Wikipedia talk:Categorizing articles about people/Ethnicity, gender, religion and sexuality/Archive 8

WikiProjects again
We're having another spate of people who can't quite believe that an article about a non-LGBT person could be supported by editors interested in gay icons, or that a WikiProject named "this genre" could be interested in helping with an article about a closely related genre. The LGBT battles routinely cite BLPCAT and this page as justification for breaking the WikiProject bots (as if even hinting that "those people" are interested in the subject matter is terribly scandalous), and we use LGBT as an example in the lead, so I have boldly added an explanation about the difference between article cats and project cats using that example. WhatamIdoing (talk) 16:52, 29 August 2012 (UTC)

New RfC about Categorization of persons
Please see Requests for comment/Categorization of persons: "Should we categorize people according to genetic and cultural heritage, faith, or sexual orientation? If so, what are our criteria for deciding an identity?" Thank you, IZAK (talk) 02:32, 30 August 2012 (UTC)

Actors/actresses
There's an RFC here on the use of "actors" versus "actresses" in categorisation: Village pump (policy) - I believe this has been discussed intermittently in the past, though I'm not sure when, and interested editors may want to comment there. Andrew Gray (talk) 11:07, 17 October 2012 (UTC)

"As another example, separate categories for actors and actresses are not needed, but a female heads of government category is valid as a topic of special encyclopedic interest." What is the support for this? Women are chosen because they are women and men because they are men for a substantial if not overwhelming majority of acting roles. Candidates for heads of governments in Israel, India, Britain, Iceland, or the U. S. run and are elected or not as people, not as women. It seems to me that if we want to make a distinction, it should be the reverse of what is given here. Kdammers (talk) 05:28, 30 October 2012 (UTC)

Discussion at Wikipedia:Categories_for_discussion/Log/2013_April_24#Category:American_women_novelists
You are invited to join the discussion at Categories_for_discussion/Log/2013_April_24. Obi-Wan Kenobi (talk) 04:02, 26 April 2013 (UTC)

Diffusion verses non-diffusion
I would say that in sports where people compete seperately by gender, and in acting, singing, dance and modeling, and possibly among comedians (although I would only really support this view if we had Category:American male comedians), we should diffuse totally by gender. In these cases gender is totally, without question central to a very large part of the person's career. On the other hand, in writing, politics and such gender is less important. Is the fact that Mia Love is female central to her being mayor of Saratoga Springs, Utah or her former and possible present campaign for the US house? Well, she has been mentioned as "the first African-American female mayor in Utah" and as "potentially the first Afircan-American female Republic in the United States house", so clealry her being female is noticed. However, I think we could also safely argue that if she had gone with her first career goal and become a broadway actress and made it big there, being a female would be much more determinative of what she did in her career than it is at present. Terri Hatcher, Amy Adams, Erica Durance, Rolly Bester, Dana Dalany and every other person who has been cast in the role of Lois Lane was cast as such in part because they were female. There are exceptions to casting males as males and females as females, but they are rare. So I think we can work out this system where in these specific places we disperse by gender. I would at the same time argue that we should not disperse by ethnicity at the same level. Dean Cain being of Japanese descent did not prevent casting him as Superman, and Laurence Fishburne was cast as Perry White despite all previous castings being with actors of Euro-American origin. Ethnicity is still arguably a large factor in how the careers of actors develop, and historically was an even bigger facotr (and some might point out that Cain "does not really look very Japanese"), but it is not as controlling a factor as ethnicity and so I do not think we should disperse on it in these cases. On the other hand I think we should disperse Category:American religious writers into specific religious sub-cats.John Pack Lambert (talk) 03:05, 30 April 2013 (UTC)
 * Hi John - I can see your points, but my fear is, the more exceptions we add, the harder this will be for editors (and NY times bloggers) to understand. We already have a guidance that makes an exception to the general rule of diffusing categories and makes all G/E/S/R categories non-diffusing) - the quiz above points out how hard it is to actually implement this in practice. If you now have a set of special exceptions to the non-diffusing rule, this will make things even messier...--Obi-Wan Kenobi (talk) 04:58, 30 April 2013 (UTC)


 * Comment My gut suspicion is that if we should just diffuse everything, but at the same time in most cases work against the creation of bottom rung ERGS cats, except in cases like acting and religious writing where gender or religion are the central way to define the connection of the person to the category.John Pack Lambert (talk) 16:43, 30 April 2013 (UTC)

Heck this guideline is a mess
I've been rereading this to try to determine just what is the standing guidance on subcategorisation but this is rather confused, messy and either the examples are outdated or else no-one's propely diffused them. Breaking down the final two paragraphs:


 * Whenever possible, a valid occupational subcategory should be structured and filed in such a way as to avoid "ghettoizing" people, but at the same time, Wikipedia rules about redundant categorization should also be respected. It is entirely possible to meet both of these expectations simultaneously; if you can't, consider alternative ways of defining the category.

"Consider" is one of the most ineffective words we have in guidelines.


 * For instance, if you cannot create "Gay politicians from Germany" without ghettoizing people from Category:German politicians, then it may be more appropriate to eliminate the more specific category and simply retain Category:Gay politicians and Category:German politicians as two distinct categories, or to refile people from the parent category into more specific subcategories based on the particular political body their career is associated with (e.g. "Members of the German Bundestag", "Chancellors of Germany", "German Bundesland presidents" or "Mayors of Berlin").

Well we have Category:LGBT politicians from Germany‎ along with a whole host of other subcategories by office, "state" (sic), former country, gender, ethnicity, party, type of death and so forth. We also have 333 entries in the main Category:German politicians - is this a backlog of unsorted entries or cases that are impossible to subcategorise? And without checking every entry in Category:LGBT politicians from Germany (or Category:German women in politics or Category:Jewish German politicians), I'm not sure if all those entries have some other ones added as well.

And of course politicians are usually an easy group to divide up but sometimes you get awkward cases, mainly people who have big political influence but don't actually operate through either parties or elections. Does one awkward case invalidate an entire set of sub-categories? And there's nothing in this guidance about what happens if the only existing sub-categories that fit said awkward case are EGRS ones.

Then of course you get groups where subcategorisation is harder - the current example being novelists where not every novel fits clear genres and some book shops or libraries are organised alphabetically but others are organised by genre and some are organised with breakouts for some genres (e.g. romance novels), some formats (e.g. classic editions or graphic novels) and a catch-all general section.


 * If a category is not otherwise dividable into more specific groupings, then do not create an E/G/R/S subcategory. For instance: if Category:American poets is not realistically dividable on other grounds, then do not create a subcategory for "African-American poets", as this will only serve to isolate these poets from the main category. Instead, simply apply "African-American writers" (presuming Category:Writers is the parent of Category:Poets) and "American poets" as two distinct categories.

This doesn't really address cases where a category is partially dividable.

Further up, and the distination of WP:Cat gender, we have:


 * As another example, a female heads of government category is valid as a topic of special encyclopedic interest, though it does not need to be balanced directly against a "Male heads of government" category, as historically the vast majority of political leaders have been male. Both male and female heads of government should continue to be filed in the appropriate gender-neutral role category (e.g. Presidents, Monarchs, Prime Ministers, Governors General). Do not create separate categories for male and female occupants of the same position, such as "Male Prime Ministers of the United Kingdom" vs. "Female Prime Ministers of the United Kingdom".

The problem with this example is that nobody ever intended to put Margaret Thatcher, Kim Campbell, Jenny Shipley, Angela Merkel and Julia Gillard, or for that matter John Major, Brian Mulroney, Robert Muldoon, Helmut Kohl and John Howard, in at the level of a catch-all Category:Heads of government for every HoG in human history. This doesn't cover very well cases where there are a lot of entries in the head category.

The more I read this the more I think it's impossible to say clearly whether anyone has or hasn't followed guidance that doesn't cover a messy situation well. We can either clean up or replace the individual examples but we need a much stronger and clearer guideline about EGRS subcategories and exactly how they should be handled in partially dividable cases. Timrollpickering (talk) 16:15, 30 April 2013 (UTC)
 * My general view is there are two sets of by gender categories. One if the heads of goverment case, where it makes sense to have a general female one because in other cases we have dispersed by more specific criteria, everyone is the head of government in a specific place.  However with sports, acting, modeling and singing I would say it makes sense to disperse by gender nearly to the lowest level.  In singing we have categories like Category:American tenors which are more specific sub-cats of Category:American male singers.  The tricky place is things like writers, and I have come to the view that probably we should not divide Category:American women writers at all, or at msot divide into fiction and non-fiction sub-cats.John Pack Lambert (talk) 17:15, 30 April 2013 (UTC)

Even as the person who originally wrote most of the original version of this guideline back in 2006, I absolutely agree that it's not good enough anymore and needs significant updating to reflect the state of things in 2013. My original work was a starting point based on the situation as it stood at the time, and was always meant to be a work in progress that would get updated and changed as our consensus and our clarity about what's appropriate or inappropriate at the level of categorization improved and evolved — but this document absolutely has not been keeping up adequately with that evolution. Part of the problem is that the document hasn't always been updated to keep up with the actual state of current consensus — and sometimes when it has, other users have challenged or reverted or editwarred or watered down those changes. And another part is that when the document was originally written, a lot more of it was based in theory than in actual evidence of how such categories play out in actual practice, because the schema of identity-label categorization wasn't nearly as developed as it is now: people hadn't tried things like yet, people hadn't tried comprehensively filtering every LGBT category on Wikipedia into individual L/G/B/T quadrants, people hadn't actually battled over conflicting interpretations of what's allowed or disallowed by individual sections of this document as extensively as they have since, people hadn't actually tested and exposed the limits and the flaws and the contradictions yet, and on and so forth.

I would absolutely favour, and contribute to, a process to review and revise this document to comprehensively identify and correct its flaws — the places where it's not clear enough anymore, the places where consensus has evolved past what's written here, the places where it just plain needs to be redone from scratch. Bearcat (talk) 18:20, 30 April 2013 (UTC)
 * Comment Do we have any guidelines on dividing ERGS categories by other ERGS criteria. For example, Category:American women poets only has by ethnicity and one by place sub-divison.John Pack Lambert (talk) 19:48, 30 April 2013 (UTC)
 * Comment Another issue that I am not sure we have faced up to is even if Category:American women novelists works, it seems less likely that we would need Category:Nigerian women novelists, since we have such a smaller group of Nigerian novelists.John Pack Lambert (talk) 19:49, 30 April 2013 (UTC)
 * Thanks Bearcat - I'm with you, I'd love to help refine this guideline, it is not very good. My suggestion is that we start from a practical place - take the quiz I made above, or even James Baldwin or Maya Angelou, and carefully go over these bios and figure out where they should fit in an ideal world - and try to come up with a set of rules that we could then apply to a new bio that would get us to the same place - in other words don't start with generic guidance, start with a real world example, solve it, and then see what can be gleaned from so solving. Working on Winona LaDuke was extremely useful to me, and gave me lots of ideas on an algorithm. So why not start by taking the quiz! --Obi-Wan Kenobi (talk) 21:45, 30 April 2013 (UTC)

Cat header template
This one seems like a quick and obvious win - what if we created a template, that would go at the top of any E/G/R/S category - that would say something like "Pages in this category are for XXX. They should also be placed in categories X, Y, and Z". Any thoughts? it could be a replacement for Distinguished subcategory, that is specific for EGRS.--Obi-Wan Kenobi (talk) 23:49, 6 May 2013 (UTC)

Question on nurses
I did a bit of re-organization of the tree. I note that currently the tree is ungendered, but there is a subcat of as presumably this is a topic of special consideration. However, we also have, which I nominated to make ungendered but this was resisted at CFD, so we created a non-gendered parent instead, with a matching. So now my question is, what should be done with all of the people in, which is now a sibling instead of a child of. My gut is, they should be also placed into either or male accordingly, since this sub-tree is now fully divided by gender. Thoughts welcome. --Obi-Wan Kenobi (talk) 16:16, 15 May 2013 (UTC)

Conflicting guidelines
Under categorization it says "Subcategories defined by ethnicity and sexuality are often non-diffusing subcategories. See also the gender, race and sexuality categorization guideline." I have two major issues with that a-it does not mention gender at all, b-we do not categorize by race. What the difference between race and ethnicity is is not always clear, but we should be clear categories need to be based on and implemented on ethnicity, not race. We should not group people from clearly distinct cultures just because they look similar and may if you go back far enough have the same ancestors.John Pack Lambert (talk) 16:00, 2 May 2013 (UTC)
 * I just fixed that guideline - that was just wrong I think - the page used to be called race a long time ago, and that redirect still exists - but we shouldn't use it, and possibly delete (or cripple?) the redirect to discourage further use?--Obi-Wan Kenobi (talk) 17:49, 2 May 2013 (UTC)

While some people have already called attention to certain issues that we need to discuss and resolve above, I'm going to list a few more here just for reference's sake: These are just raised for contemplation at this time, and I may add more here; active discussion about them should wait until we're actually ready to review the guideline in detail rather than jumping the gun. Bearcat (talk) 17:49, 2 May 2013 (UTC)
 * Status of subnational territories which occupy an uneasy middle ground between "nationality" and "ethnicity" (e.g. Puerto Rico, Greenland, Guam): How should we define what level of categorization is warranted for these people?
 * Filtering of LGBT categories into separate subcategories for each quadrant of the community: while this is fully warranted in some cases (e.g. for writers or politicians), it is not useful or desired in many others (e.g. "LGBT people from Country" categories). How can we better clarify the distinction?
 * Are ethnic-occupational subcategory groupings like, , , etc., really necessary or warranted at all?
 * Hi Bearcat - it's so funny, I was literally about to post a question here about Puerto Rico. I guess we're on the same wavelength. However, a suggestion - swing by here and take a look: Wikipedia_talk:Category_intersection - this is something we could do now, and we wouldn't have to worry about LGBT or quadrants or even whether American writers of Jamaican descent was worth creating - we would just use category intersection for that. my proposal is, we ice all of the gendered/ethnic/sexuality cats (religion have to think about) - except at the highest level - and then use category intersection to build anything beneath. Take a look and comment there - if we can get this working, this whole guideline may be completely different.--Obi-Wan Kenobi (talk) 17:52, 2 May 2013 (UTC)
 * I've really got to caution you to cool those jets a little bit — proving that your system works with two categories over four articles is, for starters, a long way from proving that it scales up well enough to adequately handle thousands of categories distributed over millions of articles. People have been proposing that we rejig the category system along similar lines for years without success, for the record. While I'm not wedded to the idea that we need to keep the system we have right now if there are better ideas out there, it's a pretty big leap from "hey, this works on one test case" to assuming that we can immediately junk the entire category system as it currently exists — for example, can you yet prove that your proposal won't unwittingly create new problems that you haven't adequately anticipated yet? So for the time being we still do need to work within the system as it currently exists, including discussing and repairing the flaws in this document — your proposal is certainly worth examining and exploring, but that doesn't necessarily mean it's actually ready for prime time already. Bearcat (talk) 05:03, 3 May 2013 (UTC)
 * Thanks and I appreciate your caution, and will definitely rely on more assistance and input from you as we move forward - you've been a thoughtful contributor to this challenging space. I probably misspoke above - I did not mean to suggest we would immediately dismantle the category system and thousands of cats - my proposal is to refine the UI of the prototype a bit, and then go live with it in a particular sub tree (say, or  or something), and test out how it works within the context of that tree, and get user feedback. Regular category browsing would remain, but if you ever wanted to have an intersection by gender/ethnicity/sexuality/etc, you would click the links to bring you to a pre-filled set of category intersections. It's at least worth a try, don't you think?--Obi-Wan Kenobi (talk) 05:18, 3 May 2013 (UTC)


 * My general take on the issues. A-Pueto Rico and Guam will work just fine, as long as we make sure the people so categorized actually are clearly "nationals" of that place. There are lots of people who clearly belong in Category:American people of Puerto Rican descent and not in Category:Puerto Rican people (although some people maybe belong in both, I am not sure.  it needs to be decided on a case by case basis.  2-If someone can writer an article American writers of Jamaiacan descent that is more than a list, we can have that category, if not, than no.  The problem is lots of descent categories have cropped up without really being merited.John Pack Lambert (talk) 02:30, 14 May 2013 (UTC)
 * Actually one problem with intersect, is that we have not really agreed to universally categorize by gender, and ethnicity is a complexed mess as well. In the case of gender, we only categorize by it when the overlap of the gender and something else is notable.  We do not put people in Caegory:American women for example.John Pack Lambert (talk) 00:23, 17 May 2013 (UTC)

totally diffused by gender categories
We have at present lots of categories that are 100% diffused by gender. The example that comes to mind the fastest is the models tree. For example Category:Albanian models is a holding cate with two sub-cats Category:Albanian female models and ;Category:Albanian male models. The the articles are in one of the other sub-cats. Considering how modeling works in the real world, this is probably a logical split. The same probably also works for singers (although at least Category:American singers has lots of gender neutral sub-cats as well, although we also have Category:American female pop singer-songwriters, which is one of the most indepth intersect categories we have. It works for actors although while in potential we will have actresses and male actors categories, in reality most places outside of Category:American actors at the top level, we have not fully split out to those two. It also works for dancers, prostitutes (although the size of that category maybe does not justify it being so segmented), and maybe a few other like categories.  It is also in theory how we work a lot of sports categories, although those are a little different because although you can see that Category:BYU Cougars men's basketball players and Category:BYU Cougars women's basketball players are a split by gender, it is really a split by team. I would say these should be limited to a-occupations where gender is not only notable but controling, the people not only are seen based on their gender but that really controls how they do their occupation, and what their occupation is. Gwenerally it should be limited to things like entertainers. For example, with singers, the most specific singer categories in some ways are things like Category:American operatic tenors, and that is clearly by gender.John Pack Lambert (talk) 00:34, 17 May 2013 (UTC)
 * We should mention this issue directly in the guidelines, which we really do not. In fact for a long time we seemed to say you should not have both a male singers and a female singers category.  The guidelines were just ignored, although at one time I did try to do a CfD to merge them, but it failed miserably.John Pack Lambert (talk) 00:34, 17 May 2013 (UTC)
 * I am not sure we even want to apply this to all people who might be called "entertainers". Even though we apply it to singers, I am not sure we want to apply it to any non-voice female musicias.  Category:female guitarists probably should not be exempt to non-diffusing rules for example.John Pack Lambert (talk) 00:36, 17 May 2013 (UTC)
 * Other probably exceptions are comedians, although maybe if we created Category:American male comeidans such a diffusion would work. Also radio and television personalities, and the related sub-cats that are not yet divided by gender should probably be left as is.John Pack Lambert (talk) 00:40, 17 May 2013 (UTC)

Proposing to expand the Ethnicity and race section
If there are no objections I propose to add some text to the Ethnicity and race section. Let me know what you think about it. FonsScientiae (talk) 16:06, 20 July 2013 (UTC)
 * The section is fine as it is, as it is short and to the point: "Ethnic groups are commonly used when categorizing people; however, race is not." Census categories cannot be used for categorization because they frequently overlap with national and "racial" groups. They are also inherently dependent on anonymous self-reporting; a respondent can tick anything s/he likes and there's no way of telling what that may be. In short, racial categorization should be avoided altogether, as it is ultimately subjective. Ethnicity -- as in what ethnic group(s) a person's family hails from; e.g. Hazara people) -- however, is not. Middayexpress (talk) 17:26, 20 July 2013 (UTC)


 * I question the reference to race being used officially for classification purposes in the UK. It is ethnicity (self classified) that is used for census purposes, and this is what is said in the Race and ethnicity in censuses article. I cannot think of an official 'racial' classification being used distinct from ethnicity - insofar as they might be understood as different then 'race or ethnicity' are covered under a single ethnicity label. Are the other examples correct and how confident are you that they are the only ones? I agree with Middayexpress anyway that the passage is fine as it is. --AJHingston (talk) 22:18, 20 July 2013 (UTC)

Help, please with LGBT Categories
Okay, here's what is going on. I've been working on cleaning up categories, creating, merging and deleting. I came across the parent category Category:LGBT actors which, mostly, consists of subcategories Category:Gay actors, Category: Lesbian actors, Category: Bisexual actors, etc.. There are a small group of individuals (say, a dozen) people who are in Category:LGBT actors rather than in a subcategory.

I was going through and recategorizing those handful of individuals into their applicable subcategory (determined by reading the article on the person) and I'm being reversed on nearly every change. Is there a reason why, for example, a self-identified gay actor should be in Category:LGBT actors rather than Category:Gay actors? Because, if there is any question about whether an actor is gay or not, they shouldn't even be classified in Category:LGBT actors, should they? I haven't been challenged on any of my other recategorizations on other topics so I'm a bit mystified. I guess I should add that I was an IP account the past year and just registered this username today.

I should say that I have no "stake" in how a person is categorized, I'm more concerned with having an orderly structure to how the categories are organized more than anything else (logical parent-child arrangements). But in this instance, if anyone can fill me in on why I would be reversed, I'd like to know. If there is something I'm not getting about WP:EGRS, I'd like to know now to avoid these reverts in the future. Thanks! Newjerseyliz (talk) 21:43, 25 July 2013 (UTC)
 * Be completely honest. I was reverting you because you were adding people, dead and alive, to a specific sexual orientation category who have not identified by a specific sexual orientation. That's also why I pointed you to the section immediately above this one. If the person is alive and is in the LGBT category, it's because that person has identified as being a part of the LGBT community (such as saying "I came out for myself and to represent LGBT visibility"), but has not disclosed a specific LGBT sexual orientation. I didn't revert you on this, for example, because the sourced Personal life section says he is openly gay. However, that matter needs to be looked over more closely because if that actor has never publicly identified as gay, he should not be in the gay actors category. If the person is dead and is in the LGBT category, then it is because there are conflicting sources about the person's specific sexual orientation or there are no sources affirming any specific sexual orientation...rather just behavior or claims about their sexuality and/or sexual orientation...or because the terms lesbian, gay, bisexual and transgender and their conceptualization did not exist at that time period. See what is stated in the section immediately above this one about categorizing historical figures by one of these terms. Flyer22 (talk) 22:03, 25 July 2013 (UTC)


 * Most of the actors you refiled are people who are certainly open about being LGBT, but about whom there is still some ambiguity about which particular letter they identify with. A woman in a same-sex relationship, for instance, could be either "lesbian" or "bisexual", and you have no way of knowing which one she actually is until she actually uses one word or the other in reference to herself. And a man could be either "gay" or "bisexual", with again no way of being sure until he specifically tells you which word applies. Either one of them could reject both labels entirely, choosing instead to identify as "queer". And even a person who's attracted to and dates people of both genders could also reject the label "bisexual" and prefer to call themselves "queer" or "pansexual" or "omnisexual" instead — and some "bisexual" people even choose to identify as "gay" or "lesbian" for political reasons, while still reserving the right to occasionally date someone of the opposite sex anyway. So until they've actually told us which label applies, we can't assume.
 * I'd suggest that you review the above discussion for the reasons behind this. Yes, we would prefer to keep clean, with all relevant people being filed in quadrant-specific subcategories instead of the main one — but for some people it's just not clear which quadrant they belong in, and those people have to be left in the main category until we can properly source which particular quadrant they actually identify themselves with. We don't make assumptions about people's preferred identity labels based on their behaviour; we wait until they've specifically used one of those labels for themselves. Bearcat (talk) 22:14, 25 July 2013 (UTC)
 * Then, if this is the case, maybe we should, as I suggest, remove the LGBT tag completely from those profiles. I was "sorting" articles into "child" categories but if there is some question about whether they should be in the "parent" category, then the question is moot. They should be untagged and LGBT removed as an identifier. Newjerseyliz (talk) 22:15, 25 July 2013 (UTC)
 * Bearcat, the above comment was directed at Flyer22...you and I were posting at the same time. I'll have to ponder your words a bit more. As for now, I'll just stay away from dealing with LGBT categories and work on more neutral ones. There are plenty of other areas that need work.
 * But, as I posted above, I kept coming across LGBT tags when reading the biographies of early 20th century writers, actors and entertainers and that got me interested in tidying up categories. Newjerseyliz (talk) 22:17, 25 July 2013 (UTC)
 * I disagree with removing all LGBT categorization for them if they have identified as being a part of the LGBT community. Or if a dead historical figure (a very old one or more modern one), as in the case of James Dean, there are a lot of claims and/or speculation in reliable sources about that person being lesbian, gay or bisexual (the only LGBT category Dean is currently placed in is Category:LGBT entertainers from the United States). Furthermore, not having the LGBT categorization causes editors to label these people by a specific sexual orientation more so than if it were not there. And at least a few of these LGBT listings have WP:Consensus at their respective talk pages. The Amber Heard matter even gained its consensus at the WP:BLP noticeboard (as currently seen on its talk page). That stated, similar to what I mentioned with regard to the Amber Heard case, I'm not hard-pressed on keeping people in the LGBT category as long as we don't identify a specific sexual orientation for them when they have not publicly identified by one. Flyer22 (talk) 22:35, 25 July 2013 (UTC)

Privileging words over thoughts and deeds: attraction vs identity vs behaviour in sexual orientation
The sexuality category expects living people to come out verifiably unambiguously as lesbian, gay or bisexual before they can be labelled as having a lesbian, gay or bisexual orientation. Given that sexual orientation is more than someone's identity, why are words privileged over their thoughts and deeds? It is difficult to know their attraction without knowing their thoughts, but their behaviour can be verifiably 'LGB' without ever having the identity.

We do not demand that someone verifiably say 'As a man.. ' or 'As a woman..' before assigning them to a gender category, but without that we have no idea as to their personal gender identity. s

The privileging of identity over attraction and behaviour is problematic given how many more people have an LGB attraction and/or behaviour than an LGB identity.

Is the answer to split sexuality into the three categories, and have an LGB Attraction one, an LGB Identity one, and a LGB Behaviour one, rather than a single LGB Sexuality one? Lovingboth (talk) 18:21, 23 July 2013 (UTC)
 * oh my. Can we please *not* go down that route? Also, categories are for things which are DEFINING - e.g. something that would be mentioned in the lede. Gender is one of these things, and people usually conform to one of the gender roles, are open about it, and 3rd party sources discuss it in that fashion (and if they don't identify as man or woman, then we have other categories for them too if they come out - I don't know of any cases where there is someone of ambiguous gender who has not publicly identified as "something" and where sources do not call them man or woman, but perhaps such a case exists?) If someone is openly gay, and says so, then that is usually captured as a category, and such a thing is also usually mentioned in 3rd party sources about them. However, if someone is openly heterosexual (dating lots of actresses and models, for example), the fact that they are hetero is *not* usually mentioned in sources about them - which is why we have LGBT categories but we don't have HETERO categories. As to behavior, this only seems to come into play with long-dead people, where historical consensus is that such and such was gay, even if X never came out and said it. But to label someone currently living as gay just because they were seen entering a gay club with a cute male friend or even because of that one time at camp, no, I don't think so. I think the term "men who have sex with men" was created for this case, because they don't necessarily identify as gay, and we should not so-label them with a category accordingly. If there is sourced information that "X has stated many times he is attracted to men, but does not identify as gay" - then we can add that to the article if it is significant to the reader's understanding and covered by reliable sources - but I don't think a new category tree should be created for "Which genders X is attracted to" and "Which genders X had sex with".--Obi-Wan Kenobi (talk) 18:46, 23 July 2013 (UTC)
 * I am mostly happy not to go down that route (it'd be a pile of work for one thing) if we can agree that identity is not the only part of sexual orientation that matters here. The prompt for this is a long-running - but currently quiet - dispute at Jodie Foster over including her in a category of LGB people (or actors, I can't remember which). She is unambiguously and verifiably out about having been in a relationship with another woman, but there is no apparent verifiable source for her saying 'I am a lesbian'/'I am bisexual'. So she's certainly LGB in attraction and behaviour, but because of the policy, any edit saying she's included in an LGB category is reverted. (Oh, and yes, many people with the behaviour don't have any sexual identity.) Lovingboth (talk) 21:45, 23 July 2013 (UTC)
 * I think the reason this guideline exists is that care should be made in categorizing people into such categories - categorization is different than text in a bio - for example, you can say "Foster is known to be dating X" - but you should not say "Foster is a lesbian" because that's what putting her in a L-category means. Categories are binary, while sexuality is a complex continuum, so in an abundance of caution I think we avoid putting people into a box like that unless they've put themselves there. We don't have such worries about other things, like "Actresses from California" into which we could put people even if they never come out and say "I was born in California" - because there isn't really a social sensitivity around that. For LGBT, there is. For example, I created Stanley_Saitowitz, and found him on a list of known gay architects, but I haven't added him to as I haven't found any sources where Saitowitz himself comes out in that way. Evoking User:Bearcat who may have more to add here as I think he wrote some of this guideline.--Obi-Wan Kenobi (talk) 22:16, 23 July 2013 (UTC)

Obi-Wan is right, we have to be extremely careful about this. For instance, the fact that a female celebrity is in a relationship with a same-sex partner does not, in and of itself, tell you whether she would be more correctly described as "lesbian" or "bisexual" (that's where our problems with categorizing Jodie Foster and Tracy Chapman have always been, as neither one of them has ever quite been explicit enough about whether they're L or B). And we run into even bigger problems with historical figures — Willa Cather, for instance, was clearly not a conventionally heterosexual woman, but because she lived in a much less LGBT-friendly time period it requires speculation to determine whether she would be most properly described as lesbian, as bisexual, as wholly asexual or as a person who, given access to contemporary medical technology, would be pursuing SRS and living as a man (and based on what's actually known about her personal life, you could make a genuine case for all four of those possibilities if you tried hard enough.) Or consider the cases of Alexander Wood and George Herchmer Markland, two 19th-century Canadian politicians who faced allegations of same-sex behaviour, but for whom there is not sufficient biographical evidence on the public record for us to properly determine how true those allegations were or weren't — so while they are filed in an LGBT history category on the basis that the allegations themselves are relevant incidents in Canadian LGBT history, they aren't categorized as being LGBT people because we just don't know whether they really were or not.

And in fact, there are people in the world who identify their fundamental orientation as the opposite of what one would expect based on their public behaviour — there are, for instance, people who identify themselves as gay or lesbian even though they're in opposite-sex relationships (Bill de Blasio's wife, for instance, still identifies herself as a lesbian who happens to have found a male soulmate, rather than as a person whose fundamental orientation has changed just because she fell in love with a man), and people who identify themselves as heterosexual even in the face of significant public allegations to the contrary. It's not our job, however, to make pronouncements about whether Chirlane McCray or Larry Craig should identify their orientation differently than they do — because that's simply not what we're interested in.

I've often pointed this out in cases of dispute, but it bears repeating here: strictly speaking, Wikipedia doesn't care about what's actually going on in a person's private sex life. We're not interested in trying to pin down what sexual activities actually get a person's naughty bits roaring; we're not interested in making pronouncements on whether our article topics are lying to themselves or others by proclaiming an identity that doesn't match their personal behaviour; we're not interested in outing people. We only care about LGBTness insofar as a person is explicitly and openly identified with the LGBT community — which requires that the person make an explicit statement of their identity in the case of a living person, and authoritative published biographical sources in the case of a dead one. What makes LGBTness relevant for our purposes is not necessarily the state of being LGBT in and of itself, but the state of being culturally and socially identified with the LGBT community in reliable sources. Simply put, we're interested in their public identity as a member of the LGBT community, not in their private bedroom behaviour itself — no matter how homosexually aroused Larry Craig's man parts may be, for example, if he doesn't openly acknowledge and associate himself with a publicly gay identity then he isn't "gay" in the sense that we're interested in. (And the same goes for Anderson Cooper, too: behaviourally speaking, he was just as gay the day before he officially came out as he was the day after — but because what concerns us on Wikipedia is the public identity rather than the behaviour, he wasn't "gay" in the sense that's relevant to Wikipedia until he actually put it on the record in a public statement.) Bearcat (talk) 22:56, 23 July 2013 (UTC)
 * Well said. Now you can see why I evoked Bearcat. If you utter his name three times, he comes running :) --Obi-Wan Kenobi (talk) 23:02, 23 July 2013 (UTC)
 * I just read (and closed, just as a formality) the Jodi Foster discussion. One of the arguments was "Well, she came out, and if she came out, she's not straight, and if she's not straight, then she MUST be LGBT." Even this, I don't think is true. LGBT (really, L, G, B, T, Q, I, etc) are specific community and sexuality and gender identities that have a specific meaning at a specific moment in time and a specific place. Being a lesbian in SFO in 2013 is quite different than being one in Victorian England or Ancient Egypt, and we have no idea if "lesbians" from that era would identify with the package of meaning that "lesbians" holds in our era (which is a broader problem with historical classification of people as gay, but I digress). No-one requires you to either be straight or something else - I'm sure there are people who identify as NONE of these things - remember, these are boxes created by our culture, and not everyone fits into the boxes we define. And since we don't have a category for them, c'est la vie. Not every aspect of ourselves is categorizeable.--Obi-Wan Kenobi (talk) 23:09, 23 July 2013 (UTC)
 * Well, for what it's worth, my own view has always been that Foster has been clear enough to justify filing her in general "LGBT" categories, with the only real ambiguity being over whether we could properly diffuse her into lesbian-specific or bisexual-specific subcats instead of the general all-inclusive ones — but that's not a fight I'm willing to invest in right now. Bearcat (talk) 00:33, 24 July 2013 (UTC)
 * Yes - we don't know (and I personally don't care) whether she identifies as bisexual, lesbian, pansexual, queer, or anything else. It also may change from time to time. But she is absolutely in the general category, with a public identity as such, except that the policy says no.


 * If people don't want attraction or behaviour categories and the LGBT categories are actually 'Identifies as..' (present tense), then that's what they should be called. Lovingboth (talk) 09:50, 24 July 2013 (UTC)


 * Reading through these comments I'd tend to agree with Obi-Wan Kenobi/Bearcat. I've always felt one should be extremely careful about applying the LGBT cat, and that it should be applied sparingly (btw - same rule should apply to categories involving race/ethnicity/national identity/religious affiliation). Trying to break the LGBT category down into "identity" and "attraction" seems as though it would beg for more pointless, largely irrelevant debate and discussion about what exactly it means to be gay. No thank you. NickCT (talk) 13:19, 24 July 2013 (UTC)


 * Thinking about it some more, I think it would solve a number of issues. Currently, the same category is used for people who are a) long dead and b) would never, ever have identified as LGB because they pre-date the identity, but had same sex behaviour or wrote about their attraction and for living people who currently say 'I'm L/G/B'... but not living people who verifiably have the history of behaviour. This is inconsistent. When David Bowie dies, it looks like precedence will say that it's ok to label him as LGB because he certainly had bisexual behaviour, even if he didn't always identify as bisexual. but while he's alive, it's not because he currently doesn't.


 * I am very very aware that there are people who think that identity should trump behaviour or attraction, but they tend to have an agenda - typically to minimise the number of people who count as LGB or to maximise the number of L & G people at the expense of the Bs - that I don't agree with. Lovingboth (talk) 15:59, 24 July 2013 (UTC)
 * Categories are a blunt tool, not a scalpel. It's much better to use prose to explain the complexities of sexuality. For someone who is dead, we would only put them as bisexual of consensus of reliable sources said so. Also, you haven't presented any evidence of an agenda - either to minimize or maximize - for example, some people are a bit dismayed at the "jew-tagging" that goes on here - labeling *many* people as Jewish - and similarly, I've seen comments (for example at the Jodi Foster article) that LGBT tagging can go a bit far - so that to me is evidence of overtagging vs undertagging. As NickCT says, it's better for all of these categories to be applied sparingly - we don't need to categorize every last person by their sexuality, we only do so if it is somehow relevant and DEFINING. Indeed, if person X was gay, and told their friends and neighbors, but no reliable sources ever mentioned it, I'm not sure I'd agree with adding the category in that case.--Obi-Wan Kenobi (talk) 16:42, 24 July 2013 (UTC)
 * Concur with Obi-Wan Kenobi. Personally speaking, I do have an agenda. My agenda is combat some of the categorization shenanigans that occurs on WP due to poor/vague categorization policy. NickCT (talk) 17:16, 24 July 2013 (UTC)
 * Well, I have spent the past month or so, reading biographies of NYC/Hollywood actors, writers, directors, entertainers from the period 1890-1950 and if there is even a HINT that a person is LGBT, they've been categorized as being so. No verification required. If some Hollywood tell-all book from 20 years ago says that Actor X used to attend parties thrown by Director X (and that person was "out"), you can bet that Actor X has been tagged Category:Gay actors. For those individuals whose biography had no mention of any same-sex relationships, I removed those tags. But, truthfully, I let a lot of them stand because I didn't know what the criteria were.
 * After reading WP:EGRS and the comments in this thread, I think that some industrious editor could go through the abundance of LGBT and L-G-B-T categories and remove mentions of historical persons whose sexuality was merely ambiguous but unverified because I think there are hundreds (thousands?) of LGBT tags applied to them. Really, it could be a huge project to take on. Newjerseyliz (talk) 22:09, 25 July 2013 (UTC)
 * Concur with User:Newjerseyliz's sentiments here. We have to get aggressive about combatting LGBT overcategorization. NickCT (talk) 16:31, 30 July 2013 (UTC)


 * Well, the thing I think you're missing here is that even if you ignore a person's publicly professed identity and focus solely on their private behaviour, that really doesn't change nearly as much as you seem to think — because by and large, categorizing people on the basis of behaviour relies on (a) unreliable sources, (b) unverifiable assertions of insider knowledge and/or gossip, or (c) speculation. Putting aside David Bowie's shifting public identity and focusing only on his actual sexual behaviour, for instance, still doesn't get him into an LGBT category — because while there are certainly rumours about his past sexual exploits, none of them can be properly sourced as being true rather than just rumoured. Morrissey, for another example, keeps such a lid on his private life that there's really little to nothing we can say about his sexual orientation — although he's pretty obviously not heterosexual, we can't properly source anything whatsoever about his actual behaviour. And then there's John Travolta, who's got rumours flying all over the place, but as of yet we can't satisfactorily prove that the rumours are true — even with the massage lawsuits, as damning as they may look we can't satisfactorily prove that the allegations actually occurred, rather than the accusers making up lies to extort money from him. Until a reliable source weighs in with definitive proof one way or the other, it's simply not our place to make our own decisions about what's true and what isn't.
 * Simply put, it's flatly impossible to properly source what a person's sexual attraction is unless they tell you — and with very few exceptions it's exceedingly difficult to properly source anything about a person's sexual behaviour unless they publicly acknowledge it. That's one of the key reasons why in virtually all cases we rely on a public statement of identity as our baseline for categorizing people as LGBT: because in nearly every case, the person's willingness to speak on the record about their social and cultural identity as an LGBT person is really the only aspect of being LGBT that we can properly reference to reliable sources.
 * I can assure you that there's no agenda on my part to minimize the size of any quadrant of the LGBT community; in actual fact, I make every effort I can, while relying on reliable sources, to make sure we're getting as many LGBT people as possible into Wikipedia. Just in the process of working on Lambda Literary Awards in the past few weeks, for example, I've probably caught about 100 writers who are openly LGBT but weren't yet being described and categorized as such, and I spent about a week last November concentrating on nothing but the list of openly LGBT politicians who were newly elected to US stage legislatures. (And I'm Canadian!) So believe me, I'm not trying to tamp down our coverage of LGBT figures — I've quite easily done more than almost anybody else on Wikipedia to beef that area up.
 * But again, what's relevant for our purposes is the social and cultural aspects of being LGBT, not the physical ones — what's relevant for our purposes is a person's public life, not their private one. So no matter how gay or lesbian or bisexual a person may actually be in their personal lives, it's not relevant to us until they choose to make it relevant by coming out and claiming a gay, lesbian or bisexual identity in their public ones. Not because we're trying to minimize LGBTness on here, but because (a) it's generally the only aspect of sexual orientation that we can actually source properly, and (b) doing our own investigative research into people's personal/private lives simply isn't what we're interested in on here anyway. We're in the business of summarizing what reliable sources have already published about a person, not in the business of outing people — and that's not a double standard, because we don't publish unsourced claims about heterosexual people's sex lives either. Bearcat (talk) 20:55, 24 July 2013 (UTC)
 * Bearcat, naughty bits lol. But anyway, Bearcat is right. I agree with it all. Public record supported by reliable sources is what Wikipedia reflects. If reliable sources don't define a specific orientation, even if we know ourselves, then it doesn't exist to Wikipedia. Can't add original research. Teammm $talk email$ 21:30, 24 July 2013 (UTC)

Lovingboth's suggestion has a number of problems. We have just as much of a problem using sexual behaviour as the basis for categorisation as we do for using self-proclaimed sexual identity. Now, I'm not a notable person in the notability guidelines sense of the word notable. I'm not sure I'd like to be a notable person either given some of the BLP dramafests I've seen. But, let us say I were to become a minorly notable person and were to get a Wikipedia article. My sexual identity is out there, because I'm openly gay. But I hope to all the gods of the Greek Pantheon that my sexual behaviour does not get out into the public realm, because who I sleep with is my own damn business. You can be reasonably sure that they are going to be dudes, but the thought of someone on Wikipedia deciding on my sexual orientation according to some criteria of who I have or haven't slept with gives me more heebie-jeebies than WP:ANI on a particularly bad day. In my ideal world, none of my sexual activities would become a matter of public discussion. And if we are using behaviour rather than self-asserted identity as the basis of sexuality categorisation and description, if there are no public reports of a person's sexual relations—hetero or homosexual—does that person then become some kind of de facto asexual? That way lies madness.

And it's silly. There are straight people who have engaged in same-sex sexual activity. That doesn't make them gay. There are gay people who have engaged in opposite-sex sexual activity and that doesn't make them straight. The labels aren't necessarily precise: there can be two people at the same location on the Kinsey Scale, and one might use "gay" or "lesbian" to refer to themselves, while another uses "bi". The key is in the term orientation: a person's sexual attractions are pointed (oriented) in a specific direction, but that doesn't mean that the person always acts in accordance solely with their attractions. —Tom Morris (talk) 23:05, 24 July 2013 (UTC)
 * "There are straight people who have engaged in same-sex sexual activity. That doesn't make them gay. There are gay people who have engaged in opposite-sex sexual activity and that doesn't make them straight." Yep, that's some of what I would have stated in this discussion.


 * And, by the way, since this discussion primarily concerns LGBT people/people who may be LGBT, I alerted WP:LGBT to this discussion. So that's why some of that project's members know of this discussion. Flyer22 (talk) 23:25, 24 July 2013 (UTC)


 * I don't doubt that this is a good-faith suggestion, but I agree entirely with Tom and Bearcat's comments that it is very problematic and (even if it were desirable) would be unworkable. Andrew Gray (talk) 23:38, 24 July 2013 (UTC)

Correct categorization quiz
Many people have been accusing wikipedia of sexism and racism over this most recent case of sub-category diffusion gone awry.(see Categories_for_discussion/Log/2013_April_24 where the word sexist appears around 40 times, uttered in various contexts)

At the heart of the issue is the question of ghettoization, which is covered (briefly) at the end of this guidance, and at the beginning (at least with respect to gender), and the question of whether removal of or non-presence of someone from the parent category has sexist or racist results.

As a way of illustrating the challenges in avoiding this so-called "ghettoization" (and resultant accusations of sexism and racism), I wanted to propose the following quiz. Your challenge, should you choose to accept it, is to provide a revised list of categories for Winona Duke in a way that conforms to the wikipedia policies as outlined in WP:EGRS and any other relevant guidelines around categorization of people, and ensure that you have not ghettoized her, or classified her in a sexist or racist way. I've chosen a complex example on purpose, to illustrate the challenges inherent in our category trees: Winona LaDuke (I've given a fixed revision so you can't cheat in case others start adding cats to the real article - please don't edit the real article until this is all done...)

Your input is the list of categories she is currently part of, and your job is to add, delete, or update any and all relevant categories - providing a list of any changed or new categories. Here is the starting list of cats she's currently assigned to: (Difficulty: Easy - don't add any categories that don't yet exist in the wiki) Category:American economists, Category:Female economists, Category:American environmentalists, Category:American non-fiction environmental writers, Category:American non-fiction writers, Category:American women novelists, Category:Women writers from Minnesota, Category:American social democrats, Category:Minnesota Greens, Category:Green Party of the United States vice-presidential nominees, Category:Female United States vice-presidential candidates Category:United States vice-presidential candidates, 1996, Category:United States vice-presidential candidates, 2000, Category:Native American politicians, Category:Native Americans' rights activists, Category:Native American women activists, Category:Reproductive rights activists,Category:Ecofeminists,Category:Native American novelists,Category:Native American writers Category:Jewish American writers,Category:Jewish American politicians Category:Antioch College alumni,Category:Harvard University alumni Category:Ojibwe people,Category:American people of Russian-Jewish descent,Category:People from Ashland, Oregon,Category:People from Los Angeles, California,Category:1959 births,Category:Living people Category:Green Party of the United States politicians,Category:20th-century American novelists

I have prepared an answer key based on my reading of the guidance and my perusal of the category tree (it took me around 60 minutes to do so), and I will compare your answers with mine to determine your score. I hope that those attempting this exercise will realize the complexities in doing this well, and also understand the consequences of getting it wrong. You can have someone correctly categorized with 32 cats, slicing in every which direction, but if you miss ONE appropriate parent, you may risk being accused of sexism.

To make this fair, I'm hiding the answers received below. Please prepare your answers without looking at others' results, and no changing the answers once you're done - I'm interested in your best raw guess as an individual editor - once it's all done we could have a robust discussion about (1) what the actual "right" answers are (I'm certainly not claiming to have the authoritative result!) and (2) how to change the guidance to make it easier to follow.

Note: Please check membership in any new categories you choose - for example - you may be tempted to put her in to de-ghettoize her, but this would be wrong, as  is already a fully diffusing category - there are no other bios sitting there today. So you need to take this sort of thing into account, and will be graded accordingly. :) --Obi-Wan Kenobi (talk) 15:11, 6 May 2013 (UTC)

Quiz answers
Stick your answer in a sandbox.

Quiz discussion
Let me know what you think of this exercise.--Obi-Wan Kenobi (talk) 22:19, 28 April 2013 (UTC)
 * I find it astounding that you came up with an authoritative solution in 20 minutes. That's very impressive. This quiz is a really cool way of illustrating and thinking about these problems. -GTBacchus(talk) 02:46, 29 April 2013 (UTC)
 * Thanks. I'm looking at James Baldwin now, I think it would take me more like 30-40 minutes to sort that one, because there are grand-parenting issues. Sheesh. --Obi-Wan Kenobi (talk) 03:16, 29 April 2013 (UTC)


 * Okay, this quiz is a great illustration of the non-editor-friendly nature of the current system. However, there are solutions.
 * Simple part: Put the distinguished category tag on E/G/R/S categories and copy upward from ERGS identifier + noun to noun category.
 * Harder part: Carefully prune ERGS categories based on the Would the head article for this category be notable? standard.
 * Now, Winona LaDuke is a nice intersectional case (me too, if I were notable, for what it's worth), but let's prune the list of her down to the actual issue at hand, by eliminating cats with no ERGS subcats: Category:Female economists, Category:American women novelists, Category:Women writers from Minnesota,   Category:Female United States vice-presidential candidates, Category:Native American politicians, Category:Native Americans' rights activists, Category:Native American women activists, Category:Reproductive rights activists,Category:Ecofeminists,Category:Native American novelists,Category:Native American writers, Category:Jewish American writers,Category:Jewish American politicians, Category:Ojibwe people,Category:American people of Russian-Jewish descent.--Carwil (talk) 12:58, 29 April 2013 (UTC)
 * I like the way youre thinking - but don't give away too much! For now, try to put together the Category diff and post in the hidden area above, once this is done we can discuss optimal algos - fwiw your algorithm will still result in ghettoization in this case, more subtle manipulation is needed... --Obi-Wan Kenobi (talk) 14:28, 29 April 2013 (UTC)
 * I'll be back with my (hidden) sandbox answer when real-life work is done.--Carwil (talk) 14:42, 29 April 2013 (UTC)


 * As a novice wikipedia editor, I don't quite understand this quiz. Is the goal to pair down the starting set of categories until an irreducible set exists? Or do you also have to scour wikipedia for additional categories the article should be in?Semitones (talk) 02:58, 30 April 2013 (UTC)
 * I took the goal to be both of those. -GTBacchus(talk) 03:52, 30 April 2013 (UTC)
 * Yes - if you see categories that aren't needed (e.g. if she is in a category and its parent, and the parent is diffusing, she should be removed from the parent). However, due to WP:EGRS, gender/ethnic/religious/sexual orientation cats are non-diffusing, which means you need to bubble her up to non-gendered/ethnic cats. So your goal is to both find cats that aren't needed anymore, but more importantly, to find any cats that should be added, especially those where her lack of membership in them could lead to the NY times writing an article. I'll give you a hint - start with the cats she's in, and look at the parents of those cats.--Obi-Wan Kenobi (talk) 04:54, 30 April 2013 (UTC)
 * Not surprised by my lousy score. I got bored with the exercise after a few minutes (It's hard to care about slotting an article into every category where it might conceivably belong! -- And if a person left their birthplace before the age of three, how many people would truly consider that person to fit in a half-dozen categories related to that birthplace?) and submitted my answers "as is". I believe, however, that I correctly removed her from a couple of categories where she didn't belong. --Orlady (talk) 15:15, 30 April 2013 (UTC)
 * Boring, tedious - such is the work of categorization. I see your point re: where she is from - but even if we ignore those cats, you left her gravely ghettoized on one or two cats... Will be fun to discuss my answers and see if we could come to consensus on the full set of cats for her... --Obi-Wan Kenobi (talk) 15:22, 30 April 2013 (UTC)


 * Actually I think the rule against "categorizaing by birth" would say we should not categorize people by where they were born. The extreme is I recnetly moved someone from the German dancers to Spanish dancers cat structure because they were born in Germany but their family returned to Spain when they were 15 days old.John Pack Lambert (talk) 17:11, 30 April 2013 (UTC)
 * This was a useful exercise for me to labour through because it helped me appreciate the hard work category editors do, even though I have been a gleeful rabble rouser on the recent controversy. Since I don't go cat work regularly, the press no doubt loves me.--Milowent • hasspoken  21:04, 2 May 2013 (UTC)
 * Yeah...Does anyone use categories, and if so, what for? I mean anyone as a user, not as an editor.KD Tries Again (talk) 14:27, 3 May 2013 (UTC)KD Tries Again
 * I sometimes use categories as a reader. I read botany articles, and if I'm curious what other plants are in the same genus as a plant I'm researching, there's generally a category for that. That's quick, one-click access to a list that's usually pretty complete, and which cannot always be found in article-space. That's just the first example that occurs to me; there are others. In this connection, I've never, ever been bothered or felt hampered in my navigating because a category was too big. -GTBacchus(talk) 15:34, 3 May 2013 (UTC)
 * Botany is a great example. So take the taxonomy of Category:Lycopodiaceae. Would you be ok if all of the contents of its child Category:Diphasiastrum were also there? And thus, would that be ok if all of Category:Lycopodiaceae and its subcats were now present and visible in Category:Lycopodiopsida, and then all of those would be present in Category:Lycopodiophyta? That's what non-diffusing means - everything just keeps bubbling up the tree, until at the top level, you'd have an article for every single plant. I personally think this is a *really* bad idea. if you really want to see all Category:Cryptogams, CatScan is your friend, but I'm not at all convinced this list of thousands should be displayed to the user when they show up on the category page. --Obi-Wan Kenobi (talk) 16:08, 3 May 2013 (UTC)
 * All 4 members of Category:Diphasiastrum? Would that break the bank? In botany, what would really matter to me is consistency. How far up the taxonomic system should pages bubble? I could imagine a convention where, say, a Subfamily is categorized according to the Family and Order to which it belongs, but not the Class. In other words, 2 taxonomic level up. That would be fine, if it were applied consistently, and the inevitable large categories resulting would not bother me one bit. Some taxa are bigger than others. Honestly, I don't know what an average reader wants when they show up on a category page, and I'm not sure we should assume we know that, unless we've determined it based on feedback from actual readers who use the pages. I've now seen the claim that large categories are unhelpful repeated many times, always by editors. I'd like to see one example of a reader saying that. -GTBacchus(talk) 19:16, 3 May 2013 (UTC)
 * Ok, this is now off-topic for this particular section - I see your point, but I'm afraid your solution is not workable. I'd suggest if you'd really like this, bring it to the biology project and ask the taxonomists what they think. My guess is, they will not agree.--Obi-Wan Kenobi (talk) 20:12, 3 May 2013 (UTC)
 * It's not "my solution"; I'm just thinking aloud. That's acceptable, right? It's actually not a far cry from what they currently do with these pages, if I'm not mistaken. Maybe it's one main level instead of two, but I'm pretty sure it's been standard for years to at least bubble up past Subfamilies to Families. You're right, though, that we're off-topic. For biology, there's no danger of ghettoization, so it's all about consistency, manageability, and usefulness are the criteria. My second point was on-topic, however. We should base our notions of what's useful on what actual readers find to be useful, and not just on what we assume they will find to be useful. -GTBacchus(talk) 19:06, 5 May 2013 (UTC)
 * Fully agree, but those notions should be tempered by what's possible with the current software, and what is maintainable by an army of volunteer editors, and to avoid things that are likely to cause inconsistencies. The idea of bubbling up in some cases is acceptable (e.g, because you have a very small set of articles, and it also wouldn't make sense to diffuse this. However, larger categories like Novelists, Poets, Journalists, etc are so large that having them not be diffusing would make things a lot more complicated, both for editors *and* readers, for reasons I've outlined elsewhere, and with very little value provided to the reader in exchange for the work required to set up and maintain such schemes (esp since we can already give the reader a fully enumerated recursive list on demand). In fact, every time we have a category that isn't fully diffused (eg in a partial state of diffusion, like or fully non-diffusing (like  or , we are doing our readers a disservice, by confusing them terribly as any glance at the media around this will reveal. But this whole thing is run by volunteers, you can't exactly just force people to start diffusing or bubbling up, either way... Obi-Wan Kenobi (talk) 21:18, 5 May 2013 (UTC)

Note: Answers will be posted by next Wednesday, so if you want to submit your answer for grading please do so soon! :) --Obi-Wan Kenobi (talk) 20:24, 10 May 2013 (UTC)

Quiz answers
The following categories should be added to Winona's bio: (but not: Category:Writers from California) The following categories should be deleted:
 * Category:American civil rights activists
 * Category:American feminists
 * Category:American poets
 * Category:Native American activists
 * Category:Native American feminists
 * Category:Native American poets
 * Category:Jewish American novelists
 * Category:Politicians from Los Angeles, California
 * Category:Women writers from California
 * Category:Writers from Los Angeles, California
 * Category:Writers from Minnesota
 * Category:People from Los Angeles, California (already in sub-cats)

User:Johnpacklambert also came up with some categories that I didn't, and which I agree with: Categories to add from JPL:
 * Category:Jewish women writers
 * Category:People from Becker County, Minnesota

Categories to delete from JPL:
 * Category:American non-fiction writers
 * Category:Green Party of the United States politicians

Discussion on quiz answers
Sorry it took a while to post these - got distracted by other stuff. The most common error was leaving out Category:American civil rights activists and Category:Native American activists, or some of the by-state+gender writer categories (some people left her as a woman writer from X, but not as a writer from X. Unfortunately due to my delay in posting the "answer" key, several categories that she was in have since been deleted. In any case, I welcome any thoughts or discussion on the "correct" set of categories for Winona, and feel free to disagree with the above answer key! :) Invocation: User:GTBacchus User:Orlady User:Milowent User:Carwil User:Johnpacklambert --Obi-Wan Kenobi (talk) 14:54, 15 July 2013 (UTC)
 * Curious why you have Jewish-American novelist but not American novelist or Native American novelist. It's also the only Jewish-related category that you have here. Why Jewish novelist and not feminist, poet, politician or activist? It seems like there are inconsistencies in your list. If she is a Jewish Native American, wouldn't all of those categories be non-diffusing to two ethnicities and one nationality? That is, if those categories (like Native American politician) exist. Liz Read! Talk! 15:15, 12 September 2013 (UTC)
 * Yes, it's entirely possible I missed some. Note that some categories have changed/been created since this quiz went online. I think the main point is, serious editors who spend time still miss things, so using the term "sexism" because someone accidentally left off one category is probably an misuse of the term.--Obi-Wan Kenobi (talk) 14:29, 13 September 2013 (UTC)

Religion categories for dead people
I've removed some dead people from religion cats recently on the grounds of WP:BLPCAT principles. Even though, obviously, those are not BLPs, I thought the same principles applied. However another user has just pointed me to WP:CAT/R, which seems to suggest that simply having refs that a person belonged to a religion is sufficient. The wording at WP:CAT/R seems counter-intuitive to me. The purpose of cats is usually to help users easily locate similar topics, in this case people notable in some way for belonging to a particular religion. If we simply add everyone who belonged to a particular belief to those cats, then in many cases they will become swamped. Irish Catholics category for example could include up to 90% of dead Irish people, making it hard to see who is genuinely notable for belonging to that church. Could I have some clarification on this before I change anything else? Valenciano (talk) 15:09, 16 August 2013 (UTC)
 * Don't apply BLP to clearly dead people please. The self-identification is required to avoid sticking labels that living subjects disagree with. But for dead people we don't risk enraging them anymore. On this, WP:CAT/R is very clear. That some categories have a huge amount of entries is not a problem -after all, Category:Living people has a lot of entries. If you want to list people notable for being Irish Catholics, you can have a List of Irish Catholic Church personalities or something like that. -- cyclopia  speak!  15:16, 16 August 2013 (UTC)
 * I've no intention of making edits against consensus, I'm simply trying to clarify what it is. As I've already said, I wasn't applying BLP, I was applying the principles of BLPCAT. BLPCAT currently requires more than self-identification, it also requires that "the subject's beliefs ... are relevant to their public life or notability, according to reliable published sources." That suggests to me that there is a further criteria to be met when categorising people by religion and I wanted to know if that also applies when people are dead. It doesn't seem logical to me that if a person wasn't notable for their religion when alive, we then add them when dead. Category:Living people is a special case, since it allows us to monitor BLPs, so that seems to me to be a clear exception to the issue of cluttered categories. Valenciano (talk) 15:42, 16 August 2013 (UTC)
 * Sorry, but WP:BLPCAT is a subsection of WP:BLP, so it's nonsense to say "I wasn't applying BLP, I was applying the principles of BLPCAT". BLPCAT is what it is because it is meant to deal with living people, where a lot more caution is in order because of privacy and sensitivity of certain categorizations. In other words, BLPs are an exceptions. For long-dead people we do not have to worry, and we come down to the default state, that is "rely on sources". As written here.-- cyclopia  speak! 16:05, 16 August 2013 (UTC)
 * I agree with Valenciano here; this has also been the lived consensus around religion categories. You are not added to a "Catholic" category unless the belief is relevant to their public life or notability. If we only applied this to alive people, it would mean we would remove the categories once they died, or add the categories once they died - neither way makes sense. if being catholic is defining, it is so while they are alive or not.--Obi-Wan Kenobi (talk) 21:47, 13 November 2013 (UTC)

Blossoming of new women + job categories
I've noticed a blossoming of new women + job categories in the past few months. Under Category:Women_by_occupation you will find cats such as ; ; ;;;;;;; and new categories of this type are being created regularly. At the same time I've been criticized indirectly for nominating too many gendered categories for discussion, so I've brought this here for a broader look. I think we need to clarify the terms of this guideline, specifically re: when should gendered job categories be created, and when should they not be? To me, seeing something like "Women farmers", considering that women do the bulk of farming in the world, is a bit jarring, and it doesn't make much sense to genderize that category in that way. The reason I'm concerned with this trend is that, if it continues, we will eventually have Women+job for every job, which necessitates all of the fully developed parallel category trees to keep that working correctly, which in my experience is often not done correctly, and generally can lead to ghettoization. More importantly, however, is that this trend, especially if continued to its logical conclusion, reinforces the idea that women are a special type of human, and that the default type of human is men. I think this is harmful, ultimately, to the project and to the goals of gender equity. I would rather see fewer, rather than more, gendered categories, especially ones that are poorly populated as the above examples, and I think such gendered categories should really focus on areas where there was a major difference historically in the performance/roles/expectations/access to the jobs in question and that such differences persist somewhat to this day. I'm not convinced, for example, that Women collectors are some special, rare breed of collectors, nor are women bartenders, especially now, a special type of bartender.--Obi-Wan Kenobi (talk) 03:52, 1 December 2013 (UTC)

Would like some discussion on sports categories split by gender
Recently there was a CfD discussion (see here) where Category:American men's basketball players was proposed for deletion. An editor quoted the following from this guideline: "As most notable organized sporting activities are segregated by gender, sportsperson categories constitute a case where "gender has a specific relation to the topic". As such, sportsperson categories should be split by gender, except in such cases where men and women participate primarily in mixed-gender competition. Example: Category:Male golfers and Category:Female golfers should both be subcategories of Category:Golfers, but Category:Ice dancers should not have gendered subcategories." I would really like to debate if this guideline should be changed. The result in this case (basketball players) is what I think is ridiculous over-categorization. What we had before gender segregation was "Category:American basketball players" which was diffused into 50 subcategories for each state. Now we have the male and the female version of the parent category, which in theory would result in 100 diffused subcategories or the inclusion of 2 categories ("American men's basketball players" and "Basketball players from Idaho" for example) where one was there before. And these segregated categories serve little to no purpose to the reader in my opinion. Can we please discuss the merits of this part of the guideline? personally I would prefer these athletes not be segrgated by gender, but I realize there may be other reasons at play that are more compelling than those stated in the guideline itself. The argument that they don't compete together to me is silly - at the end of the day they are all just basketball players (or fill in the name of other sport). I really don't see the need and frankly I am not going to add these categories to the hundreds of articles I edit each month because I think they are silly (I won't delete them, but I am not helping the effort). Rikster2 (talk) 20:43, 13 November 2013 (UTC)
 * there are many opinions on this matter. My personal feeling is, the gendering is ok, especially for sports - since basketball is played by gender-segregated teams - however, I don't think we need to gender ALL sub-categories. As such, I think we should have and  and then all other sibling subcategories should be gender neutral. That way, each player would have at least 2 categories for basketball (possibly more, if they are African american, for example). That to me is much better than all of the hundreds of sub-categories.--Obi-Wan Kenobi (talk) 21:51, 13 November 2013 (UTC)
 * What is truly the value of gender splitting basketball players? In theory, women could play in "men's leagues" (as an example, Ann Meyers was signed to an NBA training camp) and these are just occupation/citienship categories so having 2-3 categories seems silly.  How would readers actually use these categories? And why would you have one system at the top level and another at the subcategory level? Rikster2 (talk) 21:58, 13 November 2013 (UTC)
 * there is long-standing consensus to split certain jobs by gender, or at least to create a woman or men category - eg Women heads of state, Male nurses. In some social categories, like actors, models, and many sports, it was decided to have a full split by gender. If you didn't do a full split by gender, you'd likely still have a women's cat, bc women's basketball players are discussed in RS as a group. Then someone might say, well I want to split out the men too - so then you end up with a gender split anyway. As to not splitting all the way down the tree, this is very common - we have women mathematicians, but we don't have Canadian women number theorists for example - so generally gender/ethnicity/sexuality always are kept pretty high up, leaving the more nuanced categories as gender/ethnicity neutral.--Obi-Wan Kenobi (talk) 13:44, 14 November 2013 (UTC)
 * I know there is a long standing consensus to split some jobs by gender. I am suggesting that in this case this split is not necessary and over-categorization and that it may be that it should be amended.  This would encompass (in this example) both the men's and women's basketball player categories.  I don't really agree that you'd necessarily have a women's category.  Frankly, I am tired of Wikipedia guidelines being cited ad nauseum without a common sense look at if the policies/guidelines still make sense and actually serve a purpose for the user.  Wikipedia doesn't exist to be a set of rules to be enforced, it exists to be an encyclopedia for users.  I am not sure I believe that end users get benefit from all these categories.  Rikster2 (talk) 14:46, 14 November 2013 (UTC)
 * Well, you can try to nominate women's + men's basketball for deletion, or you can propose a change to the guideline here. IAR permits us to ignore rules, but without some set of guidelines to fall back on, everything just becomes opinion. I personally think we have far too many gendered categories, and I try to eliminate them when I come across them if they go too far, but its not my decision, its up to consensus. Unfortunately, the current set up is anyone can create a category they are super easy and require no research, but it requires consensus to delete a category. As for basketball specifically, I'm still rather confused by your opposition here - basketball is still, to a very large extent, separated by gender. That a few exceptions exist only proves the rule (b/c those exceptions are so talked about)--Obi-Wan Kenobi (talk) 14:55, 14 November 2013 (UTC)
 * But basketball as an occupation or pasttime isn't gender specific, regardless of how the competition structure is set up. I think the fact that participation largely is gender segregated frankly isn't good rationale for creating extra categories.  I would be happy to propose a change to the guideline because I think it is futile to CfD while the guideline supports the current structure.  I think the direction that "sportsperson categories should be split by gender, except in such cases where men and women participate primarily in mixed-gender competition." should probably be changed to "sportsperson categories may be split by gender, except in such cases where men and women participate primarily in mixed-gender competition."  I really don't see any reason why this model has to be the case, yet some editors lean heavily on this guideline to create extra, and in my opinion useless, categories. Rikster2 (talk) 17:10, 14 November 2013 (UTC)
 * There are entire books written about women's basketball, they have a separate league, slightly different rules: one book. I agree, you don't *have* to split the categories by gender, but why start with basketball, which is very strongly gendered? How it's played as a pasttime is irrelevant, as we don't categorize people who play basketball as a pasttime - it is really mostly people who play professionally, where the competition is gendered for the most part. Again, think about it this way - what you're really saying is, you'd like to delete (and all country siblings), but given there are books and documentaries and television shows and lists and booster groups and all sorts of other stuff specifically around the topic of American women's basketball, to delete it would be ignoring the fact that outside sources classify people in this way.--Obi-Wan Kenobi (talk) 17:48, 14 November 2013 (UTC)
 * There are plenty of books specifically about female business leaders as well, yet Category:American chief executives is not segregated by gender (nor should it be). And basketball as a pasttime absolutely figures into these categories as "Basketball players from State X" is applied to high school and college players, not just professionals.  And the answer is I'd like to start with basketball because that's what I focus on so I do a lot of the work associated with this overcategorization (adding categories, etc).  I don't presume to speak for tennis editors or whatever else.  As an aside, when I look around we have many "ghettoes" in Wikipedia.  Whose bright idea was it to create Category:American women state governors (without a corresponding category for men, of course)? Rikster2 (talk) 19:34, 14 November 2013 (UTC)
 * for the record, that was just renamed to Category:Women state governors of the United States – Fayenatic L ondon 23:37, 24 December 2013 (UTC)
 * re: execs, see Category:American women business executives; what I meant by pastime was that this is not for hobby players - any high school or college players we categorize here are obviously top-level amateurs, otherwise they wouldn't be notable enough to be here. For the governors, I assume that's because most governors are men, so when you create a gendered category you don't always have to create a matching female equivalent. These aren't necessarily ghettos if you de-ghettoize properly (e.g. make sure they are in a neutral category as well). Anyway, good luck, but I doubt you'll get support to de-ghettoize basketball, but I'll let others opine on that.--Obi-Wan Kenobi (talk) 19:49, 14 November 2013 (UTC)

Question on deaf/blind categories
I started a discussion here ==> Wikipedia_talk:WikiProject_Disability - please join. One of the issues is whether WP:EGRS should be expanded in scope to cover people with disabilities and categorization rules for same, since there are similar issues at play.--Obi-Wan Kenobi (talk) 17:06, 21 February 2014 (UTC)

Final rung
The last paragraph of the guidelines states: "Also in regards to the 'ghettoization' issue, an ethnicity/gender/religion/sexuality subcategory should never be implemented as the final rung in a category tree."

Huh? All categories are the final rung when they are first created. This rule basically means that ethnicity/gender/religion/sexuality subcategory can never be created. Kaldari (talk) 18:36, 25 February 2014 (UTC)
 * This could perhaps be worded better. The intent is, suppose you have with no subcats underneath. It would not be allowed, per the last rung rule, to create  as a subcategory, because this would tend to ghettoize. If, on the other hand, you had existing sub-categories, such as  into which the parent cat could be diffused, then the gender/religion/sexuality subcat could be created as non-diffusing sibling.--Obi-Wan Kenobi (talk) 18:49, 25 February 2014 (UTC)
 * Well that's about as clear as mud. Perhaps suggest a better wording than the one quoted by Kaldari?Anythingyouwant (talk) 18:56, 25 February 2014 (UTC)
 * I think you mean, "Also in regards to the 'ghettoization' issue, if any category contains only one subcategory, then that subcategory should never be an ethnicity/gender/religion/sexuality subcategory." Is that right?Anythingyouwant (talk) 20:34, 25 February 2014 (UTC)
 * It's not only about being the "only" subcategory. For example, you couldn't have with subcats of  and . You could however, have  underneath  and as a sibling of  - the key is whether the head cat has diffusing children.--Obi-Wan Kenobi (talk) 21:06, 25 February 2014 (UTC)
 * Thanks, but I don't think people would understand what a "diffusing child" is. If the language I suggested merely needs to be supplemented instead of changed, then what would the supplement say?  On the other hand, if you and Kaldari are happy with the "diffusing child" terminology then I won't revert, it just seems like the language could be clearer than that.Anythingyouwant (talk) 21:15, 25 February 2014 (UTC)
 * I wasn't proposing language, I was just trying to explain... I think this whole guideline is in need of a refresh, so I don't know if we should just chip away at a little part of it. may have something to say on this.--Obi-Wan Kenobi (talk) 21:20, 25 February 2014 (UTC)

I don't have any objection if somebody wants to rewrite the section to make it clearer what's actually intended, but for the record Obi-Wan has it exactly right. The general idea is that we don't want a situation where all people in a particular occupation are filed directly in the main "occupation" category except if they're a member of a minority group, in which case they suddenly get shunted off to a "minority occupation" ghetto instead — rather, marked identity subcategories should exist only if the parent category is also subcategorized on an unmarked scheme (e.g. by country, by state or province, etc.) that enables the identity categories to work alongside other categories rather than replacing them.

For instance, we wouldn't want a situation where the unmarked directly contained writers who are white, male and heterosexual while banishing people of colour, LGBT and women writers to special marked categories, and thus removing those people from any category they would otherwise be in. Rather, since is already diffused on other grounds (state of residence, particular type of writing, time period, etc.), the marked identity category is complementing, rather than replacing, other categories in which women can still sit directly alongside men, queers directly alongside heteros, people of colour alongside whites, and on and so forth. The basic idea is that marked identity categories should be formulated in a way that doesn't preclude the person's appearance in unmarked categories; the identity categories need to be structured in a way that complements, rather than superseding, the appropriate unmarked occupational categories.

As Obi-Wan so aptly put it in an earlier discussion above, what we want to avoid is creating the perception, intentionally or otherwise, that the default setting for any given occupation is "straight white men", while anybody else who isn't one or more of those things is some special kind of "not really the real thing" consideration. So a marked identity category can sit alongside the appropriate level of "real thing" categorization, but cannot supplant filing people in unmarked occupational categories. (This is, for example, why was not able to exist until there was a consensus established to match it with  — as long as the plain unmarked  directly contained men, a separate category for actresses would have been hiving the women off to a subcategory of men instead of a parallel sibling category within a gender-neutral parent.)

If anybody can come up with a clearer way to articulate that principle, then by all means go for it. Bearcat (talk) 22:54, 25 February 2014 (UTC)
 * One wrinkle I've come across is as follows - what happens if a gendered cat is not last rung in one tree, but is last rung in another tree? We also have a conflict with the small-cat exception, whereby if one has American women X one should presumable allow Norwegian women X as well, but often these other countries are less developed and have many fewer articles so you can't actually create a set of diffusing siblings. Look at for example - it's a last rung. Should it be deleted? I'm not sure as it's also part of a series.--Obi-Wan Kenobi (talk) 01:36, 26 February 2014 (UTC)
 * Well, if you check, it's worth noting that there are numerous other women in there who haven't been added to the women-specific category yet, so it escapes WP:SMALLCAT on that basis. And the parent category can also easily be subdivided, the way several other countries' novelists category already are, by the particular century in which the person published novels — the only reason that isn't already in place is that nobody's started the appropriate categories yet. So while the parent category isn't already diffused, it is legitimately diffusable. Bearcat (talk) 01:49, 26 February 2014 (UTC)
 * After reading the above discussion about 3 times, I think I finally understand it. That said, I think there is approximately zero chance of getting more than 5 people on Wikipedia to fully comprehend it :P I can't wait for categories to die. Kaldari (talk) 04:24, 4 March 2014 (UTC)

Of descent categories
Hey! I'm posting here, after doing a lot of category work on ethnic groups and ethnic ancestry since last summer, because I'd like to get a second and third opinion. I've read over WP:EGRS multiple times and now have read Requests for comment/Categorization of persons. This topic (below) is headed towards dispute resolution and I just wanted to check in first before weighing in on that case once it is filed.

The question is "How far back does descent go when considering categorizing individuals on Wikipedia?" So, when someone is filed under a category like Category:Brazilian people of German descent, is this going back to grandparents, great-grandparents and great-great-grandparents? Or is this descent going back 500, 1000, 2000 years?

I always thought it was proximate descent so it just went back 3 or 4 generations but, obviously, there is no hard and fast rule about this. But when discussing categories (like Category:Mexican people of Jewish descent), there are some editors that want to categorize every "of Jewish descent" category as being Category:People of Asian descent and Category:People of Southwest Asian descent because, their argument goes, every person who has ever had a Jewish relative can trace their origins back to the Middle East, even if it is several millennium ago. Remember, this is not "Jewish people" but "people of Jewish descent", meaning, people who are not themselves Jewish but who have (or believe they have) had a Jewish ancestor.

With this philosophy, every Native American could be categorized as being of Asian descent and every ethnic group could be said to be of African descent since that is where human origins started. DNA gets brought into the argument along with questions of self-identity when I really think that the grounds for categorization should be clear and obvious and not reflect a particular point of view.

I guess, given the way I've framed this conversation, my opinion is obvious but I'd like to hear from other people who work with biographical categories. I know that there are some editors who wish all "of descent" categories were deleted but, right now, they exist and it's a question of how far back descent goes (and this is any descent, not just Jewish). Thanks! Liz Read! Talk! 20:49, 4 March 2014 (UTC)
 * I think there is a difference between the top-level containers, and the categories applied to people. For example, we categorize black people in the US as "of African American descent", and then all of those African-Americans are rolled up under "of African descent", even though many came to the US several hundred years ago. That said, my thoughts are:
 * For people categories, we should not categorize descent any further than grandparents. I would even be ok with not categorizing descent any further than parents, unless the grandparents descent was truly defining for that person and regularly mentioned by sources
 * For the container categories, we should use common sense; which in this case, for most Jewish people, would not be middle eastern descent - if you are a polish jew, you are in the container of "european" descent.--Obi-Wan Kenobi (talk) 22:03, 4 March 2014 (UTC)
 * In theory, people should belong in categories for any ethnic descent that can be properly sourced, rather than just asserted — though, for obvious reasons, it's often difficult or impossible to properly source anything much farther back than two or three (or very occasionally four) generations. And even when it can be sourced, it often fails the relevance test. (If I go far enough back on my mother's side, for instance, I do have some fairly distant ancestors who were actually from England instead of France — but I'm not "of English descent" in any meaningful way, because any "English" genes I may have lingering around in my DNA came to me through a line of descent that was culturally French Canadian and completely lost any substantive sense of Englishness.)
 * And it's not appropriate to conduct our own research into an article topic's genealogy either, so there's no place for us to concern ourselves with any ethnic ancestry that can't be referenced to material that's already been published in reliable sources — DNA's mostly a moot point, because appearances on Who Do You Think You Are? notwithstanding, there are very few people whose results from genealogical DNA testing are actually verifiable in reliable sources.
 * So I'd have to agree with Obi-Wan that in most cases, we shouldn't concern ourselves with anything much older than a person's grandparents. There might be valid exceptions sometimes, but those depend entirely on the quality of sourcing that's actually available. Bearcat (talk) 23:19, 4 March 2014 (UTC)
 * Thank you for sharing your opinions, Obi-Wan Kenobi and Bearcat. A major problem with this dispute is that some folks (like me) are looking at this issue in a pragmatic way (categories as a navigational tool to organize articles) while other editors see categories as making a statement, as defining the people who are contained in the category, as reflecting identity. I think that is pointy and uses categories as editorial content. Liz Read! Talk! 01:45, 6 March 2014 (UTC)