Wikipedia talk:Categorization/Archive 10

From Wikipedia, the free encyclopedia

This page is an archive. Please do not edit the contents of this page. Direct any additional comments to the current talk page.

Contents 1 Wikipedia:Category types 2 Counting articles in categories 3 editing of categories 4 DEFAULTSORT 5 Categorising sections using anchored redirects 6 Does subcategorization automatically imply a proper subset? 7 Category vs. List question 8 Excessive category clean-up template 9 Interior design category for a general university??? 10 Definite and indefinite articles 11 Category:Kurdistan 12 Commons 13 "tags" & tagging versus categorization 14 Guideline to use information in article for categorization? 15 Sorting of non-standard letters 16 Moving categories?

Wikipedia:Category types

Since getting rid of {{subst:tl|CatDiffuse)) seems to be a dead end, I'm taking a different tact. I've been working on a new set of labels for categories that I mentioned several weeks ago on this page (see above). These will hopefully help people understand how the categorization system works. The examples are based on the discussion happening now at Wikipedia:WikiProject Films/Categorization but they would be applicable to most of the categories the way they currently exist. I'd appreciate some feedback. Thanks. -- Samuel Wantman 01:08, 5 January 2007 (UTC)

Counting articles in categories

Is there any way of counting the number of articles in a category without going through the whole category and multiplying the number of pages by (usually) 200? There is a bot which does this for certain categories, but not for others. To be more specific, I am a member of WP:ALBUM and at certain times we would wish to know the exact amount of articles in Category:Needs album infobox or Category:Albums without cover art, for instance. Anybody come across a handy auto-process? Bubba hotep 09:43, 12 January 2007 (UTC)
By the way, I've just been informed that AWB does this in a matter of seconds. I hadn't thought of that and have only just got AWB. But is there any other way(s)? Bubba hotep 09:49, 12 January 2007 (UTC)

This is also very straightforward to do in the pywikipedia 'bot framework. Could you say a little more about how you'd like this to happen? Roughly how many categories would be affected, and roughly how often, for example? Any particular form you'd want the results in? Alai 07:36, 13 January 2007 (UTC)

It would literally be just the two categories mentioned above, plus Category:Uncategorised albums, Category:Album stubs, and Category:Non-standard album infoboxes. It would ideally be done on an on-demand basis and would simply display the number of articles in each category, basically to tie in with the project's "To-Do" list. So simple, I could do it with AWB and a table I fill in manually every day, actually. Bubba hotep 20:45, 13 January 2007 (UTC)

editing of categories

New users seem to have the idea that you add pages to categories by editing the category page itself. Could we add something to the boilerplate for editing Category pages, so it tells them not to edit in an article, but to add the category wiki-text to the given article? This would save me and many others a good deal of time in reverting mistaken edits to categories, and would therefore allow us to focus further on articles. Please share your opinion on this idea, --Urthogie 23:38, 14 January 2007 (UTC)

There doesn't seem to be anything handy already built in (like the "This is a talk page..." message when editing a talk page), but I wouldn't think it would be hard to add something that could do this. It sounds like a reasonable idea to me. -- Rick Block (talk) 03:19, 15 January 2007 (UTC)

Where would I go to get this implemented?--Urthogie 04:14, 15 January 2007 (UTC)

Enhancements are treated the same as bug reports, please see Wikipedia:Bug report. -- Rick Block (talk) 14:15, 15 January 2007 (UTC)

DEFAULTSORT

Wikipedia:Wikipedia Signpost/2007-01-02/Technology report announced a new "magic word" added by using {{DEFAULTSORT:''Sort Key''}}. Something about this should probably be added to project page.

Note that this "magic word" itself is case-sensitive: using "defaultsort" will not work. Gene Nygaard 16:52, 15 January 2007 (UTC) See also Help:Magic words 16:14, 23 January 2007 (UTC)

I've started using this as well (see J. R. R. Tolkien). It has its pros and cons. You can still overide the default for any specific category tag by using the pipe sorting as normal. Carcharoth 14:27, 17 January 2007 (UTC)

Awesome! This will be particularly useful for any biographical article tagged as a stub. I've always found it annoying that the stub categories were forced to sort by first name rather than last. — CharlotteWebb 01:19, 25 January 2007 (UTC)

I tried adding it to {{Lifetime}} but something went screwy with the fancy conditional code and I couldn't find an answer as to how to fix it. If anyone wants to take a look at this it's a great opportunity to sort vast swaths of biographical articles. Bryan Derksen 04:22, 26 January 2007 (UTC)

In the short term, this is actually more likely to cause problems with the stub categories, as it'll take a long time until any significant number of the bio-stubs (vast swaths indeed) have default sort keys, and stubs being stubs, one wonders if anything approaching consistency will ever actually happen by this route. And if consistently-by-first-name was annoying, people are likely to find a mashup of by-first and by-last much moreso. That's why the stub-sorting project has been non-keen on previous suggestions to do this (typically using an optional argument to stub tags). I was trying to think of a way to recode stub templates to do something more sensible by default, but firstly, I don't think it's possible without the StringFunctions being enabled here, and secondly, doing so would actually override the defaultsort, rather defeating the purpose of the exercise. Currently I can only come up with either a) recoding the stub templates to ignore the default sort, for the sake of consistency, or b) doing a vast amount of 'bot-tagging of two-word bio-stubs with "second word, first word" default sort keys. If I'm missing a better fix... Alai 06:59, 26 January 2007 (UTC)

If it ain't broke, don't fix it. That's the better fix. Don't worry about sorting of stub categories one way or the other. Nobody much cares if all those zillions of overcategorized stub categories are missorted. It isn't all that helpful to the purpose of stub categories (if they are jumbled up a bit, maybe that will help someone find an interesting one to fix). If you cut the number of stub categories to a tenth of what they are, all those stub sorters might instead have time to go expand some of those stubs. Gene Nygaard 23:40, 26 January 2007 (UTC)

OTOH, properly alphabetizing the subcategories of Category:Stub categories should have been done ages ago. People looking for a specific category to work on should be able to find it. Something like the DEFAULTSORT magic word wasn't necessary to fix that; it could easily have been accomplished with ordinary sort keys. Yet that hadn't been done, before I did it yesterday or the day before, whenever it was. Gene Nygaard 23:44, 26 January 2007 (UTC)

On the first, I largely agree. (If we filter out the side-swipes about "overcategorisation" and general "all those foolish stub-sorters wasting their time" rhetoric: are you even remotely serious, and have you seen the size of some of those puppies? And if you reduce the number of stub categories by a factor of ten, you increase the size of the categories likewise.) The ideal model of a stub type is "set of articles a given group of editors are likely to be interested in expanding any given one of", not an indexing service for people looking for a particular article. If anything, my main concern is that people don't spend an excessive amount of effort in "fixing" sort keys in stub categories by some manual method, which is essentially what people have proposed/unilaterally started in the past.

On the second, I have no idea what you mean. There aren't any "eponymous" stub categories, at least that spring to mind, so the question of last-name indexing doesn't arise. What else is not "properly" alphabeticised? Are does it come under the category of "sofixit"? Alai 00:08, 27 January 2007 (UTC)

LOL, I happened to find {{Zappa-stub}}. Might want to nip that one in the bud. — CharlotteWebb 21:46, 2 February 2007 (UTC)

The problem wasn't last-name indexing; that's far from the only sorting issue. The problem was indexing under characters other than the digits 0 to 9 and the 26 letters A-Z of the English alphabet. Gene Nygaard 00:15, 27 January 2007 (UTC)

For example, Category:Île-de-France geography stubs was somewhere off in oblivion after Z. Gene Nygaard 00:19, 27 January 2007 (UTC)

You mean top-sorting? That's a feature, not an oversight. It makes it a good deal easier to find sub-categories in large stub types (which would be a good deal larger if you had your way), by consolidating them all on the first page; and it doesn't effect alphabeticisation. If you're doing these "properly" by changing sortkeys from " blah" to "blah", then I take it back, please don't "sofixit" at all. Alai 00:24, 27 January 2007 (UTC)

I don't know what you mean by "top-sorting" in this context. What I'm talking about is where that category is found in its parent category. It is now found under I where it belongs. It used to be off in oblivion under Î which was after Z, after d (all the lowercase letters follow their uppercase version), and various other characters. Gene Nygaard 00:38, 27 January 2007 (UTC)

Another example is Category:Côte d'Ivoire stubs which now properly appears in Category:Stub categories before "Category:Country album stubs", not after "Category:Czech writer stubs". Gene Nygaard 00:42, 27 January 2007 (UTC)

Ah, those. If you consider "sorted under the accented character that the article title starts with" to be "oblivion", and under the unaccented equivalent (or otherwise, depending on language) to be "correct", then I can see why you might see it that way, though the rhetoric seems pretty over-the-top from where I'm sitting, and the whole topic pretty unrelated to the original (other via "yet more ways in which stub sorters suck"). IIRC the latter is what French uses, so I don't know why the default behaviour is as it is (perhaps other languages use the same character, and sort it differently). Alai 00:52, 27 January 2007 (UTC)

To second Alai, I have occasionally forced accented words to sort in "un-accented" order and been reverted by editors who insist that this is wrong. <shrug> I think it's the least of our worries. As for the stub cats being temporarily jumbled, we stubbers are used to making order out of chaos. <g> Her Pegship (tis herself) 01:25, 27 January 2007 (UTC)

This is why I'd like to get DEFAULTSORT put into {{Lifetime}}, since it's transcluded into so many bio articles it'll "flip the switch" on tons of them all at once. Once upon a time I argued against making {{lived}} a subst-by-default template for similar reasons, I just knew someday we'd have something more we could do with that information. Ah well, Cassandra's curse. Most of the remaining bio articles should be possible to handle with a relatively simple bot, at least. Bryan Derksen 08:44, 27 January 2007 (UTC)

But nothing like all, so even if this were an option, it'd just make the situation more manifestly inconsistent more quickly. I agree that it seems pretty straightforward. However, it'll be a lot of edits, as there are hundreds of thousands of bios, and it's likely not to be 100% accurate, either, since people do insist on having strange names that don't conform to the standard pattern. I'll ask over at WPBIO to see if they have some pertinent input. Alai 00:07, 28 January 2007 (UTC)

My notion for what the bot would be doing would be for it to look at the birth category and death category and if they've both got the same sortkey associated with them assume that this is the "proper" sorting for that person's name and add a DEFAULTSORT to match. As long as the birth and death categories were added with proper sort keys to begin with I imagine that should be a fairly safe way to approach things. But since I'm not a botsmith myself I guess it wouldn't ultimately be up to me, so take it for what it's worth. :) I don't think we should be too concerned about inconsistencies during transitions since Wikipedia as a whole is a work in progress, as long as we come out the other side with an overall improvement in quality a short-term (and in this case relatively minor) disruption is IMO worthwhile. Bryan Derksen 10:53, 28 January 2007 (UTC)

That would be a bad assumption. There are still a large number of birth/death/living categories added by foolish editors who didn't add sort keys, even though other categories in the article already had proper sort keys. Gene Nygaard 21:52, 28 January 2007 (UTC)

Possibly lazy, rather than foolish. The assumption sometimes is that someone else will eventually come along and fix it... Carcharoth 00:29, 29 January 2007 (UTC)

In cases where there's no sort key to begin with no DEFAULTSORT would be added, so no harm is done. I don't see how this makes the assumption bad. The bot could perhaps make a list of such articles for further attention by human editors if their existence is a problem. Bryan Derksen 00:48, 29 January 2007 (UTC)

No, no good will be done. Part of the harm would be the false confidence due to a belief that a bot has gone through and fixed everything so that it is all now hunky-dory.

Furthermore, if both the birth and the death categories are already correct, odds are the rest of them are correct as well. The biggest problem areas are those which have no sort keys on any of them, or maybe one category. In that case, such a bot would just be an exercise in futility, frittering away its time for no benefit. When birth/death categories do currently differ from other categories, sometimes they are correct, sometimes the others are correct, with no clear probability either way--and sometimes "all" of them are correct; all categories are not necessarily sorted the same way.

Third, there are a great many of them in which all existing sort keys are incorrect. There is more to it than how many of the names are used and in what order, too. Often this is because they include letters with diacritics. Sometimes it is because titles have not been stripped out. Then there is the pervasive problem of putting "Jr." and the like between the last name and the given name. Or not including both a comma and a space to separate the last name from the first name (this is one particular area in which you are more likely to see discrepancies between the sort keys in various categories).

Fourth, unless you are going to have the bot actually remove existing sort keys, any that now have the incorrect sort keys will not be changed, so the bot will again be ineffective. Even if a default is added with DEFAULTSORT, sort keys override it.

Fifth, OTOH, if you do remove existing sort keys, then, for one example, those categories that should be first-name sorted instead of last-name first will be made worse, will now be missorted when previously sorted properly. In that case, the bot would do clear damage. And it would be damage that might go unnoticed for some time.

Related to the last, the first-name sorting is a property of a category (family name categories, for instance), not of a person. And just defaulting to the article name for certain categories would not solve the problem, either—even first name sorted categories need to be stripped of diacritics and indexed on the basis of the 26 letters of the English alphabet. English peerage categories are another problem area. Gene Nygaard 02:14, 30 January 2007 (UTC)

Categorising sections using anchored redirects

Please contribute to the Village pump discussion started at Wikipedia:Village pump (technical)#Categorising sections of an article. Thanks. Carcharoth 14:25, 17 January 2007 (UTC)

Does subcategorization automatically imply a proper subset?

user:Ashley Y and I are having a disagreement at Category talk:Anti-Zionism. According to the article, Anti-Zionism#Anti-Zionism and antisemitism, there is reliable and verifiable discussion whether or not Anti-Zionism is a form of Anti-Semitism. According to my understanding of WP:CAT, the operative clause is “If you go to the article from the category, will it be obvious why the article was put in the category? Is the category subject prominently discussed in the article?” which is eminently fulfilled in this case. Further, WP:NPOV states “The policy requires that, where there are or have been conflicting views, these should be presented fairly. None of the views should be given undue weight or asserted as being the truth, and all significant published points of view are to be presented, not just the most popular one.”

user:Ashley Y beleives that adding category A to category B implies that All A are B, in which case placing AZ as a subset of AS implies that all AZ is AS, which I agree is not a given conclusion. I believe that adding category A to category B implies that there is an overlap between A and B and that there is a relationship between A and B, which I do not believe anyone can argue when it comes to AZ and AS. Further, being that the point here is that AZ MAY ACTUALLY BE a proper subset of AS (per the discussion in the article) it is a violation of WP:NPOV to artificially hide this by removing the category. user:Ashley Y understandably believes that its presence is a violation of NPOV.

Thus, I believe that there are two issues that need to have some consensus here:

1. Does subcategorization ALWAYS imply a proper subset, total containment of the child category in the parent category?
2. WHich is more NPOV, to deny the debate or inform people of it?

Thank you. -- Avi 19:32, 17 January 2007 (UTC)

It does not always imply exactly "proper subset", but there must be some kind of "containment" or "parent-child" relationship, not just some (typically symmetrical) linkage as implied by "overlap" and "relationship". —Ashley Y 20:18, 17 January 2007 (UTC)

That's my understanding as well. In light of the many different ways the wikipedia categories / subcats are used, it couldn't relaly be a proper subset -- they're used for taxonomies -- but they shouldn't simply be "related" categories. If they are merely "related" then a "see also Category:X" at the top of the category page is more appropriate. --lquilter 20:24, 17 January 2007 (UTC)

I would also say that no, it does not imply a proper subset. There are numerous examples to the contrary. Gene Nygaard 17:20, 21 January 2007 (UTC)

I'm not sure I understand the distinction between taxonomies and superset/subset relationships. Surely the former is an example of the latter? The felinae being a subset of the carnivora, say, and thus a constituent taxon (or taxon of a taxon, or however many levels there are currently), and this a subcategory (of a subcategory...). Alai 07:49, 24 January 2007 (UTC)

I wish we could reach a consensus about this sort of thing. I think there are 3 types of category relationships:

• X is a Y, which is used for categories that group similar people or things, like Category:American novelists and Category:Suspension bridges
• X is about Y, which is for categories that cover a topic such as Category:Literature or Category:Racism
• X is related to Y, which is mostly for eponymous categories that collect articles that relate to a single subject or person such as Category:George W. Bush.

For more about this see my proposed Wikipedia:Category types. -- Samuel Wantman 08:04, 26 January 2007 (UTC)

Suggest reading related discussion on Wikipedia talk:Overcategorization#The tree organisation and a key word: relevance. Taxonomies & subjects are two different scopes of classification, and suggest different trees. Parent/child (or ancestor/descendant) just describes the container-ness of categories. Wikipedia categories cover both taxonomies & subjects. Because it covers both, and because we have a flexible way of developing systems for subcategorizing, wikipedia categories are not always "proper subsets". Forget the jargon. Think about it this way. A proper subset would be, for instance, parts of the human body > bones > femur. We have those in wikipedia. However, we also have things like Awards > Lists of awards. You can see that "list of ..." is not an award, per se; however, it is included under "awards" because it is part of the subject. Does that make sense? --lquilter 14:40, 26 January 2007 (UTC)

But technically you can still, with the appropriate wording, integrate stuff llike this. You need to name the categories properly. Thus 'Award (topic)' would include general articles about awards, as well as include a subcategory 'Awards (index)' which would be a grouping of various indexes of actual awards, such as lists of awards, and articles about actual awards. In reality though, this can quickly become overcomplicated. Carcharoth 14:54, 26 January 2007 (UTC)

I agree, there's a stronger "subset" relation here than there is than with many of the weaker "related to" categories, which are often where the sillier forms of circularity or nonsensical Chinese whispers effects start setting in. I also agree that in principle, truly systematic category naming is the better approach in the longer term, and is preferable to the "honking great big template" solution, or overly fined-grained meta-classification. Alai 21:53, 26 January 2007 (UTC)

Category vs. List question

I'm considering a category or list that is going to be fairly dynamic (American soccer players by country currently playing in). First, a suggestion of a better name is always encouraged. Now this list/category would change fairly often, every time a player transferred. It seems that categories are more static and as such I should use a list article for this. Lists are certainly easier to keep up with but have the disadvantage that categories are easier for people to find, and thus have an increased chance of people maintaining. Any suggestions or knowledge of Wikipedia precedents on such matters would be quite helpful. Captkrob 23:01, 22 January 2007 (UTC)

I suspect that either one would be nominated for deletion as not being encyclopedic. The data is dynamic, as you say, and the intersections are not ones what are generally considered encyclopedic. But that's just my opinion. Vegaswikian 08:18, 24 January 2007 (UTC)

I partially suspected that. But the detail to which sports are kept track of on wikipedia (up-to-the-hour stats on thousands of players, teams, etc) seem to make the "dynamicness" of this seem almost encyclopedic by comparison. For example, the monthly UEFA coefficient (more dynamic than player's teams) is constantly updated for each national side, club side, and league. Captkrob 14:58, 24 January 2007 (UTC)

Excessive category clean-up template

I would like to propose an excessive category template that can be applied to articles that appear at the top of Special:Mostcategories. This could be used to indicate that some type of category clean-up is needed (whether it involves removing redundant or unnecessary categories from the individual article pages or consolidating or deleting impractical categories). This would be particulalry useful when the categories for an article grow beyond 30 or so; the categories begin to look like an unreadable blue mass after about 30 categories or so. I would be willing to write a draft of such a template. Please comment on the proposal here. Dr. Submillimeter 09:39, 28 January 2007 (UTC)

WP:SOFIXIT; just weed out the unnecessary ones. What in the world would you need a template for? There are only a hundred articles with more than 30 categories; only 35 with more than 35 categories; only 9 with more than 40; and none with more than 45. And most of them are probably legitimate and useful.

Maybe you could at least show us that it would be reasonable to reduce those 9 with more than 40 down to less than 30, or even a couple of them. Discuss it on those talk pages, etc., and see what you can come up with. Gene Nygaard 22:00, 28 January 2007 (UTC)

On second thought, I think there ought to be a place where this problem (or whether it is a problem) ought to be discussed, and some place for you to get some guidance to fall back on as you try to weed out some of the excess if there is some, and that place might as well be here. So I certainly don't wan't to discourage others from expressing their views about it here. Gene Nygaard 00:13, 29 January 2007 (UTC)

And remember that templates have talk pages, too, so that you can start a discussion with your reasons for creating it there if you create a template, and cross-reference discussions here or elsewhere, so that editors know they have a place for feedback on how well the template works. I wouldn't wan't to see some articles with a permanent ugly cleanup box, for example, so you need to consider when it would be okay to remove the template. Gene Nygaard 00:17, 29 January 2007 (UTC)

Actually, see Talk:Alanis Morissette. The Alanis Morissette article is a good example of where the category system appears to be working poorly or applied improperly (as Morissette is classified as both a Canadian whatever and an American whatever, thus doubling the number of categories in her article). From the discussion on the Morissette page and the Talk:Floods in the United States pages, it is unclear that people working on those articles have realized that the category clutter is problematic until I mentioned it. This is why I suggest a clean-up template. Dr. Submillimeter 09:46, 29 January 2007 (UTC)

Found the old version with all those categories at the bottom... Carcharoth 03:05, 30 January 2007 (UTC)

Wow. They actually put 10s of 'Meteorology by year...' categories on a list of floods article? That is... insane! :-) Putting them on the individual year articles is fine. If the flood will never have its own section in that articles (possible in some cases), then the category can go on the redirect. This puts a link in the category that takes people from the category to the article. Sending people the other way, from the article to the category, is more problematic. And talking of articles with excessive categories, I wibble when I see the category clutter at the bottom of J. R. R. Tolkien. I should know what to do, having some experience with categories, but I still wibble. Is that a word? wibble. Hmm, interesting. Carcharoth 02:34, 30 January 2007 (UTC)

Yea, he does have a few. You may want to look at Category:Natives of Free State Province and Category:Natives of Worcestershire. That sounds like an error. Vegaswikian 02:46, 30 January 2007 (UTC)

They are both true, but are probably examples of overcategorisation. He was born in Bloemfontein, South Africa, left when he was three, grew up in Birmingham, and spent most of his adult life and career in Oxford, working at the university. That explains most of the categories, but not why people feel the need to document someone's entire life using categories... :-/ Carcharoth 03:01, 30 January 2007 (UTC)

You can only be a native of one place! I'm not aware of any people that were born in two countries on two continents. Vegaswikian 03:17, 30 January 2007 (UTC)

Yah. 32 categories. I should really discuss them over at Talk:J. R. R. Tolkien, but I'll list them here to see which ones people think are excessive in terms of categorisation: Tolkien articles with unsourced statements | Spoken articles | 1892 births | 1973 deaths | Tolkien | J. R. R. Tolkien | Tolkien family | Tolkien artists | English fantasy writers | English novelists (seems to duplicate the 'English fantasy writers one') | English linguists (seems to duplicate the 'English philologists' one) | English academics (seems to duplciate the university of Oxford categories given later) | Inventors of writing systems | English philologists | Mythopoeic writers | British Army officers (he was a British army officer, but that is not what he is famous for) | British people of South African origin (well, yes, but this sort of thing is best tacked with persondata and 'what links here') | Natives of Free State Province (not relevant to Tolkien, and if the SA origin one is kept, this duplicates it and is needlessly specific) | British World War I veterans (again, not what he was famous for, though this has more merit than the 'British Army officers' one, as Tolkien is now recognised as being an author whose fiction was heavily influenced by his war experiences - possibly a 'war author' category, similar to the 'mythopoeic writer' category would be better) | Fellows of Merton College, Oxford | Fellows of Pembroke College, Oxford | Alumni of Exeter College, Oxford (three stages in his Oxford career - do they all need to be marked by a category? His time at Leeds University is not tagged with a category - maybe lose all these and keep only the 'associated with Oxford University' one, vague as that is) | Roman Catholic writers | People from Birmingham, England | Natives of Worcestershire (if origin categories are needed, then only have one - no need to have two duplicating each other) | People associated with the University of Oxford | People from Bloemfontein (see comments on the 'natives of Free State Province' category) | Commanders of the Order of the British Empire | Inklings | People commemorated by blue plaques | Marquette University (his works are archived there - probably doesn't warrant a category) | English Roman Catholics (overlaps with the earlier 'Roman Catholic writers' category and the other 'English...' categories - so not needed).

I've bolded the dodgy ones and added comments in brackets in italics. What do people here think of this as an example? Possibly Tolkien needs a lot of categories, but possibly not. Which should defnitely be removed? Carcharoth 03:01, 30 January 2007 (UTC)

Wow. I found Category:Natives of Worcestershire (before 1974) - that would fit Tolkien. That is partly why the Birmingham one can't be subsumed in the Worcestershire one, but still, the argument exists: for people that have been in many different locations, do all the locations get a category? Carcharoth 03:10, 30 January 2007 (UTC)

It sounds like I should proceed with drafting a template. Also, see boar as another example of a category system that has gone awry. Dr. Submillimeter 10:53, 31 January 2007 (UTC)

Here is a first draft of a proposed template. The here link would lead to the template page, where a series of suggestions would be presented on reducing the number of categories in the article. Dr. Submillimeter 23:25, 31 January 2007 (UTC)

Interior design category for a general university???

I don't know my question can be answered here, but I'll ask anyway, since this board discussed topics related to Wikipedia categories and related topics.

Is there a reason why University of Applied Sciences Stuttgart is classified in the Interior Design category, other than for advertising? The name of the school is not listed alphanumerically in this category; instead, the name appears about the category alphanumeric category listing.

From what I see in the main article for this school, the school is not solely an "interior design" school; the schools offers programs beyond interior design. So, can anyone enlighten me on this? I assumed that categories were designed for similar topics (such as interior decorating, furniture, and so on), not for routing readers to a specific school that uses the category as an means advertising to direct traffic to its article. lwalt 00:57, 31 January 2007 (UTC)

Definite and indefinite articles

I would like to propose an addition to the guidelines for categorisation. The addition being that when using DEFAULTSORT or indeed the old way of sorting, definite (the, la, le, der, el, etc) and indefinite (a, an, un, une, etc) articles be omitted. The reason for this can be seen on this page [1]. Transporter 2 appears before The Transporter. If, The Transporter were sorted as simply "Transporter", this wouldn't happen. I'm sure there are other examples of this elsewhere in Wikipedia. Basically, The and A aren't necessary for sorting and in fact cause problems by being included.

Any opinions? Mallanox 23:52, 31 January 2007 (UTC)

As soon as I saw the {{DEFAULTSORT:Noun, Article}} arrangement I knew that a comma after "noun" would create sorting problems, since "(Space)" comes before ",". I have often removed ", The" from category sorting, but since the defaultsort comes as above, I figured there must be some reason I ignore, so I started using it too. But I still think it makes no sense. I agree to change the guidelines to "ignore", which basically means: no comma after "noun". I am for "inteligent sorting" (in this case it would be "Transporter 1"), but this implies that one takes a look first at the category he is assigning, checks how preceeding or following items are sorted and acts accordingly. Of course this is not always practical, as when one has a lot of categorization work to do. Hoverfish Talk 08:17, 1 February 2007 (UTC)

The English articles (a, an, the) should be omitted. Any foreign articles should normally be included in the sorting. An English speaker looking for information should not be expected to know the meaning of various foreign spelling elements in order to be able to find the article.

Whether or not you include ", The" and the like at the end of the sort key when an inital English article is omitted will rarely result in any change in actual sorting, and on those rare occasions where it would matter, it is not at all clear that including ", The" would give the most desirable results. The practice of including that comes from print media, so that the fact that the article is there is shown; in Wikipedia, the article still shows up at the beginning of the articles name in the listing even when it isn't used for sorting, so we don't have that same consideration here. Gene Nygaard 20:48, 1 February 2007 (UTC)

OK, not as much opinion as I would have liked so I'll be bold and hope that might get me a (hopefully positive) reaction. Mallanox 23:20, 12 February 2007 (UTC)

Category:Kurdistan

Hi, I'd like to invite everybody here to get involved with the discussion. There currently is a confusion on what the inclusion criteria supposed to be.
At the moment unrelated articles such as airline companies, rivers, mountains, towns, cities, city squares, long dissolved countries, among other things are categorized under the same category.
--Cat out 05:22, 2 February 2007 (UTC)

Commons

I removed the following:

This article provides guidelines on creating and organizing categories. Where possible, Wikipedia categories should be kept as close as possible to Wikimedia Commons categorization schema in the main category tree and its major sub-categorizations. See the mirror page Wikipedia:Commons categories for a reference on that, and spending some time seeing if the category need has been handled on the commons is highly encouraged before creating a new category name. This will at a mininum prevent redundant category creations, or the necessity of renaming categories on either sisters WP:CFD forum to maintain compatibility.

While the two sisters do not share the same category needs in full, the task of organizing millions of media pages into a system of findable groupings probably has a solution already in place. All other language projects link via interwiki tagging to one or both category systems, so commonality is highly desireable—it saves work for a lot of people speaking many native tounges.

There is an ongoing effort to interlink matching categories, and this is simplified when these two large repositories of categories match as close as possible. The two are very similar since both (and indeed most sister projects categorization systems) were devised on the foundations Meta-Wiki recommendations as part of it's work co-ordinating across foundation projects and language projects. In some areas, that top-down direction is excellent, in others, the effort did not anticipate interlingual and ambiguity of names issues. The communal experience is clear concise unambiguous names, even when longer, are the best choices. Many redundant new categories can be prevented by searching for keywords that might be in an alternate phrase before creating a category.

I don't think the beginning is the place for this. While it might be a goal to have some consistancy between wikiprojects, we are quite far from that now, and I'm not sure that Commons has any insight to categorization that is superior to what we are doing here. Some languages are categorized very differently from English (German is a notable example). So before we all start copying commons (or any other project) there should be some more discussion. Pending some more discussion, I think it best not included. -- Samuel Wantman 07:59, 2 February 2007 (UTC)

I don't see where Commons (a repository of free content images, sound and other multimedia files) categories would have much relevance at all to English Wikipedia categories; they serve different purposes, contain different kinds of stuff. Furthermore, if anything, Commons should model its categories after Wikipedia's categories, not vice-versa. It should be driven by our better discussed, better developed, and more comprehensive categorization schemes. For some strange reason, the talk page for the copy of the Commons categorization page, Wikipedia talk:Commons categories, redirects to this talk page. So consider this my comment with respect to both of them. Gene Nygaard 22:42, 2 February 2007 (UTC)

"tags" & tagging versus categorization

A lot of people seem confused about tags versus categorization. I've noticed this in discussions, based on people's arguments; at least one argument where someone explicitly stated that categories were tags; and just the way people use categories (see, e.g., Category:Shelley. I propose that we need an FAQ that explains this. User:SamuelWantman has laid out a bigger proposal for Wikipedia:Category types which would address this, but getting consensus on that proposal seems hard. In the meantime, a simple explanation with examples of tags versus classification (WP categorization) might solve a lot of the problems. I could start drafting but thought I'd wait & see if someone had done this already. <g> --lquilter 18:06, 3 February 2007 (UTC)

Guideline to use information in article for categorization?

I thought there was something in one of the categorization guidelines regarding the issue below, but since I can't seem to find it I'll ask here to see if someone knows where to look or to see if it's something that possibly should be added to this guideline.

All editors are technically supposed to only be using verified information for article changes. That includes categorization, meaning that as a rule articles should only be categorized using verified information contained within the article itself. As an example, if an article never verifiably mentions that someone is a practicing lawyer, they shouldn't be categorized under Category:Lawyers . So even though it's possible Shaquille O'Neal might have passed his bar exam, since his article makes no mention of him practicing law he shouldn't be categorized as a lawyer. Another example from a recent cfd is that even though Margaret Thatcher obviously publically commented on just about every air disaster and terrorist attack during her political life, she shouldn't be included in Category:Airliner bombings or Category:Pan Am Flight 103 since her article doesn't discuss those issues.

So, although this appears to be common sense and based in large part on the principles of WP:V, is it worth including a short section in this guideline that says "Articles should only be categorized using verifiable, referenced information from the article itself. Do not categorize articles based on information outside the article, such as unverified information, speculation and trivial facts not notably included within the article"? If this already appears in a guideline somewhere, could someone include a link? Or if you don't think this would be a good addition or have a question or problem, I'm interested to hear any feedback. Dugwiki 18:19, 6 February 2007 (UTC)

I think that would be an important & useful feature to add. I'm not sure about the wording. Most of what you describe is about WP:V but "trivial facts not notably included" strays from Verifiability to Notability. (We should also be talking about Notability, of course, but just in trying to keep your current proposal narrowly tailored to the problem.) --lquilter 22:00, 12 February 2007 (UTC)

Sorting of non-standard letters

The rule is to convert letters with diacritics, or non-standard letters to the nearest standard letters. But it doesn't give examples for the really non-standard letters. Would you recommend I pipe "Þagað í Hel" to sort as "Thagað í Hel"? -Freekee 04:55, 10 February 2007 (UTC)

Personally I think "Thagad i Hel" is the closest you're going to get to latin alphabet notation. Mallanox 23:18, 12 February 2007 (UTC)

Yes, and that illustrates an apparent misunderstanding on Freekee's part. All the letters are sorted (and not just letters, but spaces, punctuation marks, whatever). It isn't just the thorn that needs to be fixed; the "ð" and the "í" need to be fixed as well for it to be properly indexed.

I thought I had added an example which involved more than the first letter, to make that point clearer. But I don't find it, so I may have to do that now. Gene Nygaard 09:01, 13 February 2007 (UTC)

It's not a misunderstanding, it's a willful ignorance. I don't believe that alphabetizing past the first four letters is important enough to worry about. But I haven't categorized it yet, and who knows what mood I'd be in on that day. Or how much work I'd be willing to go through to do it completely properly? For example, now I have to ask whether "Þagað" would better be spelled as "Thagath" or Thagathe" instead of "Thagad"? Or should we go with the d because it looks like an eth? -Freekee 06:12, 14 February 2007 (UTC)

I deal with documents in Icelandic language on a daily basis, including those written in Icelandic on an English keyboard. Never have I seen eth transliterated as anything other than d. Mallanox 02:04, 25 February 2007 (UTC)

Moving categories?

What is the procedure for moving a category or sub-category? Does it need to go through Cfd?--Vbd 08:14, 14 February 2007 (UTC)

Affirmative!  Regards, David Kernow (talk) 08:21, 14 February 2007 (UTC)

If you mean recategorizing a sub-category from one category to another, that would not need to be discussed at CfD. If there is disagreement, it can be discussed on the talk page of the sub-category. -- Samuel Wantman 11:39, 14 February 2007 (UTC)

Thanks, Samuel. Recategorizing a sub-cat from one cat to another is exactly what I have in mind.--Vbd 18:07, 14 February 2007 (UTC)

Yes; sorry to overlook this possibility. Since renaming involves "moving" a category (as with renaming articles), I thought this is what you meant. Regards, David (talk) 18:55, 14 February 2007 (UTC)