User talk:Lucas Werkmeister/P642 considered harmful

Discussion on “main subject”

Note: the first two comments here were originally on the essay page itself; I’ve moved them here. --Lucas Werkmeister (talk) 19:03, 28 April 2021 (UTC)

One place where I think the use of of (P642) can be particularly useful: to narrow down a value for main subject (P921). Consider an article such as Bald eagle mortality and chlorinated hydrocarbon contaminants in livers from British Columbia, Canada, 1989-1994 (Q79952831), which is about mortality of bald eagles in British Columbia. I would prefer to give main subject (P921) as Bald Eagle (Q127216) qualified by of (P642) British Columbia (Q1974) rather than as two separate subject statements. It's misleading to say that this is about bald eagle and British Columbia. It's much more accurate to be able to say that it is about bald eagle of/in British Columbia. UWashPrincipalCataloger (talk) 02:05, 20 April 2021 (UTC)

@UWashPrincipalCataloger: I think in such cases, it would be better to have a new item for "bald eagle in British Columbia" and use that as main subject. --Daniel Mietchen (talk) 17:05, 26 April 2021 (UTC)

@Daniel Mietchen: Wow, that would lead to an exponential growth of items created for topics that have a geographic aspect. And to topics that can be narrowed down by a further topical aspect as well. Do we really want to go that direction? 100s of items for bald eagle in any particular place. It seems very inefficient when of (P642) or some other qualifier (perhaps a new one for geographical aspect of main subject?) could do the trick more simply. UWashPrincipalCataloger (talk) 19:16, 28 April 2021 (UTC)

I think the problem is not the use of a detailing qualifier but the use of of (P642) for it. Better here would be location (P276). --SCIdude (talk) 16:23, 2 May 2021 (UTC)

Discussion on project chat

Please see Wikidata:Project_chat#New_essay:_of_(P642)_considered_harmful. --- Jura 14:40, 25 April 2021 (UTC)

Mostly harmless? Archived at Wikidata:Project_chat/Archive/2021/04#New_essay:_of_(P642)_considered_harmful. --- Jura 00:47, 2 January 2022 (UTC)

Use of "OF" in positions

I would be interested if P642 disappeared how you would handle the multitude position allocations through history. "Steward of (OFFICEHOLDER)", "Physician of/to (OFFICEHOLDER)", "president of", "secretary of (ORGANISATION)", "private secretary of" and then we have the religious positions that proliferate. When migrating biographical detail to WD the "OF" statement is very useful rather than having to create each and every position, especially if there is translations where you don't know the language, the more common/generic is available, easy, and practical. Thanks.  — billinghurst sDrewth 01:31, 29 April 2021 (UTC)

I think maybe it is possible to check the positions afterwards and then try to understand what the positions mean in another language and create a new item or add an existing one and removing the statement with the qualifier. Using qualifiers in SPARQL needs an further understanding and it is not easy. A more detailed version is better but to create detailed information about specific topics needs time and sources from where it is possible to understand the topics and Ideas how to put it into statements. I would create a item for every relevant position in the past with an reliable source exists. So at the moment I think that for the next months or maybe years there stay statements with P642 and I support avoiding the use of this property and I plan to go to a library to find there books from where I get information about historic positions so that I can add them in Wikidata. --Hogü-456 (talk) 19:18, 29 April 2021 (UTC)

Using of with predicate for (P9970)

See Property talk:P9970. No good solutions have emerged as of this writing, if anyone can think of a suitable substitute, feel free to share :)--So9q (talk) 09:07, 5 November 2021 (UTC)

"Collection of"

I'm persuaded by this essay and want to remove my lazy use of of (P642) from the items I've created. I'm not sure how to substitute Nasser D. Khalili Collection of Japanese Art (Q66455644) instance of (P31) collection (Q2668072) of (P642) Japanese art (Q603399). Clearly I should get rid of the qualifier and have a separate property linking Nasser D. Khalili Collection of Japanese Art (Q66455644) to Japanese art (Q603399), but not sure of the predicate. Maybe field of work (P101)? But collection (Q2668072) does not count as a "person or organisation". MartinPoulter (talk) 11:14, 26 November 2021 (UTC)

@MartinPoulter Hm, that’s a good question. A few museums seem to have main subject (P921), but not many of the ones I checked, and in this case I’d say it’s dubious. genre (P136) doesn’t feel inappropriate to me, but from the description it doesn’t sound like it would be allowed. I looked through Wikidata property related to art (Q27918607) properties and nothing jumped out at me either. Lucas Werkmeister (talk) 19:05, 26 November 2021 (UTC)

how about differentia in subclass

The first comment was originally added on the essay page itself; I’ve moved it here. --Lucas Werkmeister (talk) 13:41, 11 December 2021 (UTC)

I agree with most of your objections.

But how about using it in subclass claims to capture the differentia (in Aristotle's sense)? I find the use of "of" and "use" indispensable for such cases --Vladimir Alexiev (talk) 13:37, 11 December 2021 (UTC)

applies to part (P518)

As mentioned at project chat, I find P642 mostly harmless, as it generally can be ignored.

applies to part (P518) might be worth looking into instead. --- Jura 00:51, 2 January 2022 (UTC)

Phase-out usage

@Lucas Werkmeister Could we "phase-out" usage of the property by adding a constraint warning everywhere it's used? It would ask to remove it "in favor of a more-specific property". This should help people move the values to properties that are more-specific, have people propose new more-specific properties covering the instances your essay outlines, and eventually get to a point where of (P642) is not used anymore and we could delete it. Lectrician1 (talk) 03:10, 6 January 2022 (UTC)