Wikipedia talk:Categorization/Archive 17

In the last months, user The Transhumanist has changed the WP:SORTKEY for many outline articles from "asterisk" to "space". For example, see here: , and many similar edits were made by him, see list . Note, "Outline of mining" is not a main article for Category:Mining. Only Mining is a main article for this category. The articles typically called "History of X" ("Types of X", "List of X", "Outline of X") should be categorized with asterisk, per WP:SORTKEY.

From WP:SORTKEY: The main article/s of a category, if existent, should get sorted with a space as key so that it/they appear at the very top of the category. Example: [[Category:Example| ]] Those articles are typically homonymous or at least synonymous to their category. Furthermore other general articles that are highly relevant to the category should be sorted with an asterisk as key so that they also appear at the top of a category but beneath the main article/s. Example: [[Category:Example|*]] Those articles are typically called "History of example", "Types of example", "List of example" or similar.

I think, it's a very bad situation when "Outline of X" is located above the main "X" article in the related category. See Category:Wine for example, Outline of wine is located above Wine. The readers want to see the main article first, not "outline" of something else, so space as a sortkey should be used for only one main article (in most cases). I ask to restore the correct categorization with "asterisk" for all outline articles. 46.211.1.121 (talk) 15:16, 17 April 2018 (UTC)

(The Transhumanist replying): Here's a copy of the end of the corresponding thread from my talk page (46.211 never checked back for an answer, and there was no way to {{ping}} them):

Thank you for correct explanation finally. Now I understand the reasons of your edits, but your logic is absolutely wrong. It's a very bad situation when "Outline of X" is located above the main "X" article in the related category (see Category:Wine for example, Outline of wine is located above Wine). Readers want to see the main article first, not "outline" of something, so space as a sortkey should be used for only one main article (in most cases). I will start a discussion on related forum to ask what the other editors think about your version of sortkeys for outlines. 46.211.2.10 (talk) 14:14, 17 April 2018 (UTC)

No need. Thank you for sharing your opinions. Your wine example convinced me. I agree with you that all other links should fall after the key article. I hadn't considered whether or not "Index" and "Outline" should appear ahead of the root article. It's not hard picking the root article out from them. But, from library classification and publishing points of view, that would be presenting the table of contents and the index before the subject itself, which does seem awkward, when you think about it. To ensure that they fall below the bare subject, per WP:SORTKEY #10, I'll start correcting outlines' placement with my next maintenance pass, or sooner, if I can figure out how to use WP:AWB to do it. And I've added "Outline of" and "Index of" to the appropriate place in WP:SORTKEY (#10). Thank you for your persistence. I'll try to be more open minded in future discussions, with whomever they happen to be with. Keep up the good work. We're lucky to have you here.    — The Transhumanist   12:04, 18 April 2018 (UTC)

 Done Adjusted sort key for those outlines that appeared before key article in the root article's category, per WP:SORTKEY #10.    — The Transhumanist   12:35, 18 April 2018 (UTC)

SORTKEY #10 states:

10. Use other sort keys beginning with a space (or an asterisk or a plus sign) for any "List of ..." and other pages that should appear after the key article and before the main alphabetical listings, including "Outline of" and "Index of" pages. The same technique is sometimes used to bring particular subcategories to the start of the list.

Now, "outline of" links conform to #10 above, and fall below the root articles in all eponymous categories, though I might have missed a couple or so (there are over 740 outlines).    — The Transhumanist   08:33, 19 April 2018 (UTC)

I changed a few other sort codes from " " to " 1". There are plenty more beginning with A-O (e.g. Geography): are we happy to ignore those because the topic sorts naturally above Outline? A few Portals seem to be sorting oddly: see Category:Logic (unexpectedly wrong) and Category:Powderfinger (unexpectedly correct). I have a list but haven't addressed these as I'm not sure what's going on. It may be that last week's temporary bug is still affecting things. Certes (talk) 10:08, 19 April 2018 (UTC)

@Certes: It is only necessary where the sort order needs to be forced. By the way, thank you for the clean up.    — The Transhumanist   07:32, 20 April 2018 (UTC)

• Now, the situation is better than before. Thank you. But " 1" is a very unclear sortkey for casual editor. But why don't we use the different sortkeys. For example, "+" as sortkey for outlines and portal pages, "*" for lists and history of X pages. So my proposition is next:
  • Use " " (space) as sortkey for main article only.
  • Use "+" (plus sign) as sortkey for portal pages and oulines/indexes.
  • Use "*" (asteriks) as sortkey for lists, history pages and other high-relevant pages.

With related (more clear) changes in #10 rule. Normal or not? 46.211.155.173 (talk) 16:54, 19 April 2018 (UTC)

We have the problem that editors change the sort key regardless of what it is. Therefore, maintenance passes are required to set them back to whatever standard is being used. The important thing is the result we are after: " 1" ensures outlines stay below the root article, but above everything else. Another thing is that most outlines have been up there for years. Readers have come to expect them there. Each outline serves as the table of contents for their subject. But, deferring to the root topic is important, so that the root always comes before the table of contents.

There is no way to force an entry to the top of the "+" or "*" sections. If you follow those with anything, then the entry falls after all the entries that have just the solitary symbol as sortkey.

By putting outlines below the root level, you would have the problem of "Index of" pages coming before "Outline of", which is as you would put it would be a "very bad situation". As I mentioned before, Outlines serve as the tables of contents for subjects, and they come at the beginning of a book. Indexes should never be presented before the table of contents. That is upside down.

Also, "+" comes after "*". Putting outlines in the + section would have other things come before the table of contents. And so "History of ", or "List of ", or "Portal of" could come before "Outline of". And we all know that chapter content should never come before the table of contents; the table of contents should always be at the beginning of the book, after the title.

There is also the problem of what is currently being done. Many things occupy the "*" and "+" sections. It would be a constant battle to keep cleaning out the "*" section to be reserved for outlines and indexes.

If there was any other way to ensure that an outline was the second item of the category, other than including it as the second item in the root section, that would be great, but there isn't.    — The Transhumanist   07:49, 20 April 2018 (UTC)

For my part, I only found about a dozen articles needing to be changed. I agree that " 1" is not an obvious choice but it's what the bulk of the other articles used, and I felt it was better to be consistent. Certes (talk) 17:08, 19 April 2018 (UTC)

See my response to 46.211 above.    — The Transhumanist   07:32, 20 April 2018 (UTC)

• @The Transhumanist. So, use the "-" sign as sortkey for outline articles. In this case, the outline articles will be located below main article, but above all others. Section of "-" is above the "+" or "*" sections. 46.211.24.155 (talk) 12:24, 25 April 2018 (UTC)

Further input is requested at Category talk:RuPaul's Drag Race contestants#Sorting --woodensuperman 09:24, 11 May 2018 (UTC)

Further to above, editors have blanketly removed defaultsort keys from all of the articles, so some further input is desperately needed. --woodensuperman 09:30, 14 May 2018 (UTC)

A discussion is taking place at Template_talk:Very_large#Template_name concerning potentially changing the naming scheme from "very large" to an alternative. Input is invited. --Bsherr (talk) 14:39, 15 June 2018 (UTC)

It has come to my attention that several language categories have been inappropriately emptied and tagged for speedy deletion by Jkrn111, and seemingly blindly/carelessly deleted by other administrators (RHaworth, Anthony Bradbury, perhaps others, but these are the prevalent ones in the samples I looked at). Take for example Category:Irian Highlands languages, which was created in April 2011 by another user, was speedily deleted with the rationale "No use, Existing Category:West Papuan Highlands languages". This is obviously not a valid speedy deletion criterion. It was replaced by Category:West Papuan Highlands languages, which was created by the aforementioned user in April of this year without any discussion, and without proper attribution. Similarly, Category:Marind languages, Category:Morehead and Upper Maro River languages, Category:Kaure–Kapori languages, among various others, were emptied and deleted under similar fashion. This really has become a mess that should have gone through the CFD process to begin with, and it would be appreciated if others more knowledge of these subjects can look into it. ℯxplicit 00:23, 15 June 2018 (UTC)

WP:CSD#C1 allows speedy deletion of categories that have been unpopulated for at least seven days, but this wasn't even 30 minutes. For example, Category:Marind languages was deleted 23 minutes after being emptied by the nominator Jkrn111. Renaming a category by moving all pages to a new category without discussion at Wikipedia:Categories for discussion is also against process. It can be difficult for administrators to determine when a category was emptied. I have an alternative account where I sometimes watch a single category and disable "Hide categorization of pages" at Special:Preferences#mw-prefsection-watchlist. Then the watchlist set to period 30 days will show page removals from the category in that period. It only works when the category still exists. PrimeHunter (talk) 01:18, 15 June 2018 (UTC)

Here is another method that sometimes works, to trace the original contents of an empty category. I go back to the start of the page history of the category, then look at the contribs of the editor who created it, at the same date and around the same time. If this shows what pages were initially put into the category, I then look at the recent edits on those pages, to see whether they have simply been removed, or were added to a replacement category out-of-process. This method only works where the same editor created and initially populated a category; it cannot help where an article was put into a non-existent (red-linked) category, and the category page was only created later by another editor.

Anyway, I agree that there is a problem here. @Anthony Bradbury, Anthony Appleyard, and Pppery: would you mind explaining how you come across these empty category pages, and how you decide to blank or delete them? I note that Category:Empty categories awaiting deletion states that categories emptied out-of-process are not eligible for speedy deletion under WP:C1, although that rule is not currently stated at WP:CSD. – Fayenatic London 21:43, 15 June 2018 (UTC)

Uh, what do I have to do with this? The only thing I do that could be construed as C1ing categories is to convert other users' freeform nominations into proper {{db-c1}} tags. {{3x|p}}ery (talk) 21:45, 15 June 2018 (UTC)

@Pppery: you removing the invalid "delete" tag from Category:Marind languages but then just blanked the page, instead of adding "db-c1". – Fayenatic London 16:31, 16 June 2018 (UTC)

Must have failed to notice that the category was empty. {{3x|p}}ery (talk) 16:32, 16 June 2018 (UTC)

• @Pppery, Fayenatic london, Anthony Bradbury, PrimeHunter, and Explicit: I found the empty categories via Category:Candidates for speedy deletion. Anthony Appleyard (talk) 22:41, 15 June 2018 (UTC)
• @Pppery, Fayenatic london, Anthony Appleyard, PrimeHunter, and Explicit: Likewise. ----Anthony Bradbury^"talk" 10:10, 16 June 2018 (UTC)

  The problem was that the nominator emptied the categories and immediately nominated them for speedy deletion so they showed up in Category:Candidates for speedy deletion right away. Considering it's difficult to check how long a category has been empty, maybe we need a rule that speedy deletion of an empty category must wait seven days after the nomination. The deletion nomination could be dated like {{Di-orphaned fair use}}. PrimeHunter (talk) 10:25, 16 June 2018 (UTC)

  Template:Db-c1 already does this... or at least it tries to. It has the following code:

  {{cat handler
  | all = {{ #ifexpr:{{#time:U}} >= {{#time:U|{{REVISIONTIMESTAMP}} + 7 days }} |[[Category:Candidates for speedy deletion]][[Category:Candidates for speedy deletion as empty categories]]|[[Category:Empty categories awaiting deletion]]}}
  | nocat = {{{nocat|}}}
  | category2 = {{{category|¬}}}
  }}
  

  What this does is that the page is put in Category:Empty categories awaiting deletion if the seven days have not yet been reached, instead in both Category:Candidates for speedy deletion and Category:Candidates for speedy deletion as empty categories. We do have some trigger-happy admins who are perhaps looking in the wrong category. --Redrose64 🌹 (talk) 11:04, 16 June 2018 (UTC)

  You are right about how Template:Db-c1 works. Looking at the deleted page histories, Jkrn111 didn't use {{Db-c1}} but {{delete}} which redirects to {{Db}}. This does add the category to Category:Candidates for speedy deletion right away, but it didn't give a valid criteria for speedy deletion of a category. So the admins looked in the right category but shouldn't have accepted the nominations. Some of the deletions didn't even claim that empty category was the reason although the admins may have been thinking it. PrimeHunter (talk) 11:49, 16 June 2018 (UTC)

  Nobody should use {{delete}} directly. I am aware that Commons encourages it (see c:Template:Delete and indeed c:Template:Speedy), but we are not Commons. If people can't be bothered to state explicitly which speedy deletion criterion they are claiming, then WP:CSD cannot apply and the tag should be reverted. Can we do something (perhaps in Lua) that will allow {{db}} to detect direct use without a wrapper such as {{db-c1}}? If we can, it should display an error message and not the speedy deletion pink box. --Redrose64 🌹 (talk) 21:11, 16 June 2018 (UTC)

  {{db}} is intended for direct use. I'm not aware of any other template which uses it. It is {{db-meta}} which is used by other deletion templates. PrimeHunter (talk) 22:08, 16 June 2018 (UTC)

WP:CAT#T says "Templates should be categorized... not by template content", but this would seem to defeat the whole categorization systems for templates. If I cannot put a template in any non-template category then I cannot find any templates via categories... even if I know they do exists. I would have to know the exact template name or exact template category name... and these are often highly unpredictable.

How does it possibly improve Wikipedia to not let Template:History of Christianity be in Category:History of Christianity? tahc ^chat 03:58, 14 June 2018 (UTC)

That template is already well categorized (i.e. grouped with similar pages) by Category:Christian history navigational boxes etc (as well as wikiproject category on talk page).  Readers have no need to go to the template page (they just see the template contents on article pages) and editors can easily navigate to the template (e.g. by clicking on "V" link) - and thence to similar templates (and other pages intended only for editors to see) via categories etc.

See here for more reasons why putting wp infrastructure pages into article categories is bad. DexDor ^(talk) 07:34, 14 June 2018 (UTC)

(edit conflict) it would go in Category:Christianity templates or one of its subcategories. --Redrose64 🌹 (talk) 07:38, 14 June 2018 (UTC)

But that only works if it is linked up to the pages it needs to be linked to. If someone starts removing a navbox from pages that need it, or just pages that I expect it, then editors cannot find the navbox to put these all back. Sometimes I hav started to create navbox only only to find someone already made a very similar navbox. There seems to be not need for this. tahc ^chat 04:26, 17 June 2018 (UTC)

Just categorize the navboxes properly, as templates. If people are competent at Wikipedia templating, they'll find them, where they belong. If they are not, then, yes, they will sometimes create redundant templates, and we'll merge them as always.  — SMcCandlish ☏ ¢ 😼  14:08, 17 June 2018 (UTC)

I have set up Category:Category redirects with possibilities, for names that are currently redirected but where there is potential to helpfully populate a separate category.

{{R with possibilities}} can now be added to a category redirect page, and it will put the page into the above category. – Fayenatic London 09:37, 21 June 2018 (UTC)

Should Category:SpaceX commercial payloads be the subcategory of Category:Spacecraft launched by Falcon rockets? Maybe correct as of now, but incorrect when SpaceX will have more than one Falcon rocket family (BFR etc). Or it should be the subcategory of parent Category:SpaceX directly with addition of all "spacecraft/payload" articles to both categories. 91.124.117.29 (talk) 23:17, 1 July 2018 (UTC)

Until it's wrong it's not wrong. (Cf. WP:NOTPAPER, WP:NODEADLINE).  — SMcCandlish ☏ ¢ 😼  07:26, 2 July 2018 (UTC)

The current hierarchy looks good to me. We could create Category:SpaceX military payloads and Category:SpaceX scentific payloads in addition to Category:SpaceX commercial payloads. I will also place Category:SpaceX payloads contracted by NASA under Category:Spacecraft launched by Falcon rockets, to be consistent with the payload hierarchy. In response to IP91's concern, I have created a subcategory Category:Future SpaceX commercial payloads to accommodate missions on the manifest which have not yet been launched. We'll handle BFR the day it flies… — JFG ^talk 12:56, 2 July 2018 (UTC)

• A wrong logic. What do you say about the possible situation when "SpaceX payloads" will be launched by another rocket but not Falcon? Be sure, it will be done in the future. For example, Cygnus CRS OA-4 which was the "Orbital Science payloads" (possible subcategory) was not a Category:Spacecraft launched by Antares rockets because it was a Category:Spacecraft launched by Atlas rockets. So, I think, Category:SpaceX commercial payloads, Category:SpaceX military payloads should be the subcategory of Category:SpaceX and all spacecraft launched by Falcon should be listed in Category:Spacecraft launched by Falcon rockets without any subcategories. 91.124.117.29 (talk) 08:04, 4 July 2018 (UTC)

Again, any SpaceX payload, past or future as currently planned, gets launched by a Falcon rocket. We can revisit the issue when BFR starts flying, and even then it may still be called a Falcon rocket (you never know what may happen with Elon's naming schemes). The Cygnus/Antares/Atlas situation is different and has no bearing on the Falcon/SpaceX discussion. — JFG ^talk 09:42, 4 July 2018 (UTC)

I have made several edit requests to semi-protected WP:BLP articles, which involve compliance with my interpretation of WP:CATV. The way I understand it is that all attached categories must be verifiable, and they must also be supported by some kind of prose or indication in the article that indicates membership in the category. So if the late Kate Spade is in Category:American Roman Catholics, then we should expect the article to read, somewhere, "Spade is a baptized Catholic and goes to Mass every Sunday. She spoke about her faith in a CNN interview.[1]" but I have been repeatedly rebuffed by editors who tell me that we do not need to worry about what the article says, as long as the fact is indicated in a source... somewhere (the fact in question is not actually mentioned in any source at all.) So when this guideline says "It should be clear from verifiable information in the article why it was placed in each of its categories." does it really mean what I think it means, or am I simply misguided? 2600:8800:1880:91E:5604:A6FF:FE38:4B26 (talk) 05:12, 15 June 2018 (UTC)

I would be curious to see any discussion in which an editor was advising you that an article could be placed in a category without there being supporting information for the category in the article itself, as, as you indicated, the very second sentence of CATV clearly indicates the information must be in the article being categorized. In short, unless I've somehow misinterpreted you, I entirely agree with your interpretation. DonIago (talk) 13:46, 15 June 2018 (UTC)

More, for the sort of category you discuss here (a religious one) we need both exactly the kind of source you describe (one that clearly states the subject's public self-identification with the religion) and a reason why the religious categorization is relevant to the subject's public life or notability; see WP:BLPCAT. So even if a baptism (by which I assume you mean infant baptism) were documented by reliable sources, it would not be enough. —David Eppstein (talk) 14:00, 15 June 2018 (UTC)

I can trawl through my edit history for all the examples, but the other most recent one is Talk:Andrea James#Semi-protected edit request, 1 June 2018 in which James is listed in several very sensitive WP:EGRS categories and the article never bothers to define her as a member. In fact there was previous discussion about that very issue: Talk:Andrea James/Archive 3#Mentioning "trans woman" in the lead in which it was explicitly decided as expeditious to delete all mention of her self-identification, but the commensurate categories were not removed. 2600:8800:1880:91E:5604:A6FF:FE38:4B26 (talk) 19:02, 15 June 2018 (UTC)

For a slight contrast, I went through Category:Reportedly haunted locations in China (as well as some other countries) and noticed that pretty much none of those article substantiated their category membership. So I emptied out the category. I had to make a request at Talk:Great Wall of China#Semi-protected edit request, 24 May 2018, which was apparently not understood by @Kuyabribri:. My other WP:BOLD edits seemed to go through unchallenged, at pretty low-traffic obscure articles. 2600:8800:1880:91E:5604:A6FF:FE38:4B26 (talk) 20:45, 15 June 2018 (UTC)

• I think for a large part this is a typical problem of Category:People by religion. Long time ago I have been trying to clean out Category:Romanian Calvinist and Reformed Christians and I think back then it contained some 20 or 30 entries, mostly politicians. All my edits were reverted. Meanwhile now in that particular category there are only 7 biographies left, but still only 2 articles really belong here. I would almost be inclined to propose that all people by religion categories to be deleted except for people by their religious occupation or their religious role, expecting that 99% of the content to be kept will consist of religious leaders (including clergy), monks & nuns, religious converts, saints, and religious writers (including theologians, religious scholars, religious poets). While the many people in these categories who are just baptized, do some voluntary work for their local church, or are buried from a church, will be purged. Marcocapelle (talk) 13:39, 24 June 2018 (UTC)

  Sounds great to me, Marco. The patent WP:OR about this stuff is like a firehose, and sometimes it rises to disruptive levels as in all the WP:BATTLEGROUNDing about the "Jewishness" of Bernie Sanders.  — SMcCandlish ☏ ¢ 😼  15:14, 24 June 2018 (UTC)

If an article's content does not support membership in a category, then the category can and should be removed. If supporting content is there, but editors decide (whether for BLP or any other reason) to remove it because it is not sourced, then the article no longer supports membership in the category...and the category should then be removed. postdlf (talk) 15:40, 24 June 2018 (UTC)

I think that part of the problem here is that, once upon a time in many articles, there WAS supporting content for the category, but since the content and the categories are two different entities, they come out of sync. Someone comes along, and for whatever reason, removes information (especially about ethnicity, religion, gender or sexuality) in the article and its corresponding source. But they leave the category hanging, maybe because they don't notice it or don't care. And so there it is, even for many years afterwards. Yes, I agree that many categories should be audited and cleaned up. I recently eviscerated Category:LGBT Roman Catholic bishops which was populated by the creator relying on WP:OR rather than any kind of sourcing or WP:EGRS compliance. The creator was blocked last year for adding unsourced content to articles. 2600:8800:1880:91E:5604:A6FF:FE38:4B26 (talk) 19:34, 24 June 2018 (UTC)

That's not so much a problem as it is just par for the course for Wikipedia. No one is compelled to finish any job, nor is it always easy to see what consequences one edit may have on other content. If you just edit one section of an article, for example, you can't even see what category tags may relate to what you just edited. Nor do all editors care about (or understand) categories. So after editor #1 removes the content relevant to the category from the article body, it often falls to editor #2 to realize there is now an unsupported category tag on the article and to remove it. Seeing an unsupported category may be reason to ask "can this be supported", but it's pretty uncontroversial that if the article doesn't even mention a category then it can just be removed. postdlf (talk) 21:02, 24 June 2018 (UTC)

I know, right? But if it were indeed "pretty uncontroversial", then I never would've needed to post here. 2600:8800:1880:91E:5604:A6FF:FE38:4B26 (talk) 21:41, 24 June 2018 (UTC)

@SlimVirgin: is having trouble coming to terms with this on Talk:Andrea James. 2600:8800:1880:91E:5604:A6FF:FE38:4B26 (talk) 21:58, 24 June 2018 (UTC)

From what I see there, there is disagreement regarding what categorization the content of the article supports, not regarding whether an unsupported category can be removed. postdlf (talk) 23:28, 24 June 2018 (UTC)

It is a very interesting case for me: the article previously supported the categories explicitly, but the editors chose, by consensus, to intentionally remove the person's self-identification entirely, and they didn't even consider that the categories would necessarily need to be removed. It seems that some are now of the opinion that hints and allusions in the article are enough to "clearly" support the categories per WP:CATV. 2600:8800:1880:91E:5604:A6FF:FE38:4B26 (talk) 23:33, 24 June 2018 (UTC)

If that's an accurate summary of the situation, then it's clearly not tenable.  — SMcCandlish ☏ ¢ 😼  00:37, 25 June 2018 (UTC)

Via edit request, I brought up a situation on Talk:Adolf Hitler and was told, quote: "Your explanations are not sufficient, and some are outright stupid. This appears to be a pro-Nazi request." by @Beyond My Ken:. 2600:8800:1880:91E:5604:A6FF:FE38:4B26 (talk) 23:48, 28 June 2018 (UTC)

That's too vague to even tell how it relates to anything.  — SMcCandlish ☏ ¢ 😼  02:03, 29 June 2018 (UTC)

To take one specific example: Adolf Hitler has no mention of Anti-Catholicism but there are plenty in Religious views of Adolf Hitler, which is in subcategory Nazi persecution of the Catholic Church. Read literally (and how else are we to read it?), WP:CATV does ban the main Hitler article from the category. So the only sensible options I can see are 1) remove the article from the category; 2) amend WP:CATV to allow categorisation where the required text is on a related page; or 3) claim that this is a special case (why?) and WP:IAR applies. Certes (talk) 02:04, 29 June 2018 (UTC)

Why is it not a sensible option to (4) write a sentence that summarizes Hitler's anti-Catholic persecution and place it appropriately in the main article with an inline citation? Copy/paste would do the trick! The lede section is full of "gimmes". In fact, there is material in the article you mention - Religious views of Adolf Hitler - about his persecution of Freemasons too, which covers another category. Why not just adhere to CATV like normal articles do, by summarizing what the sources say? 2600:8800:1880:91E:5604:A6FF:FE38:4B26 (talk) 03:58, 29 June 2018 (UTC)

Yes, that's also a sensible option, if you feel that anti-Catholicism is one of Hitler's defining characteristics per WP:CATV. Certes (talk) 10:55, 29 June 2018 (UTC)

No, WP:DEFINING does not apply to what's in the article, the operative policy is WP:DUE. If WP:RS say it then we can write about it. It does not have to be defining or notable, it just has to be verifiable. The categories, on the other hand, do need to be defining, so the threshold for inclusion of cats is higher than that of article prose: a very unfortunate situation for Herr Hitler, whose supporters wish the opposite to be true. 2600:8800:1880:91E:5604:A6FF:FE38:4B26 (talk) 02:23, 5 July 2018 (UTC)

The IPv6 anon appears to be entirely correct on this to me (aside from an error of omission). DEFINING has jack to do with permissible article content. The omission is that WP:NOT#INDISCRIMINATE is also operative policy on this, and even more of a gatekeeper (DUE tends to have more to do with how much you can talk about something in an article; if it's undue to include it at all, it also would failed INDISCRIMINATE, or it was simply off-topic, i.e. WP:COATRACKing).  — SMcCandlish ☏ ¢ 😼  05:49, 5 July 2018 (UTC)

Does anyone else think Category:Racially motivated violence against European Americans may be a bad idea. It might be a better idea to set up a category with a clearer inclusions criteria like "Crimes committed by the Nation of Islam". For now, the category summary says This is a list of specific incidents, individual racists, or hate groups that have committed violent attacks against people because they were European American (or otherwise White people who reside in the United States of America). I find the use of the "European Americans" terminology in a racially-charged context particularly troubling. The terminology itself conflates race with nationality, and mixed race Europeans being considered "non-European" has a long and troubling history. Seraphim System ^(talk) 00:35, 26 July 2018 (UTC)

[NB: I redacted my earlier comment here] Wellllll... I mean, I get pretty much what the category is trying to say... so if "European Americans" terminology in a racially-charged context particularly troubling", what would you prefer? White people? Caucasians? (Either is fine with me FWIW.)

The category is kind of over-populated tho... let's see, I went thru each article, I'll give a YES if it should be in this category, NO if not. From the top:

• Murder of Christian Prince -- NO. Robbery gone bad, basically.
• Art Agnos -- YES -- Zebra murders.
• Melissa King assault case -- NO. Middle school playground incident, "the incident arose from a vendetta between two girls"
• 2017 Chicago torture incident -- YES, at least partly.
• Anthony and Nathaniel Cook -- NO, nothing in the article or refs, except one unsupported sentence in lede.
• 2016 shooting of Dallas police officers -- YES, basically.
• Felipe Espinosa -- NO. He's a 19th century serial killer, the one ref (that I can access) has "according to legend" stuff... one racial claim, that he expressed his "intention to murder 600 'Gringos', including the governor himself, if he and the other members of his gang were not granted property" is unref'd and anyway the "gringo" is mere vulgar abuse not a primary motive IMO.
• Mark Essex -- Hmnh. Arguable. He mainly was after police in general and did shoot a black policeman, but... I dunno. He was just a mean son of a bitch it looks like -- "character and behavior disorders". He did join the Black Panthers tho and was quoted (not very reliably IMO) as being "after honkies"... I'll say YES, I guess.
• Fountain Valley massacre -- NO. Weird case. The police tagged it as a robbery gone bad, but the defense (Bill Kunstler!) "argued in part that the accused were politically motivated victims of systematic race-based civil rights deprivation". I dunno if I buy that; one of the victims was black, and one of the defendants later hijacked a plane. These people were criminals first, sounds like. Anyway the jury didn't buy it either I guess.
• 2017 Fresno shootings -- NO, I think. There's a lot of refs and I didn't read them, but this guy was a career criminal, shot a guy in an argument, and then went nuts and started shooting people generally... "Chief Dyer said that the incident was 'a random act of violence'...A federal law enforcement official said the shootings did not bear the hallmarks of a terrorist attack and appeared to be more of a 'local, criminal matter'", but OTOH there was an investigation into whether it was a hate crime, but doesn't say how that came out. A lot of the article assumes quite a bit of knowledge about the guy's internal mental state -- "The driver of that truck was spared from injury, since he was Hispanic"... have to vet all the refs to be sure. Too much work for now.
• Malaika Griffin -- YES, I suppose... "Griffin became angry when [her neighbor] laid his tools on the sidewalk in front of her house after work" and then shot him... so it was really just an argument... OTOH her dairy was full of stuff like "I am so sick of looking at white people!! I am so goddamn tired of them!! I wish I could kill those no good faggot, pedophilic, rapists, thieves & make it painful, (very)". So I dunno -- racist or just crazy? Kind of odd that this is even an article, but whatever.
• Kill Haole Day -- Uhhhh, I guess NO. The article really only talks about name-calling which isn't really violence, and it's more of a schoolkid meme thing than a real thing, and it's not clear if it's a two-way street or just against whites.
• Knockout game -- NO, there's no examples of this being anti-white racial and plenty of examples of it not, if it's even a real thing rather than a meme.
• 1993 Long Island Rail Road shooting -- YES, I suppose so. I mean really this guy was just seriously batshit insane, but he did talk a lot about hating white people. "During one occasion, Ferguson complained that a white woman in the library shouted racial epithets at him after he asked her about a class assignment. An investigation concluded the incident never occurred" kind of gives you a feel for this guy I guess.
• March 14, 1891 New Orleans lynchings -- NO... Uh well its kind of an odd case... the perpetrators were all white, and so were the victims. What happened here is that a bunch of white people lynched some Italian-Americans largely (or anyway partly) because they were Italian-American, so if you consider "Italian" to be a "race", then on one level it would indeed be "Racially motivated violence against European Americans"... "Italian" is not a race tho, in my book, and this article is a pretty different situation from the other articles in this category.
• Marine Park, Brooklyn racial attack -- NO, and it kind of trivializes what we're trying to do here, which is to document an actual issue... Uh why is this even an article??? It was mainly some name-calling between two groups of middle school girls, altho there was some "reciprocal face smacking"... I mean I'd have to say NO based on it being a fight between some whites and some blacks, rather than a racially motivated attack by blacks on whites... "A crowd circles before calls to police and parents put an end to the 20-minute spectacle. Moments later, the white girls finger the victors to the cops and five black girls are booked for misdemeanor assault"... well yeah, that figures, but "The question of who threw the first punch, and why, is [undetermined]". Sounds like the black public school girls and the white private school girls went at each other and the private school girls' parents then got revenge by pulling strings to get it investigated as a hate crime... nothing to see here. If the article is to be kept I'd suggest moving it to The time some butt-hurt white private school moms turned a playground fight into a Federal case or something...
• Murders of Alison Parker and Adam Ward -- NO, I'd say. First of all, the guy was just stone crazy ("He said Jehovah had told him to act and expressed an admiration for Eric Harris and Dylan Klebold, who together perpetrated the Columbine High School massacre; and Seung-Hui Cho, the perpetrator of the Virginia Tech shooting" etc.), which doesn't mean it's not also racially motivated, but then "In the fax, titled 'Suicide Note for Friend & Family', he described his grievances over what he alleged to be racial discrimination and sexual harassment committed by black men and white women in his workplace, believing he was targeted because he was a homosexual black man"... so... his (imaginary I guess) grievances also including gay harassment, so... I guess not. It's a long article and maybe I'm missing something.
• Yahweh ben Yahweh -- YES, I mean "[His followers] murdered white people as an initiation rite to his cult" is kind of a giveaway, if it is true... is it? All the refs say he was "accused" of this, but I guess it's credible given that his cult was black supremacist and he was... he was... let's just say I wouldn't call this guy Mr Happy exactly...
• Zebra murders -- YES. Actually there's no discussion or indication of racial motivation in the article, beyond "a string of racially motivated murders" in the lede... Huh. But all four of the perps were black, and were Black Muslims, and all known 23 or attempted victims were white, so that can't be coincidence.

So let's see, that is 8 kept in, 11 to removed (pending a wait for objections/discussion, if any). Eight is enough for a valid category. But it these eight, rather than the full 19, that should be the focus of whether this category makes sense, I guess. My two cents is that the category makes sense. Herostratus (talk) 03:26, 26 July 2018 (UTC)

There are a lot of people of mixed descent would would not be considered "white" who have European ancestry. I agree that some of these appear to be racially motivated, though some obviously need to be removed from the cat like Felipe Espinosa — but none of the articles discuss European Americans - even the quotes you pulled out like ""[His followers] murdered white people as an initiation rite to his cult" confirm this.Seraphim System ^(talk) 03:38, 26 July 2018 (UTC)

"White people? Caucasians?"

They are not interchangeable definitions for groups:

• European Americans. People of known (or self-reported) European ancestry. In the 2010 census, 223,553,265 people claimed European ancestry, 72.4% of the American population.
• White Americans. People of known (or self-reported) ancestry from "any of the white racial groups of Europe, the Middle East, and North Africa. Most of the Middle Eastern Americans are considered part of this group.
• "Caucasian" may refer to either the Peoples of the Caucasus (50 different ethnic groups which are primarily located in the Caucasus region), or to the so-called Caucasian race. As a proposed racial classification of humanity, it includes "some or all of the ancient and modern populations of Europe, the Caucasus, Asia Minor, North Africa, the Horn of Africa, Western Asia, Central Asia and South Asia. By the definition of physical anthropologist Carleton S. Coon (1904-1981): "This third racial zone stretches from Spain across the Straits of Gibraltar to Morocco, and thence along the southern Mediterranean shores into Arabia, East Africa, Mesopotamia, and the Persian highlands; and across Afghanistan into India [...] The Mediterranean racial zone stretches unbroken from Spain across the Straits of Gibraltar to Morocco, and thence eastward to India[...] A branch of it extends far southward on both sides of the Red Sea into southern Arabia, the Ethiopian highlands, and the Horn of Africa." Dimadick (talk) 12:12, 26 July 2018 (UTC)

'I agree, I think if we rename it "Racially motivated violence against white Americans" and remove the articles Herostratus listed above the remaining eight should be enough for the category - there has been some racially motivated violence against white Americans "stated he wanted to kill white people" etc. Seraphim System ^(talk) 14:51, 26 July 2018 (UTC)

Tend to agree that a move to "Racially motivated violence against white Americans" is an improvement. I agree that it's a real thing that people would want to be able to read up about, so the category is helpful. Altho I think that in either case most readers will understand what we're getting at here, I guess "white Americans" is better than "European Americans" for various reasons. "Caucasians" is becoming a kind of outmoded word so that's out I'd say. However, I think that maybe "white people" would be best, as there are surely many incidents in colonial countries etc. and I can't think of a good reason why these shouldn't be eligible also. Herostratus (talk) 15:07, 26 July 2018 (UTC)

We have an article on White people, but no relevant category. Dimadick (talk) 18:13, 26 July 2018 (UTC)

We have Category:Racially motivated violence against white people - "Racially motivated violence against white people in the United States" is another option for subcategorization Seraphim System ^(talk) 14:29, 27 July 2018 (UTC)

Oh, did not know that! Yes let's do this. Herostratus (talk) 19:30, 27 July 2018 (UTC)

Yeah, that would be the most consistent approach.  — SMcCandlish ☏ ¢ 😼  01:36, 1 August 2018 (UTC)

I'm thinking about creating a new Category:People with disputed ancestry claims to include people like these:

• Rachel Dolezal#Racial identity controversy
• Jimmie Durham
• Jamake Highwater#False ancestry claims
• Andrea Smith (academic)#Cherokee ancestry claim

I would add a note to the category page like this:

Note:This category is for people who made claims about their own ancestry which have been the topic of substantial disputes, regardless of whether these debates have been settled.

I reckon this may be a controversial category, so I wanted to check here if anyone had input on inclusion criteria or had an idea for a better category name. Daask (talk) 20:16, 30 July 2018 (UTC)

What's the encyclopedic purpose of this? I just wrote WP:Race and ethnicity a few weeks ago, and this fits right into the issues raised there of over-focus (especially American and British over-focus) on racialist thinking, which a category like this is apt to encourage.  — SMcCandlish ☏ ¢ 😼  01:36, 1 August 2018 (UTC)

There's a lot of instruction about where Paris belongs, but nothing really about where Category:Paris belongs. Because that category contains people, sport, crime, buildings, history, and a slew of stuff that are clearly not Category:Cities in France, ought Category:Paris not be included in Category:Cities in France, or frankly any of its current parent categories? Obviously, that's not what's intended (or is it?) but having categories having both articles and identically-titled categories included isn't ideal, especially when the categories are supposedly diffusing: contrast Category:States of the United States with Category:Ceremonial counties (of England). Which of these approaches is correct per WP? Or are we to assume that those looking for subdivisions of one country want one thing and those looking for subdivisions of another aren't? Harmony may never be achievable, but I would hope that we could come to consensus and, if necessary, amend the page to reflect the consensus. Carlossuarez46 (talk) 21:03, 7 August 2018 (UTC)

There needs to be structure in the categories. "people, sport, crime, buildings, history, and a slew of stuff" relate to Paris in specific ways that need to be captured, e.g., "people :verb Paris", "sport :verb Paris", "crime :verb Paris", "buildings :verb Paris", " history :verb Paris", "slew of stuff :verb Paris" etc. where :verb is some kind of relation between :topic_noun and Paris. I propose that the structure be of the form S v O, an English sentence with subject S, some kind of relation v, and object O. --Ancheta Wis   (talk | contribs) 21:44, 7 August 2018 (UTC)

Sometimes, putting a child category C into a parent category P implies the very useful information that all Cs are Ps. (All Scottish musicians are British musicians.) Sometimes it doesn't. We desperately need the missing piece of metadata as to whether the implication is true (or at least intended) in each case. I don't think there's currently a syntax for doing that, though we can do some heuristics (e.g. it's likely to be true if P is diffusing). There are several similar discussions in this talk page's archives. Certes (talk) 22:10, 7 August 2018 (UTC)

• Ought we venture to craft some language to include on the page to capture this. I participate at WP:CFD often, and many well-reasoned policy-based arguments are made where articles in a daughter category aren't properly in its parent category - or if some category were deleted or renamed articles would become miscategorized. The all Cs are Ps reading of categorization is used/implied by many WP users. That becomes problematic most clearly in WP:BLPs where such implications may be contentious and unsupported by reliable sources. Should the page say that "all Cs are Ps" is the norm at WP, unless the category page advises otherwise, or some language to not parent categories in such a way that implies the contrary. That way could make use of a Category:Categories named after cities in France (P) and advise that Category:Paris (C) belongs there and on P's page state "that not all C's daughters (and sub-, etc.) are P's"? Carlossuarez46 (talk) 19:45, 8 August 2018 (UTC)

Category:Set categories is one attempt to do the job. If it were used universally, editors and tools such as PetScan could limit their search to set categories. For example, we could safely conclude that Nancy, France is a French city via set Category:Prefectures in France without also concluding that Nancy Mitford is a French city via non-set Category:Paris (and set Category:People from Paris: any non-set link breaks the chain). But there may be a million set categories to label. Certes (talk) 20:39, 8 August 2018 (UTC)

That would only work if articles are non-diffusing from set categories to child non-set categories. E.g. if our article for Paris is not listed in the set category for cities in France (because it is in non-set subcategory Category:Paris instead) we would not be able to conclude that Paris is a city in France. I think the real problem is not distinguishing between two types of categories, but between two types of category membership: some subcategories indicate that the articles in the child category are a subset of the articles of the parent category, while others indicate only that the main topic of the subcategory belongs to the parent category. —David Eppstein (talk) 20:51, 8 August 2018 (UTC)

Is it valid to diffuse to a non-set, or does diffusing a set mean moving articles into subsets? The general solution is certainly to mark the membership rather than the category but, as we have many more memberships than categories, marking categories would be easier if it works. Certes (talk) 21:20, 8 August 2018 (UTC)

What about using the quantifier (logic) annotation — ∀ meaning "For all", ∃ meaning "For some" and "There is a". A default categorization might have to state ∃P "There is a Paris", meaning the eponymous category based on the main article about Paris (a non-set category). Later, as more articles about P arise, and more experience is gained, it might be valid to assert ∀P (a set category).

In other words, we could distinguish is-a "∀P" categories from relates-to "∃P" categories, using the quantifiers. There could many more kinds of "relates-to" categories than "set membership" categories, because the "relates-to" categories would be more general sentences than "is-a" sentences. The relation could be stated in words on the sub-category page. --Ancheta Wis   (talk | contribs) 11:02, 9 August 2018 (UTC)

Yes, set categories do correspond to is-a. I've not looked into the nuances of other types of category but the important thing is that they're not "is‑a". Only with is-a categories can we validly use set logic such as ${\displaystyle \forall C\subset P:a\in C\implies a\in P}$ to conclude that article ${\displaystyle a}$ qualifies for parent category ${\displaystyle P}$. It's not yet quite clear whether is‑a‑ness is a property of C, P or category membership (${\displaystyle \subset }$). But I was trying to stick to English! Certes (talk) 16:51, 9 August 2018 (UTC)

There is a discussion ongoing at Template talk:Main#Category namespace that may have relevance to certain sections of this guideline. Any constructive input would be appreciated. -- Black Falcon ^(talk) 04:13, 10 August 2018 (UTC)

I've had a disagreement with one editor who is very keen on project tagging large numbers of category pages, so I'm coming here for wider input. Should the mass creation of talk pages of categories (containing only wikiproject tags) be encouraged or discouraged?

The only advantage of tagging I could think is that the category will show up in the project's article alerts systems if the category is nominated at CfD, but I believe it's much more efficient to tag categories only if (and when) they do come up at CfD. Other than that, are there any reasons a project might want to track its categories? Given the large number of categories out there, and the lack of distinctions in quality or importance ratings, I'm not sure I see any point.

On the other hand, the existence of a category talk page can be a minor maintenance nuisance. First off, it adds an extra step in the process every time a category is renamed or deleted, though that's not really significant. A more important consideration is in the same direction as the reason why the {{WikiProject Disambiguation}} banner should not be placed on dab pages: when making major changes to a category, it's helpful to see if there have been previous discussions on the talk page, and if talk pages aren't generally project tagged, then this involves simply glancing at the talk page link: if it's blue, then there might have been a discussion, if it's red, then there isn't. This wouldn't work if all these links are blue.

What should be the relative weight of the disadvantages and the benefits? Are there any considerations I'm not aware of? – Uanfala (talk) 09:41, 14 August 2018 (UTC)

I would recommend that users take a look at Category:Category-Class articles and its subcategories (e.g. Category:Category-Class Architecture articles). Category tagging has happened on over 100,000 talk pages and has been happening for at least 12 years. ―Justin (koavf)❤T☮C☺M☯ 09:47, 14 August 2018 (UTC)

A difference that I can see in the case of disambiguation pages is that there's already a template to mark them, making WikiProject Disambiguation tagging redundant. I've tagged categories for WP:SKEPTIC at times myself, although not massively. Some categories are obviously of interest to some projects and I don't see a problem with marking them if done correctly. —PaleoNeonate – 16:04, 14 August 2018 (UTC)

{{WikiProject Disambiguation}} is often used to mean "this page intentionally left blank", especially when the corresponding mainspace page has changed from a redirect into a dab. We rarely create talk pages just to hold that banner. Of course, category pages can't get repurposed in that way. Certes (talk) 16:28, 14 August 2018 (UTC)

I've generally added wikiproject tags where I've come across a category page with a redlink talk page (including category pages I've created) - partly to "fix" the redlink (category pages, unlike articles, don't normally have redlinks). Of course, if consensus is that it's better to leave it as a redlink then I'd stop. DexDor ^(talk) 19:44, 14 August 2018 (UTC)

Whether to add a WikiProject banner template to a category talk page (or any other kind of talk page, for that matter) is a decision that each WikiProject reserves for itself. If a WikiProject tells you (either directly, on your user talk page or through the edit summary of a revert; or indirectly, by having a "Project scope" section (or similar) on their main WikiProject page) that they don't want cat talk pages to be tagged, it's best to honour their wish. This is one of the few areas where WP:OWN does not apply. --Redrose64 🌹 (talk) 20:29, 14 August 2018 (UTC)

Sounds reasonable. But what should be the default choice if a wikiproject hasn't stated a preference? – Uanfala (talk) 20:43, 14 August 2018 (UTC)

One thing I would try is to add the WikiProject template and preview without saving. If one of the rows in the banner begins with a yellow rectangle containing the word "Category" in blue, followed by the text "This category does not require a rating on the project's quality scale", you're probably safe. But if it shows a white rectangle, with "NA" in blue followed by "This category does not require a rating on the quality scale.", I would omit the banner template. --Redrose64 🌹 (talk) 20:55, 14 August 2018 (UTC)

Thanks! I've just had a look at {{WikiProject Languages}}, and the behaviour on previews seems to suggest tagging the category is fine. But then I wasn't able to find anything in the template's code that explicitly does anything for categories, so does that mean that this behaviour is the default of the metatemplate? That is, a white rectangle with "NA" will only show if the specific banner template has been specifically tweaked in a way that discourages categories? – Uanfala (talk) 21:34, 14 August 2018 (UTC)

The code for Template:WikiProject Languages includes |QUALITY_SCALE=subpage, which means that the various page types are defined in the custom class mask. --Redrose64 🌹 (talk) 21:47, 14 August 2018 (UTC)

Uanfala wrote "I believe it's much more efficient to tag categories only if (and when) they do come up at CfD" – but who would tag the categories? IMHO it's unrealistic to impose this as a duty on nominators. Like DexDor, I make it a habit to add project tags on redlinked category talk pages, mainly in order to generate future alerts, and regardless of whether projects currently make other use of the info. I believe that assessment as "class=Category, importance=NA" is automated in most cases. – Fayenatic London 23:17, 14 August 2018 (UTC)

Yes, class and importance are autodetected for all namespaces except Talk: (the talkspace of article space). Class and importance are also autodetected for the talk pages of redirects in all namespaces. The preceding two sentences apply for all WikiProject banners that are built around {{WPBannerMeta}}, which in practice means all except about six of them. In short: you only need to worry about these two parameters for the talk pages of articles and the talk pages of disambiguation pages. --Redrose64 🌹 (talk) 18:58, 15 August 2018 (UTC)

Just to note that one of the examples in the Non-diffusing subcategories section appears to have been changed since the documentation was written. Category:Western Europe does not include the countries - they are within the Category:Western European countries subcat (although the {{All included}} template is still present). Nzd (talk) 04:45, 12 August 2018 (UTC)

I have changed the example to Category:Mountains of Switzerland (more or less randomly). Happy for anyone else to change this if there is a more appropriate example. I've also removed the {{All included}} template from Category:Western Europe. Nzd (talk) 09:42, 13 August 2018 (UTC)

I've also just noticed that this is used as part of the main example in the Guidelines for articles with eponymous categories section, which is obviously now incorrect. Nzd (talk) 21:40, 15 August 2018 (UTC)

We have standard templates to display links to decade and year categories for (dis)establishments, e.g. see Category:20th-century disestablishments in Germany.

Template:EstcatCountryCentury has a parameter to suppress the table for centuries where the detailed categories have been merged, e.g. Category:14th-century establishments in Luxembourg.

Where a country's name changed during the century, some editors have been making tailored tables, covering only the relevant years. I have been compiling a list of these at Template talk:EstcatCountryCentury. In some cases these only cover part of one or two decades.

Hike395 (talk · contribs) recently deleted some of these tailored part-century tables with the edit comment "rm odd formatting in category space using AWB". I reinstated some of these, but then experimented with combining the years for different country names into one template to be used on the century category for both names.

Do editors find the partial-century table or the multi-name table more useful, or have any other suggestions for improvement? – Fayenatic London 20:37, 18 August 2018 (UTC)

The multi-name table is more informative, and it keeps navigation more consistent than the partial-century table. Thanks for your efforts! — JFG ^talk 20:45, 18 August 2018 (UTC)

Your feedback is requested regarding a possible issue of over-categorization. Please discuss at Talk:The Aversion Project#Over-categorization. Thanks, Mathglot (talk) 23:30, 14 October 2018 (UTC)

Hello. Koavf and I are disagreeing on the proper categorization of eponymous categories. See the discussion: Category talk:Black Francis#BRD discussion: Eponymous categorization. It has become clear that, regardless of whoever is right, there may be a lot of pages that would have to be changed. So the question is: Which categories should eponymous categories be placed in, and under what circumstances? — Mr. Guye (talk) (contribs)  01:32, 16 October 2018 (UTC)

Should a subject be redundantly included in main and subcats, or just the most-specific subcat without redundancy? Not factoring in exceptional cases.

As an example, say a rapper Soulja Boy. Should he be included in all three of:

• Category:American rappers
• Category:African-American rappers
• Category:African-American male rappers

Or just the last one? The project page is a little vague in it's wording about this topic of diffusion, I wish it were more straightforward.

Then there's also Category:American male rappers as well. There's just way too much redundant categorization for a lot of subjects. Is this encouraged or discouraged? DA1 (talk) 14:02, 19 October 2018 (UTC)

Category:American rappers bears a {{Diffusecat}} banner which means that if an article qualifies for any of its subcategories (these include Category:African-American rappers and Category:African-American male rappers), that article should not be in Category:American rappers at all. Although Category:African-American rappers does not also bear that banner, I really don't think that articles that qualify for Category:African-American male rappers should be in Category:African-American rappers as well. This is, basically, WP:DIFFUSE vs WP:DUPCAT. --Redrose64 🌹 (talk) 20:45, 19 October 2018 (UTC)

In general ethnic or gender-based subcategories should be marked as non-diffusing, to avoid ghettoizing members of those categories and keeping them from being visible in the main categories. I don't understand why Category:African-American rappers etc haven't been marked in that way. —David Eppstein (talk) 21:14, 19 October 2018 (UTC)

@Redrose64: Thanks for the response. I also agree that "African American male rappers" should be a diffused subcategory of "African-American rappers". There just seems to be an excess of redundant categorization among articles of hip hop artists. Since the "American rappers" article is already marked with diffused, that means that most rappers should only be included at Category:American male rappers or Category:American female rappers. So let me ask you, should [Soulja Boy] be included in both "American male" and "African-American male" or just the latter?

@David Eppstein: In the case of rappers, the overwhelming majority of rappers are actually African-American, so the risk of "ghettoizing", I assume that means marginalizing, isn't a risk on this particular topic as it is in others. DA1 (talk) 11:26, 20 October 2018 (UTC)

I think that Soulja Boy should be in the most specific category, no need to put them in parent categories too. --Redrose64 🌹 (talk) 22:05, 21 October 2018 (UTC)

@Redrose64: I only used his name as an example BTW. He's not even the one effected but there's hundreds of rappers articles that seems to be redundantly categorized. DA1 (talk) 22:33, 21 October 2018 (UTC)

I realize that this has been brought up previously but I thought that I would bring it up again because I want to see some consistency on this project. There are a select few subcategories within Category:Unincorporated communities in the United States by state that only contain subcategories (NJ, RI, NY, MA) For Category:Unincorporated communities in New Jersey, is it wrong to place all entries within this category, like all of the other 46 states have included (2 redirects are currently within this category). My reasoning is that these unincorporated communities can be categorized both by county and state. So the reader has a choice of searching through either by the communities specifically sorted just in that county, or have a whole list within the entire state. See Category:Unincorporated communities in Pennsylvania that contains over 1400 entries. Each of these entries contain both the county category and state category. How come NJ and others should be treated differently? Tinton5 (talk) 21:28, 13 October 2018 (UTC)

@Tinton5: You asked exactly the same question at Wikipedia talk:WikiProject Categories#Subcategory placement nineteen hours earlier. Please see WP:MULTI. --Redrose64 🌹 (talk) 22:25, 13 October 2018 (UTC)

Well because nobody answered there. I moved it to one place, here. It’d be nice instead of pointing out I posted something twice, that we can hear your feedback on the topic of categorization. Tinton5 (talk) 07:53, 14 October 2018 (UTC)

So nobody answered after nineteen hours. Boo-hoo. Remember that 02:30 (UTC) is the middle of the night in Europe and late evening in the eastern United States; some people only edit in the early evening, between evening meal and bedtime. We have discussion forums where it is considered good practice to wait a whole week before assuming that nobody will be answering. I am not obliged to give feedback on the topic of categorization; and nor is anybody else: we are all volunteers here. Maybe other people saw your original post, and are even now considering the best reply before posting it. Maybe they saw it but don't know the answer. Maybe the people who actually care about this haven't seen the post yet - perhaps they only check their watchlists once every 24 hours (or longer); maybe they only log in once a day. Maybe they're Jewish and refuse to use a computer on Shabbat. Maybe they participated in those previous discussions and are sick to the teeth with the whole thing and the thought of hacking it out all over again has made them turn away to something more rewarding. Maybe people simply don't care. --Redrose64 🌹 (talk) 19:23, 14 October 2018 (UTC)

This giant paragraph to say what? Would save everyone's time to just have said "Wait a few more days for a possible response". DA1 (talk) 13:47, 19 October 2018 (UTC)

Agreed, Red Rose is just plain sarcastic and unhelpful. I will just sit patient until others are willing to chime in. Tinton5 (talk) 12:41, 21 October 2018 (UTC)

@Tinton5: See if you want to post a redirect message at Wikipedia talk:WikiProject New Jersey. DA1 (talk) 22:37, 21 October 2018 (UTC)

Opinions are needed with regard to the disability categories that were recently added to the Health and appearance of Michael Jackson article, as seen here, here and here. Discussion is at Talk:Health and appearance of Michael Jackson#Category. A permalink for it is here. Flyer22 Reborn (talk) 12:27, 22 October 2018 (UTC)

Category:Non-empty disambiguation categories is a theoretically useful maintenance tool. It groups together all the disambiguation categories which are not currently empty (they should be empty).

However, there is a technical hitch. Non-empty dab categories are added to Category:Non-empty disambiguation categories only when the category page is purged. That doesn't happen unless the page is edited, which is rare.

So yesterday morning, Category:Non-empty disambiguation categories had only 18 subcats. But I ran an AWB job doing WP:NULLEDITs on all 1660 category pages which transclude Template:Category disambiguation ... and the result was 95 non-empty categories listed in Category:Non-empty disambiguation categories.

I have been busy fixing the pages in ambiguous categories, so over 60 of them are now empty ... but they are still listed in Category:Non-empty disambiguation categories.

The only way I can see to make Category:Non-empty disambiguation categories a usable maintenance tool is to have a bot regularly purge all 1660 category pages which transclude Template:Category disambiguation. I suggest that a weekly purge would be good.

I am sure that if I put in a request at WP:BOTREQ, some helpful bot-owner will put in a WP:BRFA request to run this job. However, BRFA won't approve it unless there is a consensus to to do so.

So what do others think? Would you support such a bot? --BrownHairedGirl (talk) • (contribs) 14:43, 12 November 2018 (UTC)

This is a problem that also affects the redirect categories. Timrollpickering 16:47, 12 November 2018 (UTC)

@Timrollpickering: yes, Category:Wikipedia non-empty soft redirected categories also relies on purging. However, the problem there is much less severe, because @R'n'B runs his Russbot at least once a day bypassing the redirects. The only pages which remain in soft-redirected cats are those where the categories are generated by some template which is generating the wrong category name.

There is no bot emptying the ambiguous categories; the nature of the cats is that they need to be diffused manually. --BrownHairedGirl (talk) • (contribs) 16:57, 12 November 2018 (UTC)

This sounds like a task for Joe's Null Bot (talk · contribs) - Joe Decker (talk · contribs), please confirm. --Redrose64 🌹 (talk) 20:39, 13 November 2018 (UTC)

Thanks, @Redrose64. I have posted at User talk:Joe Decker#Purging_Category:Non-empty_disambiguation_categories to ask Joe to pop in here. --BrownHairedGirl (talk) • (contribs) 23:24, 13 November 2018 (UTC)

I've only taken a very quick look at this, but in general problems like this should be extremely easy ... I would imagine that Category:Disambiguation Categories would provide me the set of categories to traverse. Does that sound right to y'all? --joe decker^talk 05:47, 14 November 2018 (UTC)

@Joe Decker: Thanks. Category:Disambiguation categories should define the set, or alternatively it could be defined by category pages which transclude Template:Category disambiguation. Those should be the same, but there might be glitches. Would it be possible to build your set as the union of those two sets?

Do you need to make a BRFA request to add this to your bot's task list? --BrownHairedGirl (talk) • (contribs) 17:56, 14 November 2018 (UTC)

I do need to BRFA and I will today. If I can just work from an existing category it's likely the request will be quick accepted, because it'll just be literally changing parameters to an existing script. --joe decker^talk 19:41, 18 November 2018 (UTC)

Filed, see Wikipedia:Bots/Requests for approval/Joe's Null Bot 14 --joe decker^talk 20:37, 18 November 2018 (UTC)

 You are invited to join the discussion at Wikipedia talk:Stand-alone lists#RfC about redirects to categories. Shhhnotsoloud (talk) 13:21, 29 November 2018 (UTC)

Looking for some feedback on when the proposed ω sort key would be applied. E.g. does this mean that categories like Category:WikiProject Volleyball would be under Category:Volleyball with this sortkey? If so, what does this mean for the division between administrative and content categories in the encyclopedia? Clarification would be handy here. ―Justin (koavf)❤T☮C☺M☯ 01:17, 9 December 2018 (UTC)

• @BrownHairedGirl: as she was interested in the same discussion and can provide perspective. ―Justin (koavf)❤T☮C☺M☯ 01:18, 9 December 2018 (UTC)

As discussed on Justin's talk page, there are some circumstances where a WkiProject is placed in a non-project administrative category, e.g. Category:Sports-related WikiProjects in Category:Sports and games Wikipedia administration. It may be aimed at that.

But whatever its purpose, administrative categories do not belong in content categories. This discussion arose out Justin's repeated attempts to add Category:WikiProject Volleyball to Category:Volleyball :( --BrownHairedGirl (talk) • (contribs) 01:33, 9 December 2018 (UTC)

I strongly support BHG's view here. My essay at User:DexDor/Administration pages are not articles discusses this further (including sortkeys). Perhaps a note should be added at SORTKEY clarifying that the existence of a special character (e.g. Greek) sortkey doesn't override normal categorization rules. DexDor ^(talk) 07:33, 9 December 2018 (UTC)

FYI

 – Pointer to relevant discussion elsewhere.

Please see: Wikipedia talk:Stand-alone lists#RfC about redirects to categories
 — SMcCandlish ☏ ¢ 😼  06:43, 10 December 2018 (UTC)

At Wikipedia:Categorization#Sort keys it says in point #5 that hyphens should be kept in sort values, so for -30- (The Wire), would the current {{DEFAULTSORT:30}} be incorrect? --Gonnym (talk) 22:55, 28 November 2018 (UTC)

I don't think that's what they had in mind when they wrote that hyphens should be kept. I would sort that article without the hyphens.--Srleffler (talk) 17:54, 31 December 2018 (UTC)

Editors interested in categorization are invited to comment at Wikipedia:Australian Wikipedians' notice board § Categorisation of Heritage listed buildings in Melbourne and SUBCAT. Mitch Ames (talk) 08:48, 12 January 2019 (UTC)

Can someone just confirm that articles such as Boriša Đorđević should use {{DEFAULTSORT:Dordevic, Borisa}} rather than {{DEFAULTSORT:Djordjevic, Borisa}}? Thanks GrahamHardy (talk) 07:18, 19 January 2019 (UTC)

@GrahamHardy: Since the categorisation changes of August/September 2016 (see WP:SORTKEY item 1, also Wikipedia talk:Categorization/Archive 16#OK to switch English Wikipedia's category collation to uca-default? and Wikipedia:Village pump (technical)/Archive 149#Sorting in categories unreliable for a few days), you no longer need special provision for diacritics. Use {{DEFAULTSORT:Đorđević, Boriša}} and see the category page. --Redrose64 🌹 (talk) 17:23, 19 January 2019 (UTC)

Makes sense - happy days, my question was more Đ vs Dj rather than Đ vs D, but either way you've answered my question... GrahamHardy (talk) 17:30, 19 January 2019 (UTC)

I think at this point there's a real need to provide some guidance about how this guideline should be enforced, specifically with regard to this section:

Apart from certain exceptions (i.e. non-diffusing subcategories, see below), an article should be categorised as low down in the category hierarchy as possible, without duplication in parent categories above it. In other words, a page or category should rarely be placed in both a category and a subcategory or parent category (supercategory) of that category (unless the child category is non-diffusing – see below – or eponymous).

We have a recurring problem in my part of the world with one editor interpreting this to mean that he should, in each and every situation and without looking at the categories he's dealing with at all, remove every article this applies to from the parent category. This is frequently resulting in category changes that, if considered in context, objectively don't make sense, and for which literally the only possible justification that can be given is "but WP:SUBCAT told me I can!" If taken to a discussion in these cases, there may not be necessarily consensus on how to fix the category tree, but there is inevitably 100% agreement that we should not simply remove all the articles from the parent category.

This is frequently emerging in cases where an article is in both a parent and child category because there's some sort of issue with the category tree, probably requiring discussion as to what to do with it. These cases absolutely need sorting out - but they don't get sorted out without working out what the problem is and what the best way of dealing with it is, and probably a trip to WP:CFD to move things around and practically deal with the issue. A mass removal of articles from the parent category in these situations just exacerbates the existing situation and creates an incredible mess that someone will have to come along and clean up later while resolving the actual category issue. There's nothing in the text I quoted above that actually suggests to people to deal with it by universally removing all articles in this situation from the parent category, but because it's happening I really think it needs explicit amendment to make clear that it's not acceptable to do it 100% of the time without actually considering why the articles are there. The Drover's Wife (talk) 02:23, 13 January 2019 (UTC)

Are you able to link to any edits showing that editors disagree about how the guideline should be interpreted? We might then be able to decide which interpretation is correct and how the guideline can be clarified. DexDor ^(talk) 12:44, 14 January 2019 (UTC)

Recent examples:

• (Heritage listed buildings... is a diffusing subcat of Buildings and structures...) and related discussion WP:AWNB#Categorisation of Heritage listed buildings in Melbourne and SUBCAT.
• (Journalists from Melbourne are Journalists from Victoria (Australia), Australian journalists by state or territory, Australian journalists, Australian non-fiction writers) and related discussion User talk:Mitch Ames#Subcategories.
• (Australian columnists are journalists, are non-fiction writers. Australian sportswriters are sports journalists, are journalists, are non-fiction writers), related discussion: User talk:Michael Bednarek#Category reversions.
• (Chiefs of Staff to the PM are public servants)
• (21st-century New Zealand writers are New Zealand writers)
• (Australian police officers are Australian public servants)

Mitch Ames (talk) 02:31, 19 January 2019 (UTC)

Here's just a sample (though there have been many, many more - this is just the most recent):

• We've seen Mitch pull all historic buildings in Melbourne out of the "buildings by type" category tree because of a random outlier heritage-listed buildings article that was in there (no one agreed that was a sensible thing to do)
• We've seen him pull bestselling non-fiction authors out of the non-fiction writers category and solely categorise them as columnists because they had side gigs as columnists
• We've seen him pull the former head of the Australian Border Force (a bit like the Secretary of Homeland Security in the US) out of the "public servant" category because he had also been a police officer before he got the job
• We've seen him pull public servants who had held six different notable public service offices out of the "public servants" category because there was a subcategory for one of those offices.
• We've seen him remove people from "People from [State]" categories because they were in "Alumni of [University from that State]" categories, though people don't necessarily go to university in their home state

These edits don't make sense in context: the only reason for them, and the only reason Mitch has ever tried to advance, is "but WP:SUBCAT told me I can". This discussion is a typical example: another editor explains, in detail, why Mitch's category edits didn't make any sense, and is met with "it doesn't matter...[because articles were listed in both parent and child categories so WP:SUBCAT says I can do it]". This is not helpful. Every time the specific edits have actually been discussed so far, "always remove the parent without any consideration of what the articles are, what the categories are, or why they're there" inevitably gets zero support, and the discussion inevitably focuses on the more sensible possible outcomes in that particular context that Mitch chooses to ignore every single time.

This desperately needs to clarify that editors need to look at the categories, look at the articles, and why the articles are categorised how they are, and to start a discussion if there's any doubt about what to do. There are cases where "automatically remove the parent" is perfectly sensible - to use a minority example of Mitch's recent edits, removing "People from State" from people who were already categorised in "People from Town [in that state]" - but adopting that as a universal strategy seems to just make an absolute mess a significant proportion of the time when a discussion would resolve the actual cause of the issue rather than removing correct categories and leaving people nonsensically categorised because there was a problem with the category structure itself.

I've noticed Mitch has now stopped starting discussions about his edit sprees when reverted because those discussions have comprehensively gone against him every time it's happened. The miscategorisation of thousands of articles is effectively doing severe damage to the Australian category tree unless all his edits are checked and undone in the significant amount of occasions where they just don't make sense. The Drover's Wife (talk) 03:01, 19 January 2019 (UTC)

Over-categorisation where parent and child categories cohabit survives around the whole of wikipedia despite whatever drovers or mitch might get caught up in. The problem is neither of the protagonists - but the whole category system and the guidelines given. Drover's interpretation and Mitch's interpretation are not adequately accommodated or explained in the current framework of what has been 'set' on fixed policy pages.

I had serious doubts about the edit history of category modifications by former now blocked editor User:Wwikix as I could not understand why some editors had not seen the labyrinthian parallel and intertwined categories he was creating (and in which in a lot of cases have never been corrected since the blocking). From the no-show of anyone to check the Wwikix alterations across a large range of edit, we have the more finer focused items by mitch and drover's. I do not think either help. Drover's notion of 'sense' is not a useful guideline, nor is Mitch's subcat rule editing. I do believe that the combination of parent and child categories needs to be re-examined and turned into more of a higher level review most peoples misunderstandings of what the parent/child category combination constitutes. Keeping it at this level of conversation between or about Mitch or Drovers is missing the point - the policy and explanation need review. JarrahTree 03:25, 19 January 2019 (UTC)

It's just a matter of encouraging discussion instead of formulaic mass edits where there are issues with the category tree. There are poorly-organised category trees all over the encyclopedia, many times due to examples like the person JarrahTree noted, where new editors have not known what they were doing and linked or created categories in a way that has made a mess. Randomly removing categories that make sense in context doesn't help these cases - it just adds to the mess. If we discuss it and proceed with consensus where the answer isn't clear and obvious, or encourage people to consider being WP:BOLD in fixing actual category tree issues instead, we can organise the category tree in a way that makes sense and reduce unnecessary parent/child categorisation with a minimum of drama. The Drover's Wife (talk) 03:41, 19 January 2019 (UTC)

Agree with this—I have generally been pretty careful about overcategorisation, and keep WP:SUBCAT in mind when applying categories. A recent mass AWB edit was removing the "Australian public servants" category where there was a lower level category such as "Australian diplomats" on the article. This was fine in many cases, but in some cases the subject of the article was a public servant in another area other than in the child category (e.g. worked at Department of Treasury, then transferred to Foreign Affairs and became a diplomat and ambassador), I don't think that this rule should apply when it was chronologically correct and not redundant at a stage of the subject's career. For example, I created most of the articles on the Directors General of Security (heads of ASIO)—if the subject worked in several departments in the APS, I included the "Australian public servants" category, otherwise the "Directors General of Security" categories sufficed (and public servants in External/Foreign Affairs were not always diplomats) and these were all removed. Same with Ainsley Gotto, who worked for several APS departments, but because she was in the PM's chief-of-staff category this was removed with the reasoning that the PM's CoS was a public servant—I agree with that, but not in the case where the subject was only in the parent category for some time. --Canley (talk) 08:26, 19 January 2019 (UTC)

Which is why I am in firm support of the exceptions to the rule rather than blanket rule as in - there is a very strong argument for stating that where a person has been something as part of their career, that removing a category in application of the rule is where everything fails in the application of the rule. Understanding the context of an article or its contents is far more important than application of a rule - and this needs to be incorporated into the categorisation process - I do not agree with the amount of categories at Jimi Hendrix - I believe there is something seriously wrong there, regardless of exceptions to the rule - as a counter argument - but Ainsley Gotto does deserve the complexity. Just my 1 dollars worth, there could be more compelling arguments from others, in relation to the difference between Hendrix and Gotto. JarrahTree 08:48, 19 January 2019 (UTC)

The theme I see above is that we should only diffuse if the subcat adequately describes the subject's whole relationship to the parent cat. If a public servant is a diplomat and serves publicly in no other way, diffuse to diplomat. If they also serve as a treasury official (and there's no subcat for that) then keep the public servant category. Do others agree with this guideline and, if so, is it written down anywhere? How would if vary there were a subcat for treasury officials: diffuse to both and remove from main category? Certes (talk) 12:34, 19 January 2019 (UTC)

Yes, I think you've articulated that point better than I did. I'm not sure that it's written down, but if it isn't, it should be. And in your treasury example - yes, in that case, diffusing to the hypothetical treasury officials category and removing from the main category would be the way to go in my book. IMHO, creating missing categories is often a great way of solving these problems - allowing for higher-level categories to be fully diffused while resulting in more helpful categories on individual articles. The Drover's Wife (talk) 20:36, 19 January 2019 (UTC)

I cant see how these issues can be resolved other than by asking people to exercise their judgement. Rathfelder (talk) 22:19, 19 January 2019 (UTC)

I mean, even stating that editors should do that (in those words) would help stop the blanket-not-exercising-judgment approach. The Drover's Wife (talk) 09:28, 20 January 2019 (UTC)

Certes proposes that "we should only diffuse if the subcat adequately describes the subject's whole relationship to the parent cat", e.g. that if an article belongs in "cat:diplomat" (a sub-cat of "cat:public servant") but also nominally in the parent "cat:public servant" for reasons other than being a diplomat (and there is no other appropriate sub-cat) the article should be (directly) in both the parent "cat:public servant" and the sub "cat:diplomat", and asks whether this is written down anywhere. The Drover's Wife agrees, and says that it should be written down.

In fact the current guidelines explicitly and unambiguously say that is not the case, multiple times:

WP:Categorization § Categorizing pages: "if a page belongs to a subcategory of C (or a subcategory of a subcategory of C, and so on) then it is not normally placed directly into C."

WP:Categorization § Subcategorization: "an article should be categorised as low down in the category hierarchy as possible, without duplication in parent categories above it. In other words, a page or category should rarely be placed in both a category and a subcategory or parent category (supercategory)", and – paraphrasing to match Certes' example – the article "Foo" need only be placed in "Category:Diplomats", not in both "Category:Diplomats" and "Category:Public servants". Because the first category (diplomats) is in the second category (public servants), readers are already given the information that Foo is a public servant by him being a diplomat.

There are explicit exceptions to the general rule - WP:DUPCAT and WP:EPONYMOUS - but they don't apply in the examples I cited above.

If there were only a few special cases where editors thought that the guidelines were not appropriate in those particular cases, then we could simply ignore the guideline in those cases, but – as is evident from my and The Drover's Wife's recent edit history – there are many cases, across multiple category trees. If, as some editors have suggested, the guidelines do not make sense in many disparate cases, then perhaps those editors could propose specific changes to the guidelines to see whether there is consensus to change the guidelines. Possibly another general exception rule, in addition to WP:DUPCAT and WP:EPONYMOUS, is required. Certes' example is an obvious starting point.

Bear in mind that while the discussion above regarding people's occupations (public servants, non-fictions authors) covers some of the examples I cited, it is irrelevant to others, eg

• (21st-century New Zealand writers are New Zealand writers).

Editors who still think that articles should be categorized as both "21st-century NZ writers" and its grandparent "NZ writers" should consider how those cases might be included in the proposed changes to the guidelines. It might be difficult to find examples of "NZ writers" who cannot legitimately be completely diffused by "cat:NZ writers by century".

Mitch Ames (talk) 08:16, 20 January 2019 (UTC)

There is nothing in the current guidelines which mandates these edits - it's just that people are interpreting them in ways that lead to ridiculous outcomes. It shouldn't be a controversial statement to say that the point of having categories should be to help readers find what they're after - but we've got editors making mass category decisions not on that basis, but on trying to categorise articles as far down the category as they can without any regard for whether that leaves articles categorised in any remotely logical way. Readers should not have to know that the former Commissioner of the Australian Border Force began his career as a police officer to be able to find him in the public service category tree. For a non-fiction author who writes in numerous areas, or a public servant who has held significant roles in multiple areas, the lowest they can logically go down the tree is the category for that area (non-fiction writers or public servants) unless sufficient subcategories exist to cover it - even if there might be one or two niche subcategories that they can be placed in for small parts of their story. It doesn't follow from a description that normal practice is to place them as low down the tree (which is almost always an obvious practice with uncontentious results outside of articles on people, and often even there) that one must always try to place articles at the absolute lowest it even where it results in a stupid outcome.

As for the NZ writers one: the problem with making masses of edits that are frequently wrong is that editors making the individual checks on those edits have to make the calls about whether those edits actually should have been made that you didn't do. I reverted them because it not clear to me that "NZ writers by century" intended to diffuse "NZ writers", If I'm a reader looking for an NZ writer, does it follow that I should know I need to look in "NZ writers by century"? I'm not sure it does. If it was, then that particular edit is unobjectionable (and I'm not going to argue if any other editor thinks, looking at the category in context, that it was), but when I'm checking the edits of someone with a mistake rate in an editing spree of anywhere between 40-100% (as opposed to just blanket reverting sprees with bad edits) I need to quickly second-guess edits to try to filter the good from the bad in hundreds of edits rather than being able to give them the benefit of the doubt. The Drover's Wife (talk) 09:28, 20 January 2019 (UTC)

it not clear to me that "NZ writers by century" intended to diffuse "NZ writers", – Sub-categories quite commonly diffuse their parent categories. This is explained in WP:CAT, in particular in Categorizing pages, Subcategorization and Diffusing large categories.

If I'm a reader looking for an NZ writer, does it follow that I should know I need to look in "NZ writers by century"? – The blue box at the top of Category:New Zealand writers, that says Pages in this category should be moved to subcategories where applicable. This category ... should directly contain very few, if any, pages and should mainly contain subcategories. does suggest that the editors might not leave the articles directly in that category, and that the reader may need to look in subcategories. Mitch Ames (talk) 11:00, 20 January 2019 (UTC)

This is very disappointing - it is still an on-going discussion between the two main protagonists - I had sorely hoped that someone other than these two enter into the conversation within the larger editing community. There were a few editors who came in on the Wwikix case who seemed to have a handle of the issues - it would be so useful to have fresh faces in this discussion to offer perspectives from out of the confines of the current on-going discussion. Thanks to them (the two main protegonists) for continuing the discussion, I hope you understand the desire for others who not part of this discussion to join in with more than just a fly-by comment...

On-going conversation is now sufficiently elaborated, long and dense - it really needs examination by someone not currently involved, a review or overview would be very useful. JarrahTree 11:08, 20 January 2019 (UTC)

I'm going to argue that the good points above apply to each reason for notability separately. If our diplomat were also notable as a pianist (not just playing for the family in evenings) then we'd follow the logic once to add them to a public service category and a second time to add them to a musical category. It's the same with diplomat and treasury official. We follow the logic once for diplomacy to add to Category:Diplomats from Wherever, then a second time for the treasury to add to Category:Public servants from Wherever because there is no treasury subcat. Of course, this supposes that service at the treasury is notable: if the subject wouldn't have an article but for the diplomacy then we don't make that second addition. Certes (talk) 11:52, 20 January 2019 (UTC)

If our diplomat were also notable as a pianist ... It's the same with diplomat and treasury official. – It's not the same; the important difference (in the context of the WP:CAT) is that pianist is not a sub-cat of public servant, whereas diplomat and treasury official are.

add to Category:Diplomats from Wherever, then a second time for the treasury to add to Category:Public servants from Wherever – Without prejudice to the merits of your proposal or WP:CAT, this is explicitly contrary to the existing WP:CAT guidelines, so could you please state explicitly whether you think we should:

• Ignore the guidelines (here and in the many similar cases)
• Change the guidelines to reflect the categorization as you would do it

Mitch Ames (talk) 12:30, 20 January 2019 (UTC)

I think we should

• Clarify whether the consensus is to follow your suggestion or mine, or whether it's a judgement call for individual editors, as we both seem to be offering reasonable but incompatible interpretations of existing guidelines. Certes (talk) 13:26, 20 January 2019 (UTC)

Could you please link to and quote the specific part(s) of the guidelines that you are interpreting (eg as I did here). Mitch Ames (talk) 13:50, 20 January 2019 (UTC)

WP:CATDD advises us to Add pages to multiple overlapping categories, and WP:Categorization#Categorizing pages says that each categorized page should be placed in all of the most specific categories to which it logically belongs. I think we all understand the guidance but are unclear as to whether to apply it to each notable attribute individually or once to the subject as a whole. So far we've found nothing in writing to decide that question either way, so I'm hoping that a consensus will establish new guidance on this point. Certes (talk) 14:34, 20 January 2019 (UTC)

WP:CATDD advises us to Add pages to multiple overlapping categories – Interesting. CATDD is an information page not a policy or guideline; it's a very short summary of the guidelines, which take precedence. The link from "multiple overlapping categories" is to Category tree organization, whose first sentence is "Categories are organized as overlapping 'trees'", so I suggest that CATDD should probably say "multiple overlapping category trees".

So far we've found nothing in writing to decide that question either way – The three sentences from the guidelines that I quoted or paraphrased in my post of 08:16, 20 January 2019 (UTC) ("not normally placed directly in [parent]", "without duplication in parent categories above it", "not in both [child] and [parent]") seem fairly unambiguous to me. Mitch Ames (talk) 12:32, 21 January 2019 (UTC)

I'm not sure if I count as "uninvolved" or not, as I initiated one of the example conversations linked early in this section. I believe that Mitch's edits are intended to be helpful. However, they look to be done based on formulae and algorithms, not on reading individual articles. For example, Tony Ayers has been secretary of five departments according to the succession box at the bottom of the article, and of course had a career before reaching that level. Very few people would argue for a line in a succession box not indicating that a category would also be appropriate. Only the last two lines have categories specific for those roles. The other three are represented only as category:Australian public servants. Perhaps the "solution" was not just to remove the higher category, but to create and add the missing three categories for secretaries of Aboriginal Affairs, Social Security and Community Services. Reading the rest of the article, perhaps it should also be categorised as Teacher in Victoria and Prison officer. SO instead of just removing one category, the "solution" was to create two or three new ones and add all of those and two others to the article in exchange for the one to be removed. The problem was not that one high-level category was on the article, but that there were several gaps in the category structure. --Scott Davis ^Talk 14:02, 20 January 2019 (UTC)

• Question - Reading through the above, I get the impression that the debate is ultimately between those who see categorization as an identification (or classification) tool vs those who see categorization as a navigational tool (for finding other, similar, articles). The former want categorization to be as narrow a as possible, while the latter want categorization to be as broad as possible. Does this accurately describe what underlies the debate? Blueboar (talk) 15:17, 20 January 2019 (UTC)
  • I've largely bowed out of this so other voices can be heard, but since you asked: no, I don't think so. Misclassifying articles to get them as low down the tree in its present form results in worse outcomes both for identifying article subjects and for navigating to article subjects. The narrow/broad thing is a red herring - in my book, categorising as narrowly as possible is fine as long as it's done correctly and not just for the sake of it (which may require, for example, creation of new categories so subjects can be both narrowly and correctly classified). The Drover's Wife (talk) 00:05, 21 January 2019 (UTC)

From my perspective, the issue is simply whether we follow the MOS guidelines or not, or change them if they are not working. (If there were only a few specific articles, we could ignore the guidelines in those specific case, but this is clearly a systemic problem, not just a few individual cases, so I'm talking about the many general cases here.) The guidelines unambiguously say – in three separate sentences, which I have cited and quoted repeatedly – that articles ought not be in both child and parent categories (with certain well-define exceptions, none of which apply here). There is no mention in any of those three sentences of "inclusion in both child and parent category is OK if there's a separate subcategory missing".

• If those guidelines are wrong – don't "make sense", and/or don't help the reader – in so many cases, then we should change them so that they are right/sensible/helpful. Anyone is free to propose changes and see if there is consensus for that change. Otherwise, in the majority of cases, we should follow the existing guidelines.
• If an editor thinks that one or more new specific sub-categories are required, then that editor should create the sub-categories, put the article(s) in those sub-categories – and leave the article out of the parent category, per the guidelines.
• In those cases where duplication child/parent categories is appropriate, mark the categories as {{Non-diffusing subcategory}} or {{All included}} to indicate that intent.

In some cases the existing category hierarchy may be wrong, so obviously we must fix the category hierarchy first, then revisit the duplicate and/or missing categorization of the articles. Mitch Ames (talk) 13:35, 21 January 2019 (UTC)

The "issue" seems to be in how people are reacting to an acknowledged problem. I don't think anybody has attempted to assert that the category graph is perfect as it is, nor that every article in Wikipedia is categorised perfectly. We all know that the category graph is not perfect, has extra bits that should be pruned, and bits missing that need to be added. It seems that at least one editor (you – Mitch Ames) is finding large collections of articles that have both a parent and a child category, and doing mass edits to remove the parent category, without looking at why this cluster of articles has that problem. Other people (including The Drover's Wife) want to fix the category structure first. I am mostly also in that camp – if you have found a way of identifying a cluster of articles with the same parent-child category problem, then they are an ideal set of articles to use to work out what was wrong with the structure that multiple editors over an extended time all thought that the solution to the individual categorisation of that page was to use those categories. Mitch, it looks like you have the skill to look at the macro problem and identify where the issues are. Unfortunately, the solution that you have chosen is causing angst with other editors. I'd like to encourage you to consider with the next batch that you find, instead of bulk editing to remove the parent categories, to characterise what is common about the problems you have found, and post the conclusions to either this page or WP:AWNB (assuming it's another Australian categorisation problem). It may be that you have found a cluster of articles that together should be in a new (sub-)category, but none of the individual article editors wanted to be the first to make a new category and only put that one article in it, or didn't think they had the skill or time to connect it properly. Canley said above that he/she was creating and editing articles with a particular focus, and included them in the higher-level category as a placeholder for missing finer categories based on other aspects of their career. It serves as a reminder to themselves or anyone else to come back later with a different focus and build those categories. --Scott Davis ^Talk 23:40, 21 January 2019 (UTC)

I think Scott has nailed it there, and I think this would be a good way forward. These are absolutely issues that need resolving, and are often the kind that get missed (and stay that way) because category structure issues are rarely a topic that gets Wikipedia editors excited. Mitch is also right about the last point in his last comment ("In some cases the existing category hierarchy may be wrong, so obviously we must fix the category hierarchy first, then revisit the duplicate and/or missing categorization of the articles") - the issue is that that there are so many of these issues that it is absolutely impossible for any editor to pre-emptively address them so that mass edits on this basis can be made without doing damage, and the assumption that any issues should already have been fixed by someone else is just not sound. The Drover's Wife (talk) 23:51, 21 January 2019 (UTC)

Presumably at the point that an editor thinks that one or more articles are missing a 2nd (3rd, etc) subcategory of the recently-removed parent of an existing (1st) subcategory, that editor could create/add the subcategory, and/or start a specific discussion about the missing subcategory, instead of simply re-adding the redundant parent. Mitch Ames (talk) 12:53, 24 January 2019 (UTC)

This is absolutely not a workable approach - it is already a lot of work checking edits that have been made en masse without any regard as to whether they should have been made. You are picking up legitimate issues with categories - no one disputes this - but your universal solution is broken in a great many of them and you know this - so flag them as you go and then everyone wins, rather than continuing making mass edits you know are largely flawed and expecting people already cleaning after you to do quadruple the workload. The Drover's Wife (talk) 21:45, 28 January 2019 (UTC)

I'm going through Category:Album covers and assessing non-free soundtrack cover images being used in various articles. The parent category is quite large; so, I'm wondering if it might be acceptable to create a new subcategory titled Category:Soundtrack album covers or something similar to make it easier to find these files. Apparently, non-free album cover filess are added to the parent category each time {{Non-free album cover}} is used. Will this be affected is a new subcategory is created for specific types of album cover art? -- Marchjuly (talk) 04:24, 8 February 2019 (UTC)

Have you proposed this at WT:ALBUMS? If so, and there is consensus, then you would need to either amend {{Non-free album cover}} to have a new parameter - say |soundtrack=yes; or create another template to be used instead - say {{Non-free soundtrack album cover}}. --Redrose64 🌹 (talk) 09:03, 8 February 2019 (UTC)

Can't a new subcategory be created and existing files simply manually added to it? For example, there exist subcategories of for album covers by artist in Category:Album covers by recording artist. I wasn't proposing that the files should automatically be added to a new subcategory for soundtrack; I was just wondering if doing so would affect how the copyright template works in adding file to the parent category. Apologize if I wasn't clear about that in my OP. -- Marchjuly (talk) 11:19, 8 February 2019 (UTC)

A follow-on from this discussion, here.

Briefly: given the relative stability in recent months of Category:20th-century male writers and similar categories for writers (and the longstanding stability of categories for male actors by century), I've begun creating and populating similar categories for male musicians and artists by century. My argument is one that's been kicking around for a few years now, in some guise or other; we have women categorized a certain way, and there's no reason we shouldn't be treating male subjects the same way. I've been treading relatively slowly, but haven't really met much formal pushback before the linked discussion. Hence opening this discussion here.

My feeling: we should have men-by-century categories for many of the professions for which there are women-by-century categories. We've got categories for men by profession and country, at least in many of the cultural disciplines, and I don't see any reason why we shouldn't extend it to by-century as well. Others may disagree: I'd be interested in hearing more discussion. --Ser Amantio di Nicolao^{Che dicono a Signa?}_{Lo dicono a Signa.} 21:56, 8 February 2019 (UTC)

I have just created Template:Uw-redcat, and added it to Template:Single notice links.

This is to warn users who add pages to no-existent categories (see WP:REDNOT), causing them to be listed at Special:WantedCategories. On average, 50–100 such redlinks appear every day, and it is nearly a full-time job to keep the list clear.

So far, there has been no standardised warning for this. I hope that the wording I have used makes sense.

I opened a discussion on it at WT:UW#Template:Uw-redcat, and suggest that any further discussion should take place there. --BrownHairedGirl (talk) • (contribs) 07:23, 19 February 2019 (UTC)

I'm sure this has been discussed before (and I've read the recent discussion above), but I can't seem to find a good answer. My specific question is whether the kebab article should be in Category:Levantine cuisine, and/or the geographical subcategories Category:Lebanese cuisine, Category:Syrian cuisine, Category:Jordanian cuisine, etc. It's also a general question about how to categorize food items and dishes, and similar things that are found in multiple geographical areas.

This guideline says each categorized page should be placed in all of the most specific categories to which it logically belongs and WP:SUBCAT says an article should be categorised as low down in the category hierarchy as possible, without duplication in parent categories above it. What does "logically belong" mean, and how low is "as low as possible"? Kebab dishes aren't exclusively Lebanese for example, so if "as low down as possible" is meant to be the category that includes all relevant subcategories, then probably it would have to be Category:World cuisine.

It seems more likely that it means that a dish should be included in all "Category:Country cuisine" categories that notably feature it, and not in any "Category:Region cuisine" categories that are supercategories of those countries. In other words, the kebab article should not be in Category:Levantine cuisine. It should also be taken out of Category:Balkan cuisine and added instead to each of the 11 geographical subcategories (Albanian, Bosnia and Herzegovina, Bulgarian, Croatian, Greek, Kosovan, Macedonian, Montenegrin, Romanian, Serbian, and Turkish), and similarly for Category:South Asian cuisine. What about Category:Arab cuisine?

This would imply that Category:Levantine cuisine shouldn't have any articles about specific dishes listed in it, and that the 100+ dishes currently in the category should be duplicated and moved down into each of the constituent country subcategories. The same would apply to all "Category:Region cuisine" categories; for example no specific dish articles should be present in the categories Category:Mediterranean cuisine, Category:Middle Eastern cuisine, Category:Asian cuisine, etc., or even in Category:World cuisine.

Is this correct? It doesn't seem to reflect current practice very well, as most of the "Category:Region cuisine" categories have many dishes listed directly under them, and often at the same time in the subcategories. It would be a big change to actually enforce the without duplication in parent categories above it part of the guideline. I'm also not sure how desirable that is. But it's inconsistent; looking at the list in Category:Middle Eastern cuisine, one would certainly expect to see the kebab article in there (there are a number of specific types of kebab listed). I can't figure out if I should add it, or remove all the specific dish articles.

It might also cause issues with verification, as the articles may have references to a dish being "Levantine", but not specifically mention the constituent countries. Are we sure that all such dishes are present in Cypriot cuisine for example? This is even more troublesome with the larger categories - do we actually have "Category:Country cuisine" categories to cover every country in Asia? Can we accurately determine to which specific countries in Asia that oolong, cocopandan syrup, and mochi - and kebab - do or don't belong? What countries exactly make up the Middle East?

One more example, Adana kebab is in Category:Cuisine of Adana and also in the parent Category:Turkish cuisine. Since it's served all over Turkey, it doesn't seem like it should be restricted only to the former category, while it wouldn't make sense to leave it out.

There's also the question of categories themselves, for example Category:Syrian cuisine is a subcategory of Category:Levantine cuisine, which is itself a subcategory of Category:Middle Eastern cuisine. It would seem then that Category:Syrian cuisine should be removed from Category:Middle Eastern cuisine. Currently Category:Lebanese cuisine is not in Category:Middle Eastern cuisine; again I can't figure out whether I should add it, or remove the other Levantine countries instead. Also, Category:Kebabs is in Category:Middle Eastern cuisine, but not in Category:Asian cuisine or any of the south/central/east Asian cuisine subcategories. Should it go in any of those, or in Category:North African cuisine, or should it be removed from Category:Middle Eastern cuisine and placed "as low down as possible" in each and every of the Middle Eastern (and Asian, African, European, and even the Americas') "Category:Country cuisine" categories?

Any comments or pointers to previous relevant discussions or consensus are appreciated, thanks. --IamNotU (talk) 17:20, 3 March 2019 (UTC)

This geographical categorization can get out of hand. We should be categorizing articles (based on the definining characteristics of the subject), not attempting to use categorization to create lists of what people eat in each country. I'd suggest not categorizing a food for more than one geographical area (based on where the food originated) - e.g. kebab may belong in Category:Middle Eastern cuisine (or a subcat of that), not in categories for Lebanon, Israel, Pakistan, Iran, Iraq ... DexDor ^(talk) 18:05, 3 March 2019 (UTC)

DexDor, thanks for your comment. I've never paid much attention to categories, so these are probably rather "newbie" questions. Do I understand correctly that you'd suggest not being strict about the "as low down in the category hierarchy as possible", and instead put the kebab article in Category:Middle Eastern cuisine, and remove it from the lower categories like Category:Levantine cuisine, Category:Lebanese cuisine etc.? Or do you think that sometimes being in multiple parent/child categories is ok? It feels odd to remove kebabs from Category:Turkish cuisine for example...

Your comment brings up another question that I didn't want to add to my already long post - should categorization be primarily about the origin of a dish, or where it is a significant part of a particular cuisine? For example, should the kebab article not also go in Category:Central Asian cuisine, or Category:South Asian cuisine (and then, be removed eg. from Category:Pakistani cuisine)? --IamNotU (talk) 02:25, 5 March 2019 (UTC)

The rule about being as low in the category hierarchy applies after you've determined what the definining characteristics of the topic are. In the case of kebab: Jordan (for example) isn't a defining characteristic (the article doesn't even mention Jordan); that's someone (wrongly) using the category to create a list - information about the popularity/history of kebabs in Jordan belongs in the text of articles/lists (e.g. Kebab and Jordanian cuisine) where it can be referenced (similarly for Turkey).  Otherwise it could lead to people creating categories such as "Cuisine of Omar's cafe" and putting the Kebab article in it. DexDor ^(talk) 06:39, 5 March 2019 (UTC)

Regarding origin - absolutely. For example, we (now) categorize weapons (e.g. missiles) only by country of origin; not by every country that uses them, every war they have been used in etc. DexDor ^(talk) 06:39, 5 March 2019 (UTC)

Is there any specific policy or protocol for placing pages within a parent category and subcategory? For instance, you'll see in Category:Public high schools in the United States by state, where N.J. is the only state that does not contain ALL public high school pages (only a subcategory of them broken down by county listing), along with categories with places of worship, municipalities, unincorporated communities, etc. They are only organized by county. Shouldn't all pages be included in these categories (hence this template) since pretty much all of the other US states follow this practice? Only a couple of editors are against this since it was discussed previously. I find it useful for the reader to have the option to view listings by both county and statewide. Tinton5 (talk) 04:14, 21 February 2019 (UTC)

WP:DUPCAT says "some [subcategories] are simply subsets which have some special characteristic of interest". It doesn't provide an exhaustive list of what constitutes a "special characteristic of interest", although it does say that "gender, ethnicity, religion, and sexuality should almost always be non-diffusing". Mitch Ames (talk) 11:41, 21 February 2019 (UTC)

The general principle is to diffuse, per WP:SUBCAT. Non-diffusion creates category clutter on articles, and is hard to maintain because experienced editors will instinctively remove the duplication, and tools such as WP:HOTCAT gives no warning.

I don't see any particular reason for a DUPCAT here. The by-county subcats of Category:Public high schools in New Jersey by county all look quite well-sized.

By contrast, some of the undivided categories for other states could do with subcatting, for example Category:Public high schools in California (957 pages), Category:Public high schools in Texas (757 pages). --BrownHairedGirl (talk) • (contribs) 23:14, 23 February 2019 (UTC)

As someone who has had a lot to say about over-and-inappropriate-diffusion, I have to say I agree with BrownHairedGirl on this specific one - I can't say I see a benefit to having undiffused categories here. The Drover's Wife (talk) 01:09, 24 February 2019 (UTC)

As someone who thinks about diffusion and occasionally rants about it, some properties lend themselves naturally to diffusion and some don't. The acid test for me is: does each article fall naturally into exactly one subcategory? By that yardstick, schools by county seem perfect for diffusion. In contrast, to take another example from above, ethnicity doesn't diffuse neatly: many notable people have multiple, unclear or disputed ethnicity. Certes (talk) 01:36, 24 February 2019 (UTC)

Shouldn't this encyclopedia follow some consistency? General practices are to diffuse and subcategorize each page and/or topic within its parent category, at least that is what I've been told and have seen. I have witnessed categories all over the place which sometimes don't even belong in their present subcategories. Tinton5 (talk) 03:54, 13 March 2019 (UTC)

I've submitted an RfC re: the categorization of all works (albums, songs) by artists by genre.

Please see Wikipedia_talk:WikiProject_Music#RfC_on_categorizing_all_works_by_an_artist_by_genre.

Thanks! ---Another Believer _(Talk) 17:02, 29 March 2019 (UTC)

One of the may deficiencies of Wikimedia's crude category system is that it does not automatically generate a table of contents for the category. Editors have to manually add a TOC if it is needed.

So a few weeks ago, I created Template:CatAutoTOC, which generates a table of contents on a category page if the category size exceeds a certain threshold. It is now used on about 35,000 categories, nearly all via category header templates.

The size thresholds I applied are:

1. < 100 pages = no TOC
2. 100–1200 pages = {{Category TOC}}
3. > 1200 pages = {{Large category TOC}}

However, I just noticed that {{Category TOC}} says it should not be used for categories containing less than 200 pages.

One way or another, that discrepancy needs to be resolved.

I can see the case for the threshold of 200, because it is one pageful, and a TOC is arguably un-needed on one page. Personally, I think that a TOC is still useful on categories in the 100–200 page range, but that may just be an oddity of mine.

What do others think?

What should the size thresholds be? --BrownHairedGirl (talk) • (contribs) 12:37, 31 March 2019 (UTC)

So, I'm thinking particularly of categories like Category:Canadian expatriates in the United States (sorted with the key "-") and Category:Canadian emigrants to the United States (sorted with the key "+") that are subcategories of Category:American people of Canadian descent, even though a significant portion of those expatriates and immigrants aren't/weren't U.S. citizens. Should the subcategorization be replaced with {{category see also}} instead? Or maybe it's enough that they all share the same parent category Category:Canada–United States relations? During the years I've noticed lots of reverting categories back and forth (, ), which is why I'd love to see a conclusion to this inconsistency. --Kliituu (talk) 20:15, 19 March 2019 (UTC)

• Who counts as an American person? Rathfelder (talk) 08:29, 20 March 2019 (UTC)
• It makes it much easier to retain the current tree, given many (if not most) of these people do take citizenship and you don't actually have to have citizenship to be regarded as American (or any other nationality) in anything other than a strictly legal sense in any case. -- Necrothesp (talk) 13:28, 20 March 2019 (UTC)
• It's certainly very unusual for biographical articles to say anything explicit about citizenship or nationality, and I think the reality is that for articles about people who migrate attribution of nationality is just guesswork. Rathfelder (talk) 20:36, 20 March 2019 (UTC)

@Kliituu, Necrothesp, and Rathfelder: I think that there are three issues here:

1. Is there is a useful distinction between emigrants and expatriates?
   Rathfelder and I had that discussion elsewhere, and we disagree: I think the distinction is worth retaining, Rathfelder thinks not. I don't think that can be resolved without an RFC
2. Should expatriates be categorised under descent categories?
   e.g. should Category:Canadian expatriates in the United States be a subcat of Category:American people of Canadian descent?
   I think this question is fairly straightforward: the WP:DEFINING distinction between an expat and emigrant is that the expat does not take up the nationality of the host country. So whenever I encounter an expat category parented in a descent category, I remove it.
3. Navigation between the various categories.
   That is the only plausible argument I have seen for categorising expatriates under descent categories. I don't that navigational convenience justifies such miscategorisation, but it is a reasonable approach. However, I have a solution to that: {{FooBarHumMigNav}}, which I have been intermittently working on for a few months as a Lua module.

There's still a little tweaking to do, but it's nearly ready for rollout. It takes no parameters, and when placed on a bilateral human migration category, it creates a navbox for the categories for descent, emigrants, expatriates and expatriate sportspeople between the two countries.

To demonstrate it I did a few tests on some pages, and self-reverted:

• Category:American expatriates in Japan
• Category:Irish emigrants to New Zealand
• Category:Chilean people of Irish descent
• Category:Spanish expatriates in the United States
• Category:Japanese people of American descent
• Category:Japanese people of American descent
• Category:Finnish emigrants to South Africa
• Category:Ivorian expatriate sportspeople in France
• Category:Venezuelan expatriate sportspeople in Portugal

I'd really welcome feedback on whether this is a good idea, and if so whether it needs tweaking.

Also pinging some other editors whose feedback I'd value: @Oculi, Black Falcon, Marcocapelle, Fayenatic london, and Ymblanter:. --BrownHairedGirl (talk) • (contribs) 14:14, 31 March 2019 (UTC)

I think this is a matter of definitions. If an expat is defined as a citizen of A but not a citizen of B living in B and an immigrant is defined as a citizen of B living in B who previously was (or still is) a citizen of A and who previously lived in A, then the distinction makes sense. (Note that by this definition, most immigrants have been expats first - I myself was a citizen of one country, lived for 15 years in another country, and then applied for the citizenship, meaning I was an expat for 15 years and then became an immigrant). This is not a definition everybody would agree with, and one would certainly need an RfC to move forward. Also, in many cases it is impossible to determine who is a citizen of what country - for example, the edit-warring in Maryam Mirzakhani probably costed my a year of my life, driveby editors would come, change her definition into "Iranian mathematician", and all my explanations that she was educated in the US, had a job in the US, and only published with the US affiliations - would be disregarded because people would insist that I prov she is a US citizen. May be one needs a much broader scope RfC on in which situation can one define a person (and, in particular, a living person) "an American (Canadian, Finnish etc)...".--Ymblanter (talk) 14:38, 31 March 2019 (UTC)

I would be happy with a rule that we forbid categorizing people with a nationality or nationalities unless we have explicit and reliable documentation of their citizenship, and that when we do have such documentation we merely include as categories all documented citizenships rather than trying to decide for ourselves how one of those citizenships relates to another. But too many editors and readers are too invested with waving their flags to make that likely. —David Eppstein (talk) 15:32, 31 March 2019 (UTC)

And I would very strongly oppose that, David Eppstein.

Nationality is one the two basic traits of en.wp's categorisation of people, but it is very rare to have an explicit source declaring citizenship. If we applied David's rule, we'd have to rip apart most our categorisation of people. At a rough guess, that principle would mean that 95% of our biographical articles would cease to be categorised by nationality.

Categories exist to provide navigation between related articles, not to serve as a legally-verified database of citizenship. Our readers are best served by categorising people according to the nationality with they have a clear association. We do not need to concern ourselves with whether they legally became citizens. --BrownHairedGirl (talk) • (contribs) 16:20, 31 March 2019 (UTC)

• I dont regard the distinction between emigrants and expatriates as worth a detailed discussion, because I think it is too messy to be resolvable. There is a great deal of subjective local usage because being an ex-pat is frequently seen as more respectable than being a migrant. You can only definitively distinguish the two in retrospect. Legally there isnt a distinction in most places. My guess is that there is explicit mention of nationality or citizenship in fewer than 5% of biographical articles. For most the best you get is places of residence. So to that extent our categorising by nationality is almost entirely suppositious. We could get round that problem if we categorised biographies by place of residence, but I dont think there will be much appetite for that suggestion.

So I am quite content with BrownHairedGirl's approach, which certainly seems to be an improvement. Rathfelder (talk) 16:49, 31 March 2019 (UTC)

• Agree with BrownHairedGirl on the principle that expatriates and emigrants are different, however in practice it will be difficult. We can probably only be certain that someone was an expatriate in case he/she meanwhile moved to another country or moved back to his/her original country, but if we would stick to that we would be limiting ourselves quite a lot. So I am actually uncertain whether it is useful to keep separate trees for expatriates and emigrants. On the other hand descent is something really different, that should only apply to children and (possibly) grandchildren of emigrants insofar they who were born in the new country. Also I agree that we should not bother about legal citizenship (as mostly unverifiable), the key criterion should be the country of living. Marcocapelle (talk) 19:34, 31 March 2019 (UTC)
• I support all the conclusions and suggestions of BrownHairedGirl. The new human migration nav template should have a longer name for clarity. I would place it below any category description line. In the case of expatriate sportspersons, it can be included at the end of {{Fooian expatriate sportspeople in Bar cat}}. – Fayenatic London 11:26, 1 April 2019 (UTC)

• If we are going this way then expatriates should not be included among people of Fooish descent. They are still Fooish people.
• I've had a little trial categorising expatriate Georgian sportspeople and I think this is the way to go. But had forgotten that sportspeople move about so much. One person may be categorised as an expat in a dozen countries. Ideally I'd like to take them out of the countries they have left, but think that is probably impractical. Rathfelder (talk) 22:20, 31 March 2019 (UTC)

I don't actually see a lot of point in retaining the expatriate categories. As far as I'm concerned, an emigrant is someone who moves to a country and intends to stay there permanently or more or less permanently (e.g. some people emigrate to Britain from the Caribbean, stay for decades and to all intents and purposes become British, but then retire back to the Caribbean; they're still emigrants, even though they eventually return to their country of birth), even if they don't actually do so, or who ends up staying permanently even if they didn't originally intend to. It has nothing to do with actual citizenship. I'm not sure what an expatriate is, as it has different definitions depending on context. Is it a person who lives in a country for a bit? So what? The trouble is, the term "emigrant" often tends to be used of people from developing countries and "expatriate" of people from developed countries, even if their situations are pretty much identical. If we do retain the two separate types of category, however, then I definitely don't think it's worth using both on one article. If someone ends up staying in a country then the emigrant category is sufficient. I also do think both emigrants and expatriates should be categorised under descent for navigational reasons. -- Necrothesp (talk) 07:40, 1 April 2019 (UTC)

• Expatriates are mis-classified by descent at present. The French ambassador to Belgium is not "of French descent". He is just as French as the inhabitants of France. Rathfelder (talk) 08:24, 1 April 2019 (UTC)
  • It's simply for ease of navigation and because so many emigrants have been miscategorised as expatriates. But why's it even worth categorising at all? So he lived in Belgium for a while. So what? Unless he lived there in any sort of permanent way (i.e. was an emigrant) why is that notable? -- Necrothesp (talk) 10:57, 2 April 2019 (UTC)

I'm quite happy to keep the expat categories if they are clearly distinct from the migrants. Ambassadors, governors of colonies and the like with a significant part in the history of the place. I guess we have to accept the sportspeople, but generally we should be looking for people who played a significant part in the place where they were an expat. And if its clear that they were really a migrant then they should be in that category. Rathfelder (talk) 13:45, 2 April 2019 (UTC)

I have opened an RFC about whether to standardise on the "Z" spelling in descriptive category names, i.e. to use "Organization" in all cases. I estimate that this affects the naming of about ten thousand categories.

See Wikipedia:Village pump (policy)#RFC:_spelling_of_"organisation"/"organization"_in_descriptive_category_names. --BrownHairedGirl (talk) • (contribs) 20:07, 4 April 2019 (UTC)

I am trying to create a redirect category page, but it doesn't work. The redirect page is Talk:Whites only, but the category that is displayed is Category:NA-Class Civil Rights Movement articles instead of Category:Redirect-Class Civil Rights Movement articles. What am I doing incorrectly? Mitchumch (talk) 23:48, 7 April 2019 (UTC)

@Mitchumch: I don't think that specific template works with class=redirect. You just need to add the functionality to Template:WikiProject Civil Rights Movement --DannyS712 (talk) 23:59, 7 April 2019 (UTC)

What do I need to do to modify the template to recognize redirects? Mitchumch (talk) 00:15, 8 April 2019 (UTC)

@Mitchumch: See the instructions at Template:WPBannerMeta#Assessment --DannyS712 (talk) 00:22, 8 April 2019 (UTC)

The following parameter appears to be set-up for "extended" in Template:WikiProject Civil Rights Movement.

|QUALITY_SCALE = extended

|class =

Do I need to use "inline" or "subpage" parameters to employ "redirect class"? Mitchumch (talk) 00:50, 8 April 2019 (UTC)

@Mitchumch: I think you need subpage, and then you have to set up the subpage itself. Sorry, I'm not the best person to ask about this - maybe try Template talk:WPBannerMeta? --DannyS712 (talk) 00:52, 8 April 2019 (UTC)

I will do that. This is more involved than I thought it would be. Thanks. Mitchumch (talk) 01:03, 8 April 2019 (UTC)

Using |QUALITY_SCALE=extended won't make the template recognise |class=redirect, because it's not one of the seven classes listed at Template:WPBannerMeta#Assessment. It needs to be either the subpage or inline method; I can do it for you, if I have a clear mandate from the WikiProject. However, I go out to work soon, I can pick this up at (say) 16:00 (UTC), bot not likely to be any earlier. BTW it shouldn't be necessary to explictly set |class=Redirect because the class is autodetected - if the WikiProject banner is not set up for Redirect-class, it defaults to NA-class. --Redrose64 🌹 (talk) 08:43, 8 April 2019 (UTC)

@Mitchumch: OK, this is ready to go. There are two edits required, both very simple: (a) on the main template, alter |QUALITY_SCALE=extended to |QUALITY_SCALE=subpage just like this; (b) on the documentation, alter |QUALITY_SCALE=extended to |QUALITY_SCALE=subpage (so that it matches the main template). --Redrose64 🌹 (talk) 19:06, 8 April 2019 (UTC)

Redrose64 Everything is good. Thank you. Spoke too soon. The project template for Talk:Whites only now displays as "Redirect". However, the talk page does not display in Category:Redirect-Class Civil Rights Movement articles. Any ideas what is going on? Mitchumch (talk) 19:26, 8 April 2019 (UTC)

You can either wait for the job queue (which might be months), or you can go to Talk:Whites only and carry out a WP:NULLEDIT. --Redrose64 🌹 (talk) 19:48, 8 April 2019 (UTC)

Everything is good. Thank you. Mitchumch (talk) 22:19, 8 April 2019 (UTC)

For WP:DRAFTNOCAT please add an info that class=Draft in WikiProject templates on "Draft talk" pages works as expected. While at it the section could also state that any {{stub}} template might violate DRAFTNOCAT, unless it is smart enough to have no effect outside of the article namespace, e.g., {{authority control}} is smart, but {{US-record-producer-stub}} is not smart and caused havoc on my first draft. –84.46.52.44 (talk) 16:02, 31 March 2019 (UTC)

I just tested {{US-record-producer-stub}} on in the Draft:sandbox. See my version.

As you can see, it doesn't categorise when used in draft space. So I can't replicate the problem. --BrownHairedGirl (talk) • (contribs) 23:55, 2 April 2019 (UTC)

Thanks, maybe two contributors tried to fix the same non-existent problem, and I went straight into a rat-hole ending up here. –84.46.53.95 (talk) 04:23, 9 April 2019 (UTC)

It shouldn't be necessary to use |class=Draft on a WikiProject banner template in Draft talk: space - when used outside the main Talk: space, almost all (there are five or six exceptions) WikiProject banners will autodetect the class when there is no |class= parameter. Same with |importance=. --Redrose64 🌹 (talk) 07:39, 3 April 2019 (UTC)

For the missing importance= it's "once bitten, twice shy", it's tricky enough to figure out which project templates support needs-image=yes or attention=yes when I need this, or insist on living=yes inside of {{WPBS|blp=yes|1=…}}. –84.46.53.95 (talk) 04:23, 9 April 2019 (UTC)

Hi. Currently, Category:Categories requiring diffusion has 6,457 subcategories, many of which have nothing to diffuse currently and together making it hard to find what needs work. As early as 2010 it was remarked that the category itself requires diffusion (Category talk:Categories requiring diffusion#Subcategories?). I'd like to suggest that all categories that only have 1 subcategory, and have no direct pages in them, be removed, which would reduce it by a few hundred. Other suggestions include adding a switch in Template:Category diffuse to only add the category once there are a certain number of pages that need to be sorted into sub categories. Thoughts? --DannyS712 (talk) 03:42, 7 April 2019 (UTC)

• Its certainly not much use as it is, so I'm in favour of both suggestions. Rathfelder (talk) 14:28, 7 April 2019 (UTC)
  • I suggest keeping it, for all categories that display {{category diffuse}}, but adding an additional maintenance category e.g. Category:Possibly overpopulated categories if the number of directly-held articles exceeds a certain number. – Fayenatic London 16:25, 9 April 2019 (UTC)

I don't work with cats much but is there a quick way to categorize or segregate pages into one cat that are in Category:All portals but not in Category:Miscellaneous pages for deletion? It would need to be something dynamic and automated because no one wants to manually tag all these pages. Both are automatically populated but with over 1/3 of the namespace at MFD it is getting harder to identify pages that should be checked. Legacypac (talk) 07:58, 10 April 2019 (UTC)

WP:AWB might do the trick. It includes a tool for comparing lists - e.g. of pages in specified categories - and then allows easy application of edits (including add, remove or replace category) to the resultant list of pages. Mitch Ames (talk) 09:09, 10 April 2019 (UTC)

I nominated Category:Films produced by B. F. Zeidman for deletion as a test case (Wikipedia:Categories for discussion/Log/2019 March 17). It was declined "without prejudice against a fresh wider nomination". I contend that, with a very few exceptions (e.g. Val Lewton), producers don't leave much of an imprint on the films they work on, and thus the vast majority of these categories are WP:NONDEFINING. How do I make a "wider nomination" without manually adding literally hundreds of entries to a mass Afd? Clarityfiend (talk) 19:22, 8 April 2019 (UTC)

It was unfortunate that you picked a case with no WikiProject banner on the talk page, so no alerts were generated.

In this case I suggest you start an RfC on the principle. WT:FILM would probably be the best place for it.

If a consensus emerges there to delete some or all categories, WP:Bot requests is then a good place to ask for help with a mass nomination. – Fayenatic London 11:07, 9 April 2019 (UTC)

@Clarityfiend: Let me get this straight... Category:Films produced by B. F. Zeidman contains 21 articles, and you want to send each of those 21 articles to Afd? That is, you wish the whole of each article to be removed from Wikipedia on the grounds that it is about a film that was produced by B. F. Zeidman? I really don't think that will fly. Please note that sending a category to WP:CFD (which is what Wikipedia:Categories for discussion/Log/2019 March 17#Category:Films produced by B. F. Zeidman was about), if successful, results in the removal of articles from that category, followed by deletion of the category page. The articles themselves are not deleted, they remain largely intact save for an edit like this which was a consequence of this CfD. --Redrose64 🌹 (talk) 20:40, 9 April 2019 (UTC)

No. I want to delete most of the producer categories, not the films. Clarityfiend (talk) 18:58, 10 April 2019 (UTC)

Then you can't use AFD - categories go to WP:CFD. --Redrose64 🌹 (talk) 19:39, 10 April 2019 (UTC)

I didn't use AfD. Note my second link. Clarityfiend (talk) 19:43, 10 April 2019 (UTC)

In your original post (which remains unchanged above), you wrote without manually adding literally hundreds of entries to a mass Afd. --Redrose64 🌹 (talk) 19:54, 10 April 2019 (UTC)

How do we apply CATDEF? What if a category is not commonly and consistently used by reliable sources to describe something, but it is a type of category that is often used? How does WP:COP fit in: is this a list of categories we should generally use, or that we may use for certain articles when appropriate? Big questions... but, more specifically, input at Talk:Michael_Gove#People_educated_at_Robert_Gordon's_College would be useful. (My view is apparent there!) Bondegezou (talk) 21:46, 4 May 2019 (UTC)

Some categories (like Category:Chess players) should have a CATDEF-type question raised before including each article in it, while for others they shouldn't, and it's it more relevant to raise in deciding whether a category should exist in the first place. Education is considered a defining biographical fact, just as birth/death years are, place of origin, etc., and categories for colleges/universities are standard and uncontroversial. For education levels below that, I believe it's determined case-by-case whether a particular school merits one (correct me if I'm wrong, but I'm not aware of a blanket consensus). And so if you want to question whether a secondary school like Robert Gordon's College should have an alumni category, there's a process for that. But so long as an education category exists and it applies to a particular article's subject, then you probably don't have a good argument for not applying it to that article. We may get into a threshold question like "well he dropped out after the first day" or something like that, but that doesn't appear to be the case here. postdlf (talk) 23:15, 4 May 2019 (UTC)

Thanks for your input. You raise a number of points.

To just tackle the big question for me... I don't understand how this is consistent with what's written in the guidelines. The text at CATDEF doesn't say there are exceptions, but you're saying certain types of category are effectively exempt...? Bondegezou (talk) 23:28, 4 May 2019 (UTC)

As I said, it's more a question of when CATDEF gets applied, at the time of category creation or at the time of category application. What I can tell you beyond that is I am describing longstanding and widespread practice, and it is at best unusual for someone to object to including an article in an education category that indisputably applies factually. So you're reading the guideline differently than most editors, and it definitely isn't a license to litigate endlessly over what are routine or standard categories. postdlf (talk) 23:42, 4 May 2019 (UTC)

A biographical article might say something like "Fred Foobar (born 1978) is an actor. His father was a builder. He attended Smalltown School and Bigcity University. He became an actor ..... (details of acting career) ... He is married with 2 children and enjoys playing chess.".  Categorization (grouping with articles about similar encyclopedic subjects) should place that article in Category:Actors (or a more specific subcategory of that) and nothing else. Wp also has categories for year of birth and year of death/blp, but those are really for use by editors rather than readers (they contain thousands of articles so aren't good for navigation) and could be hidden from readers.  All of the other details in the article are not characteristics that should be categorized.  The current existence of categories for things like schools inevitably leads to confusion and conflict because (1) (as noted above) this means that attended-school categories work in a different way to plays-chess categories and (2) people argue that because there are attended-schools categories there should be categories for other non-defining characteristics (example). If we can't delete attended-school categories etc then we should at least ensure that categorization guidance pages are clearer about the situation. DexDor ^(talk) 06:54, 5 May 2019 (UTC)

I feel practice should match the guidelines, whether that means practice needs a nudge towards the guidelines, or the guidelines need re-writing.

I concur with DexDor: I find WP:CATDEF as written easy to follow and we should treat education categories just like Category:Chess players. We include them when reliable sources commonly and consistently use them to describe the article subject, not merely when they are factually true. That is the ethos of the whole categorisation system: that it is not simply about what is true. Bondegezou (talk) 14:47, 5 May 2019 (UTC)

My feeling on this (as noted at the original discussion) is very much in line with Postdlf: we need to bear in mind that including these categories is a longstanding and uncontroversial practice. (In terms of how widespread it is, ~30% of BLPs have one or more educated-at categories). It seems reasonable to me that the community's consensus and preferences are best expressed in what's done in practice, rather than in a certain interpretation in principle of the guidelines.

If the two conflict, then, we should probably think first of all about clarifying the wording - but if we do decide it's best to go with the guidelines as written, we certainly need a larger RFC so the community can say, yes, actually, we do think this big change to common practice is a good idea. (Or not, as the case may be)

Not sure what would be a good way to do such a clarification, though; I'm not intimately familiar with the categorisation policy pages. Maybe simply expanding what's explicitly listed as "standard biographical details" on WP:CATDEF? Andrew Gray (talk) 16:24, 6 May 2019 (UTC)

I thought here would be the best place for a broader discussion. I don't want to push anything -- if there's further discussion and most people think there's no problem here, fine -- but if there's continued uncertainty, we could move to an RfC here, maybe with a revised wording to WP:CATDEF (either loosening it along the lines Andrew Gray suggests, or alternatively a tighter wording) or possibly adding clarification instead at WP:COP.

I remain of the view that CATDEF works well as it is, and is relatively easy to apply. While educated-at categories are sometimes appropriate (more so for higher education than secondary), I would be very happy to see that estimated 30% cut back significantly. I do not see how Wikipedia benefits from most categories being defining, and then a few more not being so. I note that it has always been the way that categorisation ebbs and flows, that there's a tendency for categories to accumulate and for over-categorisation to arise as a problem. As I understand it, the community has decided that categorisation is only for a limited set of defining matters and that it is not to be used as a broad ontology, but tensions over that are apparent in existing rules (e.g. WP:SMALLCAT).

If I may @BrownHairedGirl:, who had some interesting observations at Wikipedia:Categories_for_discussion/Log/2019_April_8#Category:Members_of_the_Fabian_Society_Executive_Committee that may be of relevance here. Bondegezou (talk) 17:46, 6 May 2019 (UTC)

Some additional background (so people don't have to search down prior discussions)... I have removed a number of categories from articles, chiefly UK politicians, which are about where the person was educated. E.g. I removed Category:People educated at Robert Gordon's College from Michael Gove. My argument for this is that these categories clearly fail WP:CATDEF. CATDEF says, "A central concept used in categorizing articles is that of the defining characteristics of a subject of the article. A defining characteristic is one that reliable sources commonly and consistently define the subject as having—such as nationality or notable profession (in the case of people), type of location or region (in the case of places), etc." With someone like Michael Gove, it seems clear that reliable sources commonly and consistently define him as a British Conservative politician or as a Cabinet member, &c., but they do not describe him as a Robert Gordon's College alumnus.

Some editors disagreed. They note that lots of people articles have these educated-at categories. It is a relatively common practice. Others have asked why I don't take the category to a CfD, to which I respond that I am not currently disputing that this category may be defining for some people: I'm just saying it's not defining for Michael Gove. Bondegezou (talk) 17:53, 6 May 2019 (UTC)

Thanks for the ping, @Bondegezou.

I regard place-of-education categories as as "standard biographical details" which should be applied in all cases, subject as ever to WP:CATVER. The degree of definingness obviously varies, but the categories are useful as a complete set. I would support changing the guidelines to clarify this. --BrownHairedGirl (talk) • (contribs) 18:29, 6 May 2019 (UTC)

I concur with BHG. We categorise by definingness and also "standard biographical details" (which would be included in any competent 200-word obituary). Year of birth, death, where from, school, university and a few more. Certainly the guidelines should be changed to clarify this. It's very unsatisfactory to have to agonise over whether Gove was or was not defined by his school, or college, or being adopted, or being from Aberdeen or perhaps Edinburgh. Oculi (talk) 20:30, 6 May 2019 (UTC)

Oculi, you raise a number of issues.

First, what you are suggesting seems to me quite different from what WP:CATDEF currently says. I am glad you concur that CATDEF should be re-written, if the community decides that should be our approach.

Personally, I find CATDEF as written straightforward. You look at what RS say about an article topic and it's normally pretty obvious what the defining categories are. For Michael Gove, it's that he's a British Conservative politician. It's clearly not that he's, say, a jogger or that he was educated at a particular school. There are some grey areas, yes (e.g. being adopted), but they can be discussed. I don't believe that your suggestion (defining categories + a specified subset) particularly helps: yes, certain details would automatically be in, but we'd still be using the same principles for everything else. We'd still have the same uncertainty over whether being adopted is a defining category for Gove.

I am skeptical of the "obituary" test. Obituaries normally say whether someone was married (and whether their spouse survived them), how many kids they had, and what they died of. The former are not currently treated as defining categories on any article I can think of, and the last (what they died of) is only occasionally noted. I would say that an obituary is one sort of RS about a person, but one should look at the range of RS.

More broadly, we would then have CATDEF for articles not about people, and CATDEF + a pre-specified subset for articles about people. I don't see the justification for that. It just seems like unnecessary creeping over-categorisation to me. Bondegezou (talk) 09:39, 7 May 2019 (UTC)

You’re just not reading CATDEF the way most editors approach it. You should also read WP:OC, which says that “Definingness is the test that is used to determine if a category should be created for a particular attribute of a topic.” (emphasis added) So as I said above, we don’t have to keep asking that question for every article once a category exists, and it makes absolutely no sense to do that in the context of education categories. Really even in the context of a category like one for chess players, it’s better understood as a question of inclusion criteria (defining what is meant by the category to determine who belongs in it) rather than asking the “definingness” question for every article. Maybe that will resolve your personal dilemma here, but regardless you have no basis for removing applicable and valid categories from articles. postdlf (talk) 12:47, 7 May 2019 (UTC)

I agree with postdlf. E.g. if you find a category that probably isn't defining for anyone (e.g. "People born in June") then you shouldn't remove articles from that category manually, but instead consider taking the category to CFD for deletion. DexDor ^(talk) 19:19, 7 May 2019 (UTC)

If a category isn't defining for anyone, then we take it to CfD, yes.

However, what if a category is defining for some people, but not others? WP:CATDEF is not about whether a category should exist: it is written in terms of (to quote) "categorizing articles". And we do apply a definingness test when considering the likes of Category:Chess players. I find Postdlf's interpretation inconsistent with practice and inconsistent with the current wording.

Nor is there any indication in the guidelines that education categories should be treated differently from any other categorisation. That's an idea that seems to have grown up in some editors' actions, so either we should codify it, or encourage editors to be more restrained in this area. Bondegezou (talk) 11:16, 9 May 2019 (UTC)

Postdlf refers to WP:OC, but that says (immediately before the section Postdlf quotes), "Categorization by non-defining characteristics should be avoided. It is sometimes difficult to know whether or not a particular characteristic is "defining" for any given topic, and there is no one definition that can apply to all situations. However, the following suggestions or rules-of-thumb may be helpful:" That phrasing clearly refers to the question of whether a specific category applies to an article.

I also note that it goes on, "if the characteristic would not be appropriate to mention in the lead portion of an article, it is probably not defining". Yet we have all these education categories that are not mentioned in ledes (and shouldn't be mentioned there). So, why isn't the solution to remove education categories from articles where they are not defining? Bondegezou (talk) 11:50, 9 May 2019 (UTC)

I had a delve into the archives to see if discussion around the time the current guideline wording was agreed could shed further light on the matter. The current WP:CATDEF wording came from Uniplex in this edit on 29 Sep 2011, following discussion at Wikipedia_talk:Categorization/Archive_14#proposing_adding_to_categorize_by_defining_characteristics and at Wikipedia_talk:Overcategorization/Archive_9#Non-defining_vs._trivial_clarification_proposed. The word "defining" had come earlier: it's in the very first draft of WP:OC (23 Nov 2006), with this (broader?) phrasing: "If you could easily leave something out of a biography, it is not a defining characteristic." Back in May 2006, this guideline had a simpler formulation, including, "An article will often be in several categories. Restraint should be used as categories become less effective the more there are on any given article." Bondegezou (talk) 12:07, 9 May 2019 (UTC)

"So, why isn't the solution to remove education categories from articles where they are not defining?" Because no one else thinks that's a good idea, or even agrees that there is a problem that needs a "solution". If you have some suggestions to make for how to clarify the guidelines so you are no longer tempted to read them as somehow forbidding or restricting education categories (the existence and application of which are supported by longstanding and widespread consensus), please let us know, but we're definitely not emptying them out. Thanks, postdlf (talk) 15:44, 9 May 2019 (UTC)

I would like to know if all wikipedia titles are assigned to at least one semantic category (e.g., proteins, surgical procedures). If not how to find wikipedia titles that do not have any semantic category?

Emijenne (talk) 15:55, 17 May 2019 (UTC)

All Wikipedia articles should have at least one category applied. What's a "semantic" category? postdlf (talk) 19:07, 17 May 2019 (UTC)

Category:All uncategorized pages lists the articles that are not in any (other) category. Mitch Ames (talk) 23:27, 17 May 2019 (UTC)

Only if it's tagged {{uncategorised}} - it's not automatic. --Redrose64 🌹 (talk) 11:54, 18 May 2019 (UTC)

I am seeking approval for a bot to bypass double category redirects. Your comments are welcome at Wikipedia:Bots/Requests for approval/JJMC89 bot 17. — JJMC89 (T·C) 06:28, 9 June 2019 (UTC)

Additional opinions would be welcome at Talk:Outline of Esperanto. -- Beland (talk) 05:42, 18 June 2019 (UTC)

More eyeballs requested over here: WP:Categories for discussion/Log/2019 June 27#Category:Works by Bella Thorne and here: WP:Categories for discussion/Log/2019 July 3#Category:Works by Jimmy Somerville where there is a particularly WP:LAME edit war is kicking off. Mostly between two editors, who are doing a bizarre sort of opposed tag-team deletion, tagging the opposite categories.

Jimmy Somerville is a clearly notable musician with an extensive career and back-cataglogue. Clearly we should represent them here, and through categorization, but how? We have the following:

• Category:Jimmy Somerville

• Category:Works by Jimmy Somerville

• Category:Jimmy Somerville albums
• Category:Jimmy Somerville songs
• Category:Songs written by Jimmy Somerville
• Jimmy Somerville discography

This is three levels of categorization, which many would see as too many – certainly for this few members. As I read WP:OCEPON, we should have "Works by ..." but not "<Artist>", unless we have more content needing it. The albums / songs split is perhaps a little verbose, but if we have that (it seems justified for a case this size) then we should still have "Works by ..." and the artist page should be in that.

There are half-a-dozen similar artists all listed here, and we need clarity in our general guidance first, not recurrent edit-warring item by item. Andy Dingley (talk) 18:47, 3 July 2019 (UTC)

If both Category:Jimmy Somerville and Category:Works by Jimmy Somerville are deleted (I see no policy reason for this, but it's being advocated at CfD), where should Jimmy Somerville be categorized? Should it be placed into both Category:Jimmy Somerville albums and Category:Jimmy Somerville songs (and potentially more than that), or else in neither, and thus disconnected altogether from this category tree? Andy Dingley (talk) 17:26, 4 July 2019 (UTC)

Hey, regarding WP:SORTKEY where it says Hyphens, apostrophes and periods/full stops are the only punctuation marks that should be kept in sort values. The only exception is the apostrophe in names beginning with O', which should be removed. For example, Eugene O'Neill is sorted {{DEFAULTSORT:Oneill, Eugene}}. All other punctuation marks should be removed. - if only those 3 punctuation marks are kept, what happens with titles such as these: ! (Donnie Vie album), "@", ( ) (album), ? (XXXTentacion album), [[/ (book)]] (valid redirect link, but breaks text here) and also ...Baby One More Time (song). The guideline was also quite on other signs such as these: @ !*, $ (Mark Sultan album), ^ (math), * (arithmetic), ÷ (album), & (album), ~ (iamthemorning album), ¿Dónde Está Santa Claus?. Would appreciate any help here. --Gonnym (talk) 21:39, 4 July 2019 (UTC)

I found a disambiguation page in an article category, and wanted to look up the rules about it. But I found it difficult, since there is no mention in this article, nor in Wikipedia:FAQ/Categorization nor Help:Category. The relevant info is at the very end of the Disambiguation article, at Wikipedia:Disambiguation#Categories (WP:DBC). I think it would be helpful to have a pointer to that here. I thought about putting this:

===Disambiguation pages===
{{see|Wikipedia:Disambiguation#Categories}}
Disambiguation pages should not normally be placed in article content categories, but in disambiguation categories only.

between the "Articles" and the "Files/images" sections, but I wasn't sure if that would be the best approach... --IamNotU (talk) 19:28, 9 July 2019 (UTC)

What's the best practice for the "by century" cats for people who span centuries? If a person was born in the 18th century and died in the 19th, do you put them in both Category:18th-century foos and Category:19th-century foos? Or just pick one? -- RoySmith (talk) 19:42, 20 July 2019 (UTC)

Usually in both, as long as they were active in both of them. Though in categories by occupation, such as Category:18th-century writers, they should only be added to the century in which they were active in their field. Dimadick (talk) 19:45, 20 July 2019 (UTC)

Why are people allowed to create orphan categories? See Wikipedia:Database reports/Uncategorized categories There are thousands of them.Rathfelder (talk) 10:52, 21 July 2019 (UTC)

After I cleaned up regular articles, only three articles are still using <categorytree> tags, all outlines: Outline of German language, Outline of Esperanto, and Outline of Korean language. Normally I would expect outlines to have their own content independent from categories. Klarst, who has edited all three articles, has objected on Talk:Outline of Esperanto, saying that they are useful, but it's unclear to me why or whether these particular outlines are special. How do other editors feel about this? (This tag causes the listing of pages from the category to be transcluded into the outline.) -- Beland (talk) 07:18, 31 July 2019 (UTC)

Documentation is at mw:Extension:CategoryTree. --Redrose64 🌹 (talk) 07:23, 31 July 2019 (UTC)

This is not a request for how to use the tag, the question is whether or not they should be used. -- Beland (talk) 06:46, 5 August 2019 (UTC)

Were these "not always baseball players from Foo is Fooian" edits legitimate? Even the category explanation on most of those categories says following: "This category is for Fooian baseball players who currently play or have played in Major League Baseball." 85.76.163.182 (talk) 19:30, 3 August 2019 (UTC)

They are not, no.

There are various ways to define "Japanese" or "Welsh" or "Peruvian" or whatever. One is by ethnicity, one is by citizenship, one is by geographic origin, one is by residence, and there are several others.

And different editors have different opinions. But a very usual -- maybe the most usual -- is geographic origin. Lets say you were born and raised in Burma, but your parents were French, French is your first language, and you aren't a Burmese citizen. You would be described as Burmese. Probably, to assist the reader since it's a bit complicated, as "A Burmese artist of French ethnicity", or as French-Burmese.

I mean, what else? We can hardly call you "French" if you've never even been there. What your citizenship is is useful info, but not key to defining you. (IMO; other editors might disagree.)

So, then, if you are "from" Japan -- you were raised there, say -- then you ARE Japanese. So the editors position that not all people from Japan are Japanese, is not correct. (Debatable cases can be discussed at that individual's article; but the presence of one or two rare exceptions at the margins is not sufficient to decouple entire categories.)

The edits should be rolled back and the editor engaged. Perhaps she has compelling arguments, and can present them here. Herostratus (talk) 21:21, 3 August 2019 (UTC)

This may be an argument for more categories to use "from Foo" instead of "Fooish", for clarity and consistency. -- Beland (talk) 06:49, 5 August 2019 (UTC)

There is no easy answer to this kind of problem. To say a person is "From" somewhere can also be taken to mean that they are no longer there. And as far as nationality goes - the law is complex and varies from one country to another. Not everyone born in a country takes on that nationality. It's very rare to find anything explicit in a biographical article about the nationality of the subject. Rathfelder (talk) 09:43, 5 August 2019 (UTC)

Yes, there is no easy answer. However, here we are concerned with, and only with, how people are categorized.

Assuming that we are not going to stop categorizing people by country (we're not) and that we're not going to try to push thru some strict rubric for assigning a person to a country, where they have to meet several criteria (we're not), we have to look at at the totality of the person, but generally where they were raised determines their country for categorization purposes. I believe that is how it is most often done. That is why I say that the edits OP mentioned are not legit.

Sometimes people are raised in two countries, or several. Sometimes there are other complicating factors. Son of an American diplomat, raised in Ghana and Madagascar and Libya and Thailand -- what is he? If you move to Japan at age 1, you are raised in Japan. If you move there at age 20, you're not. What if you moved there at age 9, tho? And so on.

If the person making the disputed edits had picked out specific cases, that'd be different. "This particular guy should not be considered 'Japanese' because [reason]", that's fine. But that's not what the editor is doing. He's saying "Not all Baseball players from Japan are Japanese, therefore we can't categorize any baseball player as Japanese (or American, or anything else)". Extrapolated from baseball players to other vocations, that would be an extremely radical position. It would be a huge, huge change to categorization of people, and it's not going to happen. Since it's not going to happen, doing it where the editor did is not legit. Let him first run a CENT RfC and see if he can get this radical change put thru, and good luck with that.

(N.B. Also "from" certainly does not imply removal from the place. "Where ya from?" "Oh, from here, lived here all my life." And law doesn't matter. We're trying to help the reader find other people from the same mileu. Suppose a person was born and raised in Japan, is of Japanese ethnicity, speaks Japanese as their native tongue, writes in Japanese, lives in Japan and in fact has never been outside Japan, and forth. Suppose this person changes their citizenship registration from Japanese to Israeli (let's assume this is possible). Would we remove this person from the category of Japanese writers, so that the reader would not be able to find him in that category? Why? How would this help the reader? Once you avow that this is a valid point, it's just a matter of discussion where the margins are. What if he goes to live in Israel? What if he's lived in Israel for thirty years and writes in Hebrew now? And so forth. (N.B. many people are categorized under two nationalities and more, and this would be an example of where that'd work.)

TL;DR: people are from where they are raised. People who are from X are Xish (Xian, whatever) for the purposes of categorization. Individuals may be opted out of this general rule on a case-by-case basis. Herostratus (talk) 16:31, 6 August 2019 (UTC)

Can we get some opinions at Talk:Empathy/Archive 3#Removed category:autism? A permalink or the discussion is here. Flyer22 Reborn (talk) 22:05, 9 August 2019 (UTC)

I'm surprised by the dispute over the inclusion of some cats on this article and would appreciate some more eyes and input. See Talk:The_Americans#Spy_thriller/drama_category_dispute. Thanks. --В²C ☎ 20:32, 22 August 2019 (UTC)