, as some kind of meta-data that can be worked with, and not just a soup of phrases like "Left-handed people", "Hungarians", "Left-handed Hungarians", "Men", "Hungarian men", "Architects", "Left-handed architects", "Left-handed Hungarian men", etc, etc.

What I'm really thinking of is that there should be some kind of economy of categorization. Perhaps thinking of this page, not negatively in terms of "overcategorization" to avoid, but positively in terms of "economy" to pursue, then we can better focus on what "economy" should mean in this context, and how to achieve that.

Opinions? -GTBacchus(talk) 23:18, 30 December 2006 (UTC)

  • Interesting idea. See also this higher up on the page, which is thinking along the same line of focusing on "good" cats rather than "bad" cats. >Radiant< 23:44, 30 December 2006 (UTC)
    • I made a similar suggestion above, about having both tags and categories. I'd have no probem with that if the tags were clearly presented as a user based thing, and not an attempt at creating taxonomies. -- Samuel Wantman 01:40, 31 December 2006 (UTC)
      • Tags cannot be a taxonomy, because they cannot be hierarchical.Circeus 02:19, 31 December 2006 (UTC)
        • What I meant was that the tags were not trying to duplicate the function of the category taxonomies and the categories were not trying to duplicate the user tags--that both were for different purposes. I don't want to replace one messy system with two messy systems! -- Samuel Wantman 07:11, 31 December 2006 (UTC)

Categories based on names

May I suggest a new entry: Categories based on names. It seems like this is done frequently, but it often fails to work well. Take, for example, the WP:CFD nominations of Category:North, Category:South, Category:East, and Category:West, which discuss articles with one of these words (or variants on these words) in their name. This is also not the first instance where this has happened; many other inappropriate categories based on the surnames of unrelated people or based on the names of unrelated places have been created in the past. (My personal favorite was Churches named for St. Dunstan.) An exception should be made when the category refers to articles that are directly related (such as Category:Brahe).

What do other people think? Should I draft a guideline on this? Dr. Submillimeter 09:48, 3 January 2007 (UTC)

Of some interest might be Wikipedia:Categories for deletion/Names, a forgotten subpage which nonetheless seems to have been consistently upheld (most recently in the outcome of the bitter Chinese-surname CfD, which I dearly wish someone else had started). Although these are covered by the "trivial characteristic" heading, they come up often enough (and their triviality is contested frequently enough) that a separate heading would not be amiss. -- Visviva 11:11, 3 January 2007 (UTC)

Occupation by religion categories

I propose that this be expanded to cover religion as well. It seems to me that there has recently been a proliferation of Occupation by religion categories where the religion has little or no relevence. See my recent nomination of Actors by religion as an example of this. I see such categorization as mostly divisive. And yes, I was and still am very disapointed with this outcome. -- ProveIt (talk) 18:20, 3 January 2007 (UTC)

You got your wish, it's gone.--T. Anthony 12:49, 1 February 2007 (UTC)
Well I see it as mostly just reflecting a reality. There is things like LDS cinema and therefore it's plausible some of the actors are essentially Category:Mormon actors. Granted I proposed renaming that something like "Actors in the LDS cinema" but got shot down. Still as long as there is things like the Vatican Observatory or Christian film studios it seems sensible to think there will be Catholic scientists and Christian actors. When possible some kind of rename should be done though to make the restriction clear.--T. Anthony 13:22, 6 January 2007 (UTC)
To clarify, the problem lies in people misusing the parent category Category:People by religion and occupation. The description does currently specify that it should only be for people who notably integrate religion into their occupation. So using directors as an example, it shouldn't include an director who is religious in their personal life but who doesn't particularly incorporate their religion into their movies. On the other hand, Mel Gibson or Steven Spielberg are both known in part for incorporating their religious beliefs into their movies, so they would be good candidates for putting in a subcategory of Category:Directors by religion (if that category existed).
The main thing is that in some cases, religion does play an important part in someone's professional career. But in a lot of cases, a person's religion has no effect on their career and so doesn't need to be categorized in a "religion/occupation" pairing. Dugwiki 18:41, 3 January 2007 (UTC)
I suspect almost every director incorporates their religious belief into their movies. Hitchcock is an example which immediately comes to mind where this has been widely written about. If there are to be any categories like this, there has to be clear distinctions made about what is acceptable as a category and distinctions about deciding who belongs in those categories. If this cannot be done, the category should not exist. Lists are perfectly suited for these topics. -- Samuel Wantman 20:37, 3 January 2007 (UTC)
The only issue here with lists versus categories is that a parent category already exists - Category:People by religion and occupation. Since that category is in place, it only makes sense if it is (per the name) divided by occupation and religion. Basically as long as that parent category exists, so will these subcategories.
The parent category was discussed on cfd Sep 12 2006 with a result of "no consensus". So unless it is rediscussed on cfd and deleted, not having subcategories for it isn't viable. The only option currently is to make sure the subcategories are properly limited in scope. Dugwiki 20:58, 3 January 2007 (UTC)
Also, in regards to "every director incorporating their religious beliefs into their movies", the key word here would be "notably" doing it. It has to be something that is notable enough in the director's movies to be verifiably included in their article. It's certainly possible that Roger Corman let his religious beliefs somehow trickle into Frankenstein Unbound and other movies, but since it's not mentioned in Corman's article he wouldn't be a good candidate for a directors-by-religion category. By contrast, Steven Spielberg's article mentions his Jewish heritage and beliefs and some of the impact it's had on his movies, so he'd be a good candidate for the category.
Thus I think the trick is to insist on enough notability and verifiability about a person's religious beliefs impacting his career that the information appears in his article. If something isn't worth mentioning in the article, or can't be verified, it shouldn't be used to categorize them. Dugwiki 21:08, 3 January 2007 (UTC)
I think that should be the minimum standard. I would like to apply Uncle G's prime notability criterion to these things. -- Donald Albury 23:29, 3 January 2007 (UTC)
Such a standard would result in a Delete from me. Simply because application of that standard would require citations/references, which means it should be a list, not a category. That aside, I believe that this is already covered under WP:OCAT#Intersection by ethnicity, religion, or sexual preference? (And by the way, considering so many of such discussions have resulted in "no consensus", I wonder if we're being accurate to list that section on WP:OCAT...) - jc37 11:44, 4 January 2007 (UTC)
To clarify, only information present in an article should be used to categorize it, period. If an article makes no mention of some bit of information, that information should not be used to categorize it. And, since all information in an article should theoretically be verified by a provided reference in the article, it follows that therefore all categorization should theoretically be based on cited information as well. This applies to all categories, not just this particular religion/occupation topic.
So requiring that an article make cited mention of someone's religion to categorize them by religion is actually the normal standard.Dugwiki 17:49, 4 January 2007 (UTC)
Thank you for clarifying my comment. -- Donald Albury 17:36, 5 January 2007 (UTC)

Just to add fuel to the debate, I nominated Category:Roman Catholic scientists for deletion. Feel free to comment. Dr. Submillimeter 11:14, 4 January 2007 (UTC)

  • You seem to have succeeded as it apparently lost the CfD. It was really an uphill battle on "my side" and I think I got too involved. Anyone maybe it's for the best, have a pleasant evening.--T. Anthony 03:10, 11 January 2007 (UTC)
  • Adding my three cents: (1) The entire category tree "People by religion and occupation" says, in its name, that it is simply an intersection. Defining the category to include only an invisible super-notability or super-relevance is not managable by those who are actually categorizing -- people will, inevitably, just add the category if it's an accurate cross-section, and will want to know why it was removed, and it will generate lots of discussions on lots of pages. A non-intuitive definition of a category is not workable.
... (2) If we want to use this category as an intersection, that's another thing. I think it's a bit of fluff but certainly "religious belief" or lack thereof is a highly notable thing about people, as notable to some people as their nationality. So I'm not sure why occupationXnationality is acceptable and occupationXreligion is not. (I'm not really arguing for the category as a cross-section, because I don't love it; just pointing out that it's not that different from other acceptable cross-sections.)
... (3) I don't have a problem with trying to pick up the folks this category is truly intended to capture (the ones whose religion is notable to their occupation), but it needs a less-intuitive but more accurate name. <g> Something like, "Catholicism-influenced filmmakers" or "Producers of Christian evangelist films". (I know that X-influenced is not good, but that's what folks are talking about whether they use that term or not.) --lquilter 15:07, 10 January 2007 (UTC)
  • I'd like to restate some of these points from a different perspective. I think we have to distinguish between the "x is a y" categories, which I call "index categories" and the "x discusses y" categories which I call "topic categories". Articles are put in topic categories only when the article discusses the topic. So if the topic were "Religion-influenced film", it would might make sense to put an article about a religion inspired filmmaker in this category, but only if the article contains a significant discussion about the topic. If a category is an index, it should include every item that belongs in the index of which we have an article. If we don't want to include everyone that belongs in an index category, it is probably a bad category by design and should be deleted. It is also for this reason that it makes sense to have some topic categories like "Racism", but not the corresponding index categories like "Racists". Even if we make this distinction between "topics" and "indexes", we should still be extremely hesitant about putting people in topic categories, if only for the reason that people will want to diffuse them into an index category. -- Samuel Wantman 08:23, 11 January 2007 (UTC)
I take issue with point 3 by Lquilter. Not everyone agrees that these religion categories are supposed to be used for people whose religion directly influences their work. I am sure that some people thought that Enrico Fermi did belong in a category on Roman Catholic scientists not necessarily because of his religion's influence on his work but because people want to divide people by religion and occupation arbitrarily. I have seen this elsewhere. See, for example, the arbitrary inclusion criteria for Category:Moravian writers written by Pastorwayne. If the criteria merely included people who wrote on Moravian religious topics, that would be OK. However, the category could encomass Moravians who have written on any topic, which weaken's the category's effectiveness. Dr. Submillimeter 08:52, 11 January 2007 (UTC)
I agree with Dr. Submillimeter that many people use these categories simply to document intersection of notable identity/occupation features. I said "truly intended to capture" because I thought that there must be some policy already existing on the issue -- judging by the numerous times people have raised the "identity was notable in practice of occupation" argument on CFD, that is.
There are several potential meanings for the category structure Identity-X-Occupation or Identity-X-Nationality or variants thereof.
  1. Where the intersection is itself relevant (i.e., "Catholic scientists" where their Catholicism was particularly relevant to their practice of science);
  2. Where the practice as represented by the intersection itself exists (i.e., "Catholic scientists" who practice Catholic science (a nonexistent category as far as I know), or "Catholic writers" who write about Catholicism) --the distinction from the first is a nuance but real;
  3. Where the category represents simply the intersection of identities.
I have problems with the first, because to me it's not intuitive in the category name and is completely unmanageable and unenforceable; I have problems with the second (Dr. Submillimeter's proposal) because I think, ideally, the categories should be named in such a way that makes it clear that they are not merely an intersection, as in, "Writers of Catholic literature", "Scientists of Catholic science" (maybe better to talk about actual instances of religious-based science, like "Scientists of intelligent design").
As for the third category, I think people will keep making these categories, and using them to represent intersections; I think it would be best to recognize that this is something people want to do, use the category names in the obvious intuitive sense (intuitive both by what people keep doing and by the category name itself), and pick out the major acceptable identity categories that are appropriate for various cross-sections (I would suggest, nationality; occupation; time/era; gender, ethnicity/heritage/whatever, sexuality, and religious faiths/positions (to include atheism/agnosticism); and an assortment of other identity categories that are not generally to be included, but included if there are particularly central to that person (e.g., political/philosophical identity (e.g., anarchists; Communists; fascists); disability identity (e.g., Deaf); etc.
And I would set up guidelines that nationality are (generally) diffusable but other cats should be redundant, to avoid ghettoization by identity. --lquilter 23:15, 18 January 2007 (UTC)

TfD nomination of Template:CatDiffuse

Template:CatDiffuse has been nominated for deletion. You are invited to comment on the discussion at the template's entry on the Templates for Deletion page. Thank you. --Samuel Wantman 02:26, 4 January 2007 (UTC)

List and Cat dilemma

A recurring problem is that some people only think in terms of "keep" vs. "delete", and not in terms of in what form the information is most accessible or comprehensive. For instance, we have a few topics that have (1) a list, (2) a category, and (3) a group template; arguably two of the three are redundant, but people may object to removing any of them because they misunderstand this as removal of the information.

For instance, we now have Wikipedia:Articles for deletion/List of fictional actors, a group nomination, all of which now nominated separately. Some (but not all) of these are obviously better kept as categories than as lists. But people don't realize that. Any thoughts on how to deal with this would be welcome. >Radiant< 16:07, 4 January 2007 (UTC)

Just my opinion, but one general rule might be that the larger the number of articles the more useful it is to have it sorted as a category because categories maintain and sort themselves somewhat automatically. A list with a couple of hundred entries requires more manual maintainence and would therefore be more prone to outdated entries. Along those lines, if only a handful of articles are to be sorted then a list of links to those articles can easily be included in the main article, or as a subarticle that would be easy for readers to navigate.
Another consideration is whether all the links to the related articles already appear in a main article, or can be inserted naturally into that article. For example, one of the reasons you don't need to categorize actors by film is that a full cast list for the film is typically included in the film's article. So a reader looking to peruse a list of actors in the film just has to visit the film's article to find all the links he needs. A category would be completely redundant. By contrast, if the articles to be sorted are not all easily linked from a single main article, such as all regular actors who have appeared in any Star Trek, then a category might be a good idea to collect those various links in one place.
Finally, there's the fact that a list can contain extra information that a reader might want which a category can't. For example, another reason a cast list is useful in an article is that it can show not only the actor's name, but also the character they played, or in a TV show the first episode in which they appeared, etc. Categories are only good at one aspect - displaying links of article names. Therefore if the goal is to present auxilliary information, then a list might be necessary.
Just some personal thought on the topic of lists vs categories. Dugwiki 19:02, 4 January 2007 (UTC)
I agree with both Dugwiki and Radiant!. Categories are useful for browsing. Some categories are alphabetical indexes and others help navigate from index to index. They don't present other information, there is no annotations or citations. That is where lists come in. Categories also don't structure information other than alphabetically, this is where serial boxes help. One more point, which is being discussed at WikiProject Films is that if you have multiple category intersections like the ones that we have talked about as beign overcategorizations ,you end up cluttering up articles with way too many categories. If this information is significant, it can be a list instead of a category. -- Samuel Wantman 22:04, 4 January 2007 (UTC)

Life and death

I'd like to see somthing about categorization by life and death. For example, we don't make subcats of Category:Living people. This comes up from time to time. See Living centenarians, Dead fictional characters, Irish Born Catholic Bishops (Living) and todays Living supercentenarians. -- ProveIt (talk) 14:54, 5 January 2007 (UTC)

I am in favor of that. Dr. Submillimeter 16:25, 5 January 2007 (UTC)
There are subcats of Category:Dead people though, such as Category:Deaths by cause and Category:Deaths by year. So living people and dead people aren't categorized exactly the same way. Dugwiki 16:44, 5 January 2007 (UTC)
Sure, adding a date or cause of death category implicitly makes it clear they died ... that's fine by me, but still, we don't allow any Dead dentists, Living librarians, Recently deceased vegatarians, etc. -- ProveIt (talk) 17:45, 5 January 2007 (UTC)
Agreed. I think one way of putting this is that acceptable Dead People subcategories are ones that have to do with information about the death of the person, as opposed to the random fact that the person happens to be dead. For instance, Dead Librarians would be a random intersection because the fact that the person is dead probably has nothing to do with the fact that they were a librarian. But cause of death is directly linked to the fact that the person is dead, so Deaths by cause is reasonable. Dugwiki 17:51, 5 January 2007 (UTC)

Maybe we need to add some overcategorization guidelines. The first I'd add is : Just because one category exists, you don't have to complete the entire taxonomy. If we accept "LGBT writers" it doesn't imply that we should have "LGBT engineers". If there is "Dead people by year" that doesn't mean we should create "Dead people by profession", etc... -- Samuel Wantman 21:15, 5 January 2007 (UTC)

Going by what I said above, "Dead people by profession" would only make sense if a person's profession was somehow connected in a notable way to the way they died. In general, that's not the case - there's no obvious reason to group all dead politicians together, since only a handful of them died in a way directly connected to their political work.
However, this does bring up the somewhat interesting and related category of "Occupational deaths by profession". This category might be appropriate for certain risky professions, grouping for example "Soldiers killed in the line of duty", "Fire fighters killed in the line of duty", "Pilots who died while flying", etc. In these sorts of categories the way the person died is directly tied to their occupation, and it certainly sounds like something notable that connects the articles.
So here again, it sounds to me like one rule of thumb is to look at whether or not the two topics are directly connected. If the category is just combining two otherwise unrelated statistics, it probably isn't a good category. Dugwiki 22:02, 5 January 2007 (UTC)
I agree with this rule. Personally, that is the very reason why I would have supported the segregation and maintaining of category:Living centenarian. Whether or or not a long-lived person is alive is very relevant in my opinion.Circeus 23:39, 5 January 2007 (UTC)
I agree ... these people are notable just because they continue to remain alive. However, I still voted against it, just because like People who are 20 years old, it requires constant maintainence. Plus, it's a subcat of living people ... -- ProveIt (talk) 23:44, 5 January 2007 (UTC)
I'm with you, Circ, in that I think a case could be made for Category:Living centenarian. I think the reasons for not allowing the category are more technical matters in how the category system on Wiki works than in whether or not being alive and over 100 years old is notable. But this is a case where I think if you were to allow say one or two subcategories of Living People to exist, this one would be reasonable. Dugwiki 23:56, 5 January 2007 (UTC)

Performers by Performance et al.

I originally posted this on CfD itself in order to generate some discussion. At the time, there was no WP:OCAT, so at the time I was proposing a speedy criterion. Now that we have this page, however, I think that this would be a good addition. However, I think further discussion would be helpful. I've only made one change - to remove cast members of a Radio/TV series from cast members of a show, due to recent discussions as an exception. - jc37 15:02, 6 January 2007 (UTC)

Performers by performance

Some examples:

  • Performers who have performed <action>
  • Performers who have portrayed <character name>
  • Performers who have portrayed <a type of character>
  • Performers who have performed <a specific work>

-

  • "Performers" can include actors/actresses (including porn), models, singers, dancers, comedians, etc.
  • "Action" examples: a "spit take", "anal sex", a "pirouette", a "runway walk", a "pratfall", a "sword fight", etc.
  • "Character name" examples: Darth vader, or Hamlet. This includes voicing animated characters, such as Donald Duck. This also includes doing "impressions".
  • "A type of character" examples: wealthy, poor, religious, homeless, gay, female, politician, Scottish, dead, etc.
  • "A specific work" examples: "Amazing Grace", "Swan Lake", "Hamlet" (the play), "Why did the chicken cross the road?" (a joke), etc.

-

Discussion
  • I agree that there are pretty much no exceptions I can think of to the general rule against performer-by-performance. Actors portray so many different individual roles during their career that this could easily be unwieldy. Use lists if there is a strong desire to sort out actors who played a particular role. Dugwiki 16:02, 8 January 2007 (UTC)
  • Generally opposed to performers by performance categories. If some great counter-example comes along it can be argued as an exception.--lquilter 15:12, 10 January 2007 (UTC)

I can see an argument for allowing "Performers who have portrayed <character name>" in some circumstance. Isn't there a benefit to readers to have a category that would include William Gillette, Basil Rathbone and Jeremy Brett, all of whom are most famous for their portrayals of Sherlock Holmes? At the moment, if a reader of Basil Rathbone was trying to recall who played Holmes in the ITV television series, he would have to go to Sherlock Holmes, scroll down to the "Adaptations" section, click on the link provided there to Sherlock Holmes in other media, and scroll down to the "Television" section of that article before he finds the name Jeremy Brett. That's four steps. Wouldn't it be better for this reader if a category Category:Actors who have portrayed Sherlock Holmes existed? He would be able to click on the category on Basil Rathbone's page, and scroll down until he sees the name Jeremy Brett, which would probably trigger his memory. Two steps instead of four.

But (I hear you say) the same function could be performed by a list. The problem is that it isn't immediately apparent why a List of actors who have portrayed Sherlock Holmes should be linked from Basil Rathbone — again, the curious reader would have to go to Sherlock Holmes and probably thence to Sherlock Holmes in other media. Contrariwise, it's immediately intuitive why Basil Rathbone should include Category:Actors who have portrayed Sherlock Holmes.

On the other hand, Sherlock Holmes points out that "more actors have portrayed Sherlock Holmes than any other character," so such a category could be quite large. I do see the hazards of this type of category, but I wanted to point out the potential benefits as well. I think that there are some circumstances in which the benefits would outweigh the hazards. —Josiah Rowe (talk • contribs) 10:42, 16 January 2007 (UTC)

It's true that inclusion on a list doesn't necessarily provide obvious navigational cues from the list member's page to the list. I'm still processing through whether it's desirable to have that -- because if the subject of the list is significant to the list member, then it will be separately mentioned on the page, and can be linked from there; and in other cases, "What links here" might suffice. But, that aside, it might be possible to set up a standard subhead structure within "See also"; like, "See also relevant lists". The problem is that such a solution would end up replicating the "What links here" feature, I fear, as many people started adding marginally useful items. --lquilter 23:25, 18 January 2007 (UTC)

Cast members of a show

Support or Oppose: Should Performers by performance (proposed above) be expanded to include:

  • Cast members of <show>
  • Cast members of a show by <location>

-

  • "Show" can be a film, a stage production, etc.
  • "Location" examples: Broadway, Vietnam, Hollywood, Bollywood, The Improv, Australia, the Palladium, Carnegie Hall, etc

-

  • Further question, if kept: Should "show" be specific? For example: A revival of Show Boat, would have a different cast than the original production. Or general (any version of the show)?

-

  • Comment - Due to some of the discussion below, I would like to clarify what I said at the top: This section intentionally does not include Radio/TV series cast members categories. I think they deserve a seperate discussion, and I don't want this discussion to end up "no consensus" because of it. - jc37 22:29, 10 January 2007 (UTC)
Discussion
  • In my opinion there is usually no need to categorize a cast list for a specific show. The reason is that a show's main article should have the cast list in it. So readers who are interested in surfing through article links of actors who were in the cast can easily do so by just going to the main article. Why go to "Category:Casino Royale (2006 film) actors" when you can just go to Casino Royale (2006 film) and see all the same information and see the character they played and other information about the film? Moreover, many actors have 50 or 100 or more film and TV cast credits in their bios, meaning that theoretically if you create individual actor-by-show categories for each you could have 50 or 100 or more individual categories at the end of each actor's main article. (Voice actors such as Tara Strong in particular are extremely prolific.) Therefore allowing the creation of so many categories carries a significant risk of overcategorizing individual actor articles.
Similarly, I am generally opposed to actor-by-television series categories on the same grounds (although this puts me in disagreement with some of the editors such as Tim! who support most of the actor-by-TV-series categories). You can see the cast list of a TV series just as easily from its main article as you can from a category.
The main exception where I would support an actor-by-show category is when you are categorizing regular cast for a franchise of different shows. For instance, Category:Star Trek actors covers a variety of television shows and movies, so pulling all those actors together in one spot can be useful. Thus my rule is if the category only covers a single show where the cast list already appears in the main article, you don't also need a category for it. If the category covers multiple shows with different main articles, though, a category might be useful. Dugwiki 16:17, 8 January 2007 (UTC)
  • I'm opposed to Cast members of a show, whether it's TV, stage, film, or other. I think it leads to too many categories on performer pages; requires non-intuitive and unmanagable definitions to exclude the guest stars, voice actors, and whatever else one wants to exclude; and requires non-intuitive and unmanagable definitions of notable shows to exclude every show that every existed from being added. --lquilter 15:14, 10 January 2007 (UTC)
  • I'm OK with actors by television series, provided it's for regular cast only. No Simpsons guest stars or the like. -- Prove It (talk) 18:02, 10 January 2007 (UTC)
Just to expand on that further, ProveIt, what do you feel makes a television series regular cast list differ from a film cast list? Why handle the two differently? I could take a guess as to why you think that, but rather than maybe guess incorrectly it's probably better if you clarify your reasoning for reference. Dugwiki 18:27, 10 January 2007 (UTC)
It's more than just a one time thing, they often spend years in a single role and it's often what made them notable in the first place ... for example, Alan Alda is in Category:M*A*S*H actors. I guess I'm ok with that because when I think of him, that's the very first thing that comes to mind, he spent 11 years doing that, and it would be the first thing I would say if asked about him. On the other hand, one can become a Simpsons guest star in a single afternoon. -- Prove It (talk) 21:56, 12 January 2007 (UTC)
"Often"? See, that's thinking of the category they way we would like to see it used - only on notable, long-running TV shows with stable casts. The fact is there are a million tv shows with dedicated fans who will want a cat for their show. It's not always going to be Alan Alda & MASH, it's going to be Gina Torres and Ron Glass and Summer Glau and a million other mid-level-recognition (or even low-level recognition) actors who are regular cast in many short-lived tv series. ... And frankly I think it's disproportionate, even as cast, because why favor the cast (the actors) over the crew? the directors? and so on? ... With the current software and category structure performers by performance is too unwieldy and cast-by-TV-series exceptionalism is just giving in to the TV subculture. If the actor is known by the role, like Alan Alda, then they should just go in the Category:M*A*S*H. Any actor so associated with a show will have that option because the show will have its own cat. If we keep out the scores of guests on the show, by restricting them to a "list of guest actors on TV show X", then we don't need to subcat the "Category:TV series" into "Category:Cast of TV series". --lquilter 22:59, 12 January 2007 (UTC)
But what do you do when "Category:Name-of-TV-series" becomes overpopulated? The logical course of action when a category becomes too large is to subdivide it. It seems intuitive to me that there is benefit in having a subcategory for the regular cast of a television program, rather than throwing its cast into the parent category along with other articles on the series. I agree that actors shouldn't be categorized by one-time guest appearances (except perhaps in rare cases, such as when an actor was awarded an Emmy for a guest role on a program), but having categories for regular casts of television programs is both useful and workable. —Josiah Rowe (talk • contribs) 10:57, 16 January 2007 (UTC)
The problem is categorization -- if the actor page doesn't need a "performer by performance" category, then it also doesn't need to be included within the category for that show. The characters on the show should be included, but why the performer? Actors shouldn't be categorized by particular performances, period; whether they are subcategorized by their role with the performance, or whether they are just dumped into the big category. (Maybe you can't have the single-notable-performance category in an individual's page, for that reason -- so strike Category:M*A*S*H from Alan Alda. --lquilter 23:29, 18 January 2007 (UTC)
Well, that does seem to be a logical consequence of making "no performers by performance" an absolute rule. But from my perpective, the notion of saying that Alan Alda can't be included in Category:M*A*S*H or any daughter categories shows the fundamental flaw of this reasoning — it's almost a reductio ad absurdum. The Alan Alda article has a full section on his work on M*A*S*H, which was not limited to his performance (he was a producer of the show in its later years, and directed several episodes, including the finale; he's also credited for the political bent of the show's later seasons). Just because categorization of performers by performance has been extended to ridiculous extremes is not an argument for eliminating such categorization altogether — it really strikes me as an overreaction to the issue. —Josiah Rowe (talk • contribs) 02:24, 21 January 2007 (UTC)
The thing is I'm not convinced that categorizing cast lists is useful or needed. Look at it from the reader's perspective. Let's say I'm interested in looking at articles about actors in the cast of a show. The most likely thing for me to do is to search for the show's article first. When I pull up that article, it virtually guaranteed to have a full cast list in it, including links to the individual actor pages and character descriptions. So all the information I need to find articles about actors in the cast of a show is right there in the main article, neatly organized in a list and labelled with extra information. Why, then, would I want or need to use the category system to do the same thing, especially when categories are only capable of listing actors alphabetically by name with no other information? Moreover, the more these categories proliferate the greater the potential problem of having actor pages with an unwieldy number of categories to sift through at the bottom. Plus having the category essentially doubles the amount of maintainence needed to keep cast lists up to date, because editors have to update both categories and the cast lists in the main article.
Thus I'm not convinced there is any noticable benefit to these categories. They're redundant with the main article's cast list, and there is a potential downside to having them. Dugwiki 16:10, 16 January 2007 (UTC)
Prove It, the problem is not the categories, it is the articles. Having any categories that include TV shows is not a workable solution. I'll give you an example of why. Take Jerry Orbach who is in a Law & Order category or Ned Beatty who is categorized as a cast member of Homicide: Life on the Street, in which he was one of the main cast members in the early seasons. Someone looking at Jerry's categories will wonder why he isn't in a category for Chicago, a role that brought him fame and a Tony Award on Broadway. Someone look at Ned's categories will also notice that he appeared in M*A*S*H (not one of the main characters), and inevitably will wonder, why he isn't in a category for Deliverance? Deliverance was the role that made him famous, and his role on M*A*S*H was not memorable. The TV categories will make sense, but the category listings on pages for individual actors will never look right if only some notable and not notable TV shows are listed, but no other performances. Allowing all performances to be categorized does not work because it will lead to massive overcategorization and clutter. Leaving just TV will make it seem that only the TV performances are important for actors who worked in several media. The only workable solution is removing all these categories from every media. It is also, by far, the easiest system to maintain. -- Samuel Wantman 00:51, 13 January 2007 (UTC)
Well, I've been thinking about it ... you've obviously put considerable thought into this, and I've decided I think you're right. However, I don't think it will be very popular. It's strange ... I consider myself very much an inclusionist, but for categories I keep finding myself on the delete side of the discussion. -- Prove It (talk) 02:33, 13 January 2007 (UTC)
Welcome to the difficult world of a Wikipedian categorist. I too considered myself very much an inclusionist until I started participating at CfD. For months I wondered why Radiant! was always voting "Delete - overcategorization". Now I find I consistantly do the same. Keeping an article at AfD is a much different matter than keeping a category at CfD. A single bad article residing in the bowels of Wikipedia is no big deal. Perhaps it will eventually turn into a good article, perhaps it will stay an uncited stub for years. So what. But bad categories seem to breed more bad categories. A single bad category can put a stain on hundreds of good articles. It is like a cancer that must be removed before it metastasizes.--Samuel Wantman 08:19, 13 January 2007 (UTC)
  • This discussion interests me, and the points that have been made so far have largely convinced me that all the TV performance categories should go. I may have misplaced my earlier comment above, but I meant to point out here that Wikipedia has already apparently identified 32,000 career defining performances by notable actors. So apparently this cancer has grown quite large already. zadignose 06:58, 17 January 2007 (UTC)

Overreaction

As I commented above, the notion that all performers by performance categories must be eliminated from Wikipedia seems to me an overreaction to the problem. I will grant that it's silly to have categories for everyone who's provided a voice for an episode of The Simpsons, and the "actor by television series" categories are generally overpopulated. But the consensus in recent CfD debates on the subject is to restrict these to the main cast of a series, a determination which is not difficult to make. The slippery-slope argument Samuel Wantman makes above has some merit, but I see this as an argument for caution, not elimination. The fact is that for many actors, a role in a television series will be a significant part of their biography. For many other categories, we use whether there is explicit mention of the category in the article as a criterion for inclusion. Why can't we just use a slightly more stringent criterion for performers: add a category only if there is significant mention of the role in the article. "Significant mention" would not include mention in a filmography, or even the bald statement "so-and-so appeared in such-and-such in 19xx". It would include any article that has a section devoted to the actor's work on the series, or notes about awards received for it.

Let me be clear: overpopulated and/or excessively applied categories are a problem. But eliminating an entire class of useful categories is not the solution to it. Let's not throw out the baby with the bathwater. —Josiah Rowe (talk • contribs) 02:24, 21 January 2007 (UTC)

Categorizing an article is not an isolated event. If anything, we need broaden all of our discussions about categories to discuss entire classes of categories. When considering a category, we should think about what the category will look like fully populated, which articles belong, and which don't. We also need to consider what the category does to an article. Adding one category often implies several others that belong. So it is more than just adding a category because there is a section devoted to the subject, we need to consider the eventual outcome of the entire categorization scheme. If we make a distinction that X belongs in category Y, we also need to consider all the other categories that may be created if the scheme is complete, all the categories the article will be in if the article is complete, and all the articles that will accompany that article when every stub reaches FA status. It is easier to say "no" to these categories when they are small, and not part of a massive hierarchy. That is why I've been pushing this argument. We need to make very clear distinctions and strictly enforce them. Otherwise, I'm afraid we may all be wasting our time trying to make the categorization system make any sense. -- Samuel Wantman 20:47, 25 January 2007 (UTC)

Persons by creative work

Support or Oppose: Should Performers by performance (proposed above) be more broadly expanded to include:

  • Persons by <creative work>

-

  • "Persons" in this context are those who create the creative works. Examples: Artisans, blacksmiths, violin makers, bakers, artists, sculptors, producers, cinematographers, jewellers, engravers, brewers, etc.
  • "Creative work" examples: a sculpture, a painting, a diagram, a blueprint, a theory of mathematics, a novel, a dictionary, an article, a blog, a web site, a cake, a carpet, a bookshelf, a bridge, a galley, a pocket watch, a bell, a computer, a video game program, a programming language, a slide show presentation, a log cabin, an igloo, etc.
Discussion
  • Agree. The reasoning here is the same as actor-by-performance. Game developers or architechts might work on numerous projects over the years, and if each project had its own category you could end up with tons of categories in each person's article. Dugwiki 16:19, 8 January 2007 (UTC)
  • The reasoning is the same, but this might sweep very broadly -- I can't think of any counterexamples off the top of my head but this is so broad they might exist. --lquilter 15:15, 10 January 2007 (UTC)

General Discussion

No, I'm pretty sure he's just asking for feedback on his suggestion. I posted my responses to his various proposals for the guideline above. Dugwiki 16:21, 8 January 2007 (UTC)
  • I'm going on the assumption that this is a discussion and not a poll. I would support removing all of these categories as overcategorizations. Since most successful performers would fit in dozens of these categories it clutters up their category listings with information which (hopefully) will be just above in a filmography or similar. I think this is reason enough. It becomes impractical to present this information as categories for many actors. The alternative, having a list attached to an article is just as convenient for the user. So if William Shatner is not in the "Star Trek cast" category, you click on the link to "Star Trek" and there is information about the main cast. Or click one more link to get a more detailed cast listing.
    This decision to keep or remove should be an all or none decision. Either all performer categories get dealt with this way or none of them. Otherwise, we'll be discussing this at CFD forever. I don't see what is lost by making lists out of this information, Lists avoid needless category clutter. The lists can also provide information that categories don't, the character, dates, pictures, etc... -- Samuel Wantman 20:34, 6 January 2007 (UTC)
I generally agree, Sam, but I think you picked a bad example with "Star Trek". The main exception would be when you have a category that covers a variety of individual works, such as Category:Star Trek actors which covers regular cast from all the different Star Trek spinoffs. In that case, it's a little more unwieldly and harder to use for a reader to collect information from multiple articles in one list. Using a category to cover a franchise of related shows does, I think, make sense. So a better example using William Shatner would be "Category:T. J. Hooker actors" or "Category:Over the Hedge (film) actors", neither of which I think would be necessary. Dugwiki 16:29, 8 January 2007 (UTC)
I picked Shatner because hes was the most borderline actor I could think of, by which I mean if there is any TV actor closely associated with a show it would be him. And even in his case I don't think he should be be in Category:Star Trek, if for no reason besides the slippery slope of putting more actors in performance categories. We probably need a template for all such categories that says "For actors appearing in this show see List of cast members of XXX". This seems to be one case where we need a hard an fast rule that performers by performance should NEVER happen. If we allow Shatner, next it will be Nimoy, and then Stallone, etc... etc... until every actor that appeared twice in the same character gets categorized. By that time, people will say "if him, why not her?", and it will all fall apart. -- Samuel Wantman 02:56, 9 January 2007 (UTC)
By the way, on a related tangent, note that currently we have Category:William Shatner which received a "no consensus" result on a cfd discussion July 3 2006. The topic of whether or not eponymous categories for individual people is a good idea is another ball of wax entirely. For example, since Category:William Shatner exists, should that category also be applied to every actor, director, film, TV show, book and cartoon he's notably worked with? Now take that process, and do the same for not only every actor who is popular and prolific, and imagine how many categories and cross-categories you'd have. Films and TV series would have up to one category per actor and important crew member, and actors and directors could have up to one category per actor they've ever co-worked with. Imagine how many categories and articles that might be for, say, Steven Spielberg when you add a new category for every major actor and director and producer he's ever done signifacant work with. Dugwiki 16:39, 8 January 2007 (UTC)
Six Degrees of Kevin Bacon -- Donald Albury 23:00, 9 January 2007 (UTC)
And fortunately, no one has created Category:Kevin Bacon, yet. -- Donald Albury 23:11, 9 January 2007 (UTC)
  • My intention was/is that this is a discussion for consensus, the "same way" that CfD determines consensus. So each section is a separate nomination, but the closer should take the general discussion into consideration as well, where appropriate/relevant. (My apologies for forgetting to add the general discussion section when posting.) I hope this helps clarify. Feel free to link to this discussion whereever you feel that others might be interested, but please keep WP:SPAM in mind : ) - jc37 10:19, 7 January 2007 (UTC)- jc37 10:19, 7 January 2007 (UTC)

Kitesurfing locations?

Is anybody else as surprised as I was to see that we have Category:Kitesurfing locations? -- Donald Albury 20:37, 10 January 2007 (UTC)

I'd suggest nominating this one for deletion. First off, whether or not something is a "top kitesurfing location" is probably partly subjective. Second, the main article on which this category is based is currently unreferenced and might therefore itself be subject to afd deletion.
Finally, even if you assume the category criteria is completely objective and verifiable, it's a bad idea to start categorizing locations based on the types of tourist attractions they might provide. For example, imagine how many categories Las Vegas would fall under if you started making categories like "Category:Gambling locations", "Category:Musical theater locations", "Category:Hiking locations", "Category:Heavily visited tourist locations", etc.
So cfd this category for deletion (there's already a list article, so listifying isn't needed). Dugwiki 20:53, 10 January 2007 (UTC)

I'm starting to think that a warning notice should be made that this talk page is not a "pre-CFD" discussion... It's been turning into that quite much these last few weeks.Circeus 22:57, 10 January 2007 (UTC)

I agree. The discussions here should be as a result of previous CfD discussion(s), not for discussions of pro-active strikes, even if in order to add new sections to this page. Else CfD is being undermined as a place for discussion. - jc37 23:03, 10 January 2007 (UTC)
That's preferable to the way I feel sometimes. I see a group of categories that I want to tidy up, with about 20-30 articles in them. I could easily create a new category structure, depopulate the old categories and populate the new ones, and then stick the old (now empty) categories in a Category:Empty categories for deletion category. But instead, I fear that someone will cry foul and insist I take the categories to CfD, when a long, tedious discussion may get the, ahem, 'wrong' result. What should I do when I feel like this? I have expereience in categorising. Should I just trust my instincts? Carcharoth 23:08, 10 January 2007 (UTC)
Wikipedia works by consensus, so if you take something to CFD and don't get the result you want, you probably should learn to work with that. Sometimes I get the results I want, sometimes I don't, but I respect the consensus results enough not to try to do an end run around them by simply ignoring cfd altogether and emptying categories I don't like.
Besides, odds are that if you try and completely empty a category someone will notice and revert your changes. All the articles you are altering are presumably monitored by their main editors, and the categories you wish to label as "empty" are probably monitored by the categories creators. So it's a good bet that trying to do what you described would simply result in those people reverting all your changes and telling you to discuss the proposed changes on an appropriate project page and/or cfd or afd. Dugwiki 23:30, 10 January 2007 (UTC)
Actually, if you create the new categories first, and then change the categories, many editors watching the pages whose categories you change will go - "hey, that's a good idea" (providing it is a good idea of course). I'm really thinking of borderline speedy CfDs here. Regarding the more complicated examples, I find myself arguing (or I did in the past) on CfD for a particular structure and set of category names that I think will work, and people don't seem to get it. But if I could carry out the changes and then point to them and say "what do you think of that?", then things would be clearer. It's like applying WP:BRD to categories. When it works it is good. Also, recategorising categories and creating intermediate new ones, is rather easy. That doesn't require a CfD discussion. Carcharoth 01:08, 11 January 2007 (UTC)
I think I understand what you're asking, Carcharoth... You're asking where the venue is to discuss category structure for multiple categories? This has started to become more prevalent, as our category structure has been filling out. the precedents that I have seen are: WikiProjects; WT:CAT; someusertalkpage; some semi-related Wikipedia: page (whether right or wrong, such as here); On CfD itself as a proposed struture change (which is then followed up by either bold action or by follow-up CfDs); or on CfD individually or part of a group nom (which almost always fail to "No Consensus", since what's being discussed is a structure, which is then followed up by one of the previous discussion types). Hope this helps : ) - jc37 23:37, 10 January 2007 (UTC)

Award winners

First, I did some (relatively) minor wordsmithing on award winners to try to clarify the second sentence, which was talking about what to do with an award if categorization is inappropriate. --lquilter 15:58, 19 January 2007 (UTC)

  • Actually it turns out that what I thought was simply unclarity in the phrasing was not; Radiant pointed out to me that the phrasing was meant to be article about the award. So, I'm going to revert, and suggest here that we need more clarity in that section. Maybe "articles for the award or individual award-winners"? --lquilter 16:01, 19 January 2007 (UTC)

Second, I would suggest that the last sentence is too narrow. It currently reads "If an award doesn't have an article, it certainly doesn't need a category (and not every award that has an article needs a category)." That's true, but doesn't offer much guidance. The vast majority of award-winner categories (Category:Award winners) have articles, so this sentence is just going to weed out non-notable awards or awards that people haven't yet documented in wikipedia. The real problem is that there are some fields with many awards and people/organizations/things might win many of those awards, leading to vast numbers of categories. Can't we come up with some more objective and narrow criteria? Say, for example, that award winner categories are merited only when the award is, itself, a defining -- not merely notable -- characteristic of that person. For instance, Nobel Laureates are defined by the Nobel. I honestly can't think of any other award that is so defining -- people might be nominated for and win multiple Academy Awards in multiple categories; introductions of them wouldn't always describe the person by the various specific academy awards. --lquilter 15:58, 19 January 2007 (UTC)

There's probably two seperate bars of entry to consider: the bar of entry to create an article about an award, and the bar of entry to create a category for the award. In my opinion, it should be harder to create a category for an award than it is to create an article about it. There are a couple of reasons for this:

1) Most award articles can also contain a list of past winners, or have an easily accessible link to a list article for them. So it's likely that a reader who wants to look at articles about people who won a specific award would be able to find the information within the article itself, or possibly one click away in a list article.

2) A list article will almost certainly be created regardless of whether or not you have a category. That's because for awards you need to list not only the person who won but also the year they won it and possibly information about why they won the award. You wouldn't simply want to give the names alphabetically with no other information at all. So almost without exception there will be a list in the article space giving a full breakdown of all the award winners, including year and why they won. That makes having the category for that award redundant, as a reader will already be able to find all the information he needs from that list.

3) Famous or highly notable people can win upwards of 50 or 100 awards and recognitions in their career, some more important than others. If most or all of those had their own winners category, you'd end up with 50 or 100 award-related categories in that person's article. In addition, the award is (hopefully) already listed in the person's article, probably in a "awards" section, with a link to the corresponding award article. So a reader will already have most or all of the information he needs about that person's awards in his article, and will have a link to the award's article to find other people who won the same award. From the perspective of assisting a reader searching for information, then, having these redundant award categories in the person's article offers little benefit, if any.

So, in my opinion, awards should only have their own unique winners categories if the award is extremely important or defining. That way the category could serve as a flag, of sorts, indicating exceptional importance of the individual. Knowing that someone won an Academy Award or a Nobel Prize, for example, immediately tells you that they reached the pinnacle of their industry. Therefore that level of top industry award is probably worth flagging. But less influential awards probably don't need the category. Dugwiki 18:21, 19 January 2007 (UTC)

I generally agree with your argument. My caveat is that, without objective criteria, "extremely important" is open-ended. Another issue is that we can't prescribe here. I'm concerned about being able to demonstrate a consensus for this guideline if it is too detailed and restrictive. I know that what I would like to see in policies and guidelines and what there is general consensus for is not necessarily the same. --Donald Albury 14:03, 20 January 2007 (UTC)
Dugwiki lays out many of my concerns; Donald Albury points out another of my general principles--which is not to have a hidden "super-notability" criteria for categories. ... This keeps leading me to thinking that we should have a general rule of "no categories for award winners" with a short list of pre-approved exceptions, like Nobel (maybe Academy--I've started rethinking my position about whether it is really that defining; but that's another discussion). But that runs smack-dab into another principle: no unnecessary bureaucracies, and no requirements to seek pre-approval for doing something.  ?? How to balance? Or is there a third way? ... This must have come up before, as there are other "guidelines" about what to do or not to do, and there must be exceptions; are there other guidelines/discussions we could look to as examples of rules-with-exceptions? --lquilter 19:48, 21 January 2007 (UTC)
  • While having common-sensical exceptions is acceptable, explicitly enumerating exceptions is not good (as this encourages people to add to the list, ultimately defying the point). >Radiant< 13:57, 22 January 2007 (UTC)
I agree with all the replies above. The only reason I mentioned the possibility of a "super-notability" exception is that I'm pretty sure there are editors who want to make an exception for categories for Academy Award winner and Nobel Prize winners. I don't like the somewhat subjective nature of what it would mean to be "super-notable", though, so I could accept having no award winner categories at all, even for top industry awards. Both possible methods have their own benefit and downside. (If you allow the exception, it acts as a useful flag and is something readers are likely to use, but exceptions can be abused. If you don't allow the exception, you lose some search utility that readers might want. )Dugwiki 18:35, 22 January 2007 (UTC)
I could go with no award winners categories at all, too. I initially thought we probably needed, say, Nobel and Academy (but what Pulitzer? and ...). But then if you just applied those two categories, it would be irresistible to subdivide them by the more specific award won or the year won. If they weren't subdivided, then it's really a big mix of unlike things; if we do subdivide them, then they're small enough sets to be adequately handled by a list. So the advantage, I think, is nill. ... --lquilter 13:39, 5 February 2007 (UTC)

Categorize national parks by state?

I was considering adding Category:National parks in Utah, was sort of surprised it didn't exist already, and wondered if this had been discussed and shot down before. What's the consensus of categorizing national parks (or areas supervised by the national park service) by state? Thanks, — Zaui (talk) 23:25, 22 January 2007 (UTC)

Category:United States National Parks by state already exists. I think it's overkill given the clickable map at Category:National parks of the United States. -- Rick Block (talk) 03:05, 23 January 2007 (UTC)
Definite Overcategorization, and inconsistency. These subcats shouldn't exist, and neither should the category Category:United States National Parks by state. They should all be upmerged into Category:National_parks_of_the_United_States, as there simply aren't very many, and only a few have been segregated out this way. There are a total of 61 National Parks, in 29 States, so the average category of this type can only hold 2 entries! In fact, the two largest categories would be California and Alaska, each with eight entries, and the majority of other categories would exist for only a single entry.
They should also be crossreferenced with the already existing "Parks in (State)" Categories, of which there are currently 29. See Category:Parks_in_the_United_States. This way, if anyone is interested in which notable parks exist in a particular state, they can be redirected there and find everything they need. There's no other apparent utility for National Parks by State categories. National Parks don't really belong to the States in which they're located, but still, if anyone is interested in such a geographical breakdown, besides the above mentioned clickable map, they can look at the already existing List_of_United_States_National_Parks_by_state.zadignose 04:30, 23 January 2007 (UTC)

Is Category:Disasters by year overcategorized?

Is Category:Disasters by year overcategorized? I started (or at least greatly expanded) the idea of categorising the articles in Category:Disasters by year. My intention was to see all the disasters for a particular year lined up in its own category. For example, Category:1990 disasters. I didn't intend to subcategorise any further, as Category:Disasters is already split up fairly logically (eg. Category:Transportation disasters; Category:Natural disasters). But recently I noticed someone created the categories Category:Transportation disasters by year and Category:Natural disasters by year. Now it is no longer possible to see a category with all the disasters from a single year. Instead you have to click up and down through many levels to see what is even there. Can we please stop this relentless overcategorization? I used to be able to navigate around the Disasters category system and find logical categories with 5-20 articles that would make interesting reading. Now I find these articles have been dispersed over innumerable tiny subcategories and the result is painful to try and navigate through. What can we do to avoid the end result of having each article in its own category, and how can we get the right balance between this and having all the articles in one category? Carcharoth 00:42, 25 January 2007 (UTC)

Intent of this guideline

My understanding is that this page is a reflection of consensus already determined on CfD. As far as I know, we are not adding any examples to the page, except as a result of existing consensus. (For one thing, I don't think we should pre-empt CfD that way).

I'd like some comments from the other page regulars (and any other interested parties) about this. - jc37 13:22, 26 January 2007 (UTC)

Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Overcategorization/Archive_2&oldid=1137431337"