Wikipedia talk:Categorization

(Redirected from Wikipedia talk:Categories)
Latest comment: 1 day ago by Johnpacklambert in topic Excessive sub-dividing of ambassador categories

Sort key with article and disambiguator

edit

Should the DEFAULTSORT for The Salt Path (film) be ["Salt Path, The (film)" or "Salt Path (film), The"?

A recent edit by @Fuddle: changed it from the former to the latter and I want to revert but would like to see documentation or consensus to quote when doing so, especially as it was done using a tool or gadget called "Cold Default Sort".

It seems more logical, to me, to sort all articles with the title "The Salt Path" together, after any (hypothetical) works "A Salt Path" and before "Salt Path Adventure", and keep the bracketed disambiguator as the final element.

Should this be added to the documentation? Or have I got it wrong?PamD 05:08, 27 July 2024 (UTC)Reply

@PamD: Consider that WP:SORTKEY says to remove all punctuation except hyphens, apostrophes and full stops. This would make your choices Salt Path, The film or Salt Path film, The. That apart, I can see arguments on both sides. --Redrose64 🌹 (talk) 09:51, 27 July 2024 (UTC)Reply
But commas are included as in "Smith, John", so it contradicts itself. PamD 10:42, 27 July 2024 (UTC)Reply
Commas are a special case. You remove them from the original page name before rearranging it, it does say Commas can be added when re-ordering words, as in the previous example. --Redrose64 🌹 (talk) 11:44, 27 July 2024 (UTC)Reply
If the movie were called 'Salt Path (film)', that would be the defaultsort. Fuddle (talk) 12:11, 27 July 2024 (UTC)Reply

"People from ____ County" categories

edit

Is use of a "People from ____ County" appropriate when the county is not specifically mentioned in the article? For example, "People from San Luis Obispo County, California" is among the categories for James Griffith. As far as I can see, San Luis Obispo County is not mentioned in the article. Is this an exception to WP:CATV? Eddie Blick (talk) 00:08, 4 August 2024 (UTC)Reply

In this case, Griffin died in San Luis Obispo County. There are is no consensus as to whether a death location constitutes being "from" a place or not.--User:Namiba 13:35, 4 August 2024 (UTC)Reply
Categorization is supposed to be based on defining attributes … if the article text does not even mention a category’s attribute (in this case being “from” a specific place) I don’t see how the attribute can be considered defining for that topic/subject. Blueboar (talk) 14:55, 4 August 2024 (UTC)Reply
The article does mention Avila Beach, California as the place of death. It's in San Luis Obispo County, but it's too small to have its own People from… category. - Eureka Lott 15:09, 4 August 2024 (UTC)Reply
We do have Category:Deaths by location and its many subcategories. To give just two examples: Roland Ratzenberger and Ayrton Senna are both in Category:Sport deaths in Italy but neither could be remotely considered as being "from" Italy. --Redrose64 🌹 (talk) 18:39, 4 August 2024 (UTC)Reply
Thank you, @Namiba, @Blueboar, and @EurekaLott. I appreciate your comments. Eddie Blick (talk) 19:07, 4 August 2024 (UTC)Reply
Avila Beach is also too small to have significant medical facilities. So in this case since the article says he died of cancer there, one can reasonably infer that he lived there in his later years. Whether that's enough to be defining for a category is a different question. —David Eppstein (talk) 19:42, 4 August 2024 (UTC)Reply
Not necessarily, some people with cancer die in the attempt to complete the Things To Do Before I Die. Maybe he always wanted to visit Avila Beach, and having finally made it there, realised there was nothing else worth visiting. We just don't know. --Redrose64 🌹 (talk) 20:59, 4 August 2024 (UTC)Reply
It's a nice place to visit; I've been there more than once. But it seems an odd choice for a bucket list. Nevertheless, as you say, we should not operate on guesswork. —David Eppstein (talk) 22:40, 4 August 2024 (UTC)Reply
  • Our guidelines on categorization specifically say that people should not be categorized by place of death. We have not well enforced this rule. We allow say Deaths from lung cancer in California because deaths from lung cancer itself is too large. We do not however place biographical articles in deaths in California, that is a container category for articles on types of death. Just because you died somewhere does not mean you are from there. However to address the actual question, I would say if we have statements in the article that say that someone is "from Avila Beach" either explicitly, or implicitly based on other things said there (but place of death alone is not enough, lots of people die on trips, short visits, in a hospital in a place they never lived etc.) then if we do not have a category for Avila Beach, we can place them in the category for the county. Although you could solve all that issue by first editing the article to explicitly state that Avila Beach is in San Luis Obispo County. We have to watch that we only put people in these categories where they make sense. We would not place someone in the San Luis Obispo County category if they died or moved away years or decades before the county came to be. Yet in the case of German states formed after World War II we have people from the 18th-century being placed in categories based on these states. Such things ought not to be.John Pack Lambert (talk) 13:52, 6 August 2024 (UTC)Reply
In a related question, should Burials at X cemetery be in City X or people from City X? User:Alansohn and I have differing views on this. I contend that Category:Burials at Hackensack Cemetery should be in Category:People from Hackensack, New Jersey while the aforementioned editor contends that it should be in Category:Hackensack, New Jersey. I contend that a category with people is most useful as a subcategory of a people-based category. Moreover, being buried in a specific place for all of eternity does, in fact, demonstrate the kind of long-lasting connection one might expect from those in the from category. What do you think?--User:Namiba 18:29, 13 August 2024 (UTC)Reply
So… one of my ancestors purchased a 12 grave plot in a cemetary located a few towns away from where he lived. It was a brand new cemetary at the time he purchased it, and he thought it was nicer than the old one in his town.
Since the family already owned the plot, his children and several grandchildren were also buried there - even some who had moved to other parts of the country. The point being… none of them ever lived in the town where this cemetery is located. It’s simply where they were buried. I don’t think it accurate to say they are (in any way) “from” this town. Blueboar (talk) 21:39, 13 August 2024 (UTC)Reply
Agreed. A person's place of burial doesn't necessarily correspond with where they lived. There's a separate category tree for burial locations. In this instance, Category:Burials at Hackensack Cemetery should be placed in Category:Burials in Bergen County, New Jersey. - Eureka Lott 21:58, 13 August 2024 (UTC)Reply
My POV is that the "from" categories also include people who have a close connection to a certain place. Is not the permanent location of your body for all of history a close connection? Moreover, isn't it more likely that a user looking to find where someone is buried to look in the "from" category alongside other biographies?--User:Namiba 15:02, 14 August 2024 (UTC)Reply
I certainly don’t think so. My family continues to have a tie to the town where my relatives lived… but not to the town where they are buried.
Then consider soldiers that are buried on battlefields or national cemeteries. A WW II soldier from the California might be buried in Normandy, France… a Union soldier from Maine might be buried in Arlington. Those soldiers were not “from” France or Virginia… they were “from” California or Maine.
The fact is, people can be buried in places they are not “from”. Blueboar (talk) 17:05, 14 August 2024 (UTC)Reply

<Language> words and phrases categories

edit

As far as I understand Category:Words and phrases by language must contain articles about words, such as Troika (European group), Obrazovanshchina, Gaoli bangzi, not on subjects which happen to me named by a non-English word such as Bezirk or barangay or Balalaika. Sometimes the classification is not immediately evident, e.g., for Atel (slang). But I'd like the issue to be clarified, becaause it itches me to launch a massive decategorization in Category:Russian words and phrases. - Altenmann >talk 01:44, 14 August 2024 (UTC)Reply

What is the correct head category for Burials in X cemetery?

edit

Based on the above responses, it is clear that the consensus that burials in X cemetery do not belong in "people from" categories. However, the better question is: what are the best head categories? I propose that it should be as follows: Category:Cemeteries in City and Category:History of City. If there is no Category:Cemeteries in City, then Category:City will suffice. Does this make sense?--User:Namiba 14:09, 16 August 2024 (UTC)Reply

@Namiba: We already have an extensive tree under Category:Burials by location. What's the problem with that? --Redrose64 🌹 (talk) 21:12, 16 August 2024 (UTC)Reply
Perhaps I wasn't clear in my initial post. Let me use a concrete example. There is Category:Burials at Liiva Cemetery. It is only in Category:Burials in Estonia by cemetery for the moment. My question is about adding a second category for the specific location within Estonia. Should it also be in Category:Tallinn, Category:History of Tallinn or Category:Cemeteries in Tallinn?--User:Namiba 23:48, 16 August 2024 (UTC)Reply
We want to avoid categories of one item, so I would go with “Cemeteries in X location” if the city/town has more than one cemetery … but bump it up to the next regional level if the city/town only has a single cemetery. Blueboar (talk) 01:02, 17 August 2024 (UTC)Reply
We have two overlapping trees at Category:Cemeteries by country and Category:Cemeteries by continent. --Redrose64 🌹 (talk) 08:26, 17 August 2024 (UTC)Reply
I really think most bu continent categories are unneeded. Just break down by country and leave small counties in the general category.John Pack Lambert (talk) 00:58, 29 August 2024 (UTC)Reply

Should we make sure a false Category has other entries before removing it

edit

If I come across a person who lived 1712-1798 and find that they are in Category:19th-century French merchants (which oddly enough does not exist) am I justified in removing it even if it is the only article in that category, or do I have to instead leave it there and file a formal petition to delete the category. Either way this illustrates that we need to come up with much better rules against overly narrow intersection categories, because the uncontroversial edit I outline above should not require such an extensive process to accomplish.John Pack Lambert (talk) 20:37, 1 September 2024 (UTC)Reply

What is the point of by century categories

edit

I was under the impression that by century categories were a convenient way to subdivide other categories that are to large. I have seen people argue that x-nationality by century categories are container categories. I think we should formalize this view. By century categories are supposed to be limited to the intersection of something defining amd that century. I think therefore we need to define what that definingness is. I think we need to limit all by century categories that have biograohical articles to being the intersection of an occupation, broadly defined, and that century. Or sub-cats thereof. The general 18th-century people category and any non-occupational sub-cats should only function as container categories for by century categories that are occuaptional in a broad sense.John Pack Lambert (talk) 20:42, 1 September 2024 (UTC)Reply

In our frenzy to subdivide large categories we seem to have moved far far from the principles that categories should be defining and that separate defining properties should be listed as separate categories. Instead, I have to somehow remember or look up which combinations of subsets of characteristics are named as categories and that are not covered by even more specific combinations of subsets, and then which of those combinations are diffusing and which are not. So that if, for instance, someone has American and Dutch nationality, works as a mathematician, specializes in geometry, of African descent, is a women, and has been active in both the 20th and 21st centuries, I need to remember that American women mathematicians are subcategorized by century but Dutch women mathematicians are not, that American mathematicians are subcategorized by specialty but Dutch mathematicians are not, that we have categories for African-American mathematicians and African-American women mathematicians (non-diffusing in their mathematician parent categories but not in their African-American or African-American women parent categories respectively), that Dutch people of African descent are categorized by place of descent but not by occupation, etc. etc. Most of these are non-defining intersections of defining characteristics. I end up with a random assortment of a dozen categories much longer than the single sentence of my description above. I don't even think it's possible to automate this process of going from characteristics to categories because most of our categories are not annotated as to whether they are pure intersections of their parent categories or whether they add extra information beyond that. Getting it right is so tedious and annoying that I seldom can or do. And then once all our articles are categorized in this way it becomes difficult or impossible to list or even count the articles in a parent category because they are all broken out into a complicated hierarchy of overlapping subcategories. I think these intersections are broken and I think we should stop using them. —David Eppstein (talk) 21:07, 1 September 2024 (UTC)Reply

Excessive sub-dividing of ambassador categories

edit

We have a huge number of ambassador catrgories that only have 1 article. Thus seems to stem from a view that it is reasonable to have every one of the in excess of 20,000 possible ambassador categories at least as long as any example has at least 1 article. The potential is probably worse than that, since there are in addition to all the existing countries, past countries like the Ottoman Empire, Yugoslavia, the Soviet Union and so on each of which both sent and received Ambassadors. So Ambassadors to Prussia (which really should be Ambassadors to the Kingdom of Prussia) has 4 sub-cats with 2 or less articles. Size is not the only issue though. Many Ambassadors had multiple assignments over time, and in some cases the same ambassador served as the agent of their country to several countries all at the same time. I think most of the current problem will require CfD nominations, but it would be nice if we could stop the creation of more small categories now. Especially since ambassadors are not default notable so the fact Country X has sent multiple Ambassadors to country Y in no way shows that they will all be noable.John Pack Lambert (talk) 21:32, 1 September 2024 (UTC)Reply

Also should ambassador categories A-hold only those who held the title Ambassador. B-include every formally appointed top representativd of countey Y to country X regardless of title. C-cover every person who acted as the top representative of ciuntry Y to country X. I think B is what we should do, but C is the current reality. I think very short term appointees, consuls who were at the time the top representative and other things not even close to regular appointments should nit be included, but on the ither hand since we categorize bt shared name we should not worry if the title was anbassador, minister etc as long as they were the top representative of nation X in nation y.John Pack Lambert (talk) 21:36, 1 September 2024 (UTC)Reply