Wikidata:Property proposal/parent superproperty
parent edit
Originally proposed at Wikidata:Property proposal/Person
| Description | parent of the subject |
|---|---|
| Represents | parent (Q7566) |
| Data type | Item |
| Domain | persons |
| Example | Yog-Sothoth (Q621413) → The Nameless Mist (Q10575312) |
| See also |
Motivation edit
I propose this property as a superproperty (superclass) for father (P22) and mother (P25). It can be linked with subproperty of (P1647) to those two other properties. I explicitly do not propose the removal or transfer of any values currently stored in the properties father (P22) and mother (P25).
- This property will make it easier to write queries for requesting information about all parents of certain people together, because you can filter for any subproperty of this property.
- It can be linked to external properties, such as schema.org/parent.
- And in the exceptional and rare case, when the gender of a real or fictional being is unknown or neither male nor female, the parental relationship could be added using this property, and be queried together with all other parents.
Of course I can't add examples for my first and second reasons for this property, but that does not matter, the main point of this property is making queries easier. (But nevertheless, I did find an example for my third reason. Feel free to add more examples.) Robin van der Vliet (talk) (contribs) 00:00, 25 August 2020 (UTC)
Discussion edit
Support Robin van der Vliet (talk) (contribs) 00:00, 25 August 2020 (UTC)- Support That seems reasonable, as it would align this property with most others, most prominently child (P40) and spouse (P26). It's also repetitive, as gender is recorded in its own statement. And now I'm wondering why so much effort is spent on categorising given names into male and female? But somehow I sense the vague notion that the best course of action here is to quickly think about somethings else and not mention it again --Matthias Winkelmann (talk) 00:23, 25 August 2020 (UTC)
- @Matthias Winkelmann: I think the main reason for it is that many languages (not English) have grammatical structures that differ depending on the gender of the subject of a phrase. ArthurPSmith (talk) 22:03, 1 September 2020 (UTC)
- Support NMaia (talk) 02:00, 25 August 2020 (UTC)
Comment I don’t see how this will make querying easier, at least not with regards to the “subproperty” relationship. SPARQL has no syntax to use this as far as I’m aware, and some things even become completely impossible when you introduce that kind of indirection; for instance, while it’s possible to query for “X is descendant of Y” using something like X (wdt:P22|wdt:P25)+ Y, this is not possible if you want to get the property in question via a subproperty of (P1647) relation, because SPARQL property paths can only contain IRIs, not variables (i. e.,(wdt:P22|wdt:P25)+is legal,(?wdt)+is not). --TweetsFactsAndQueries (talk) 08:54, 25 August 2020 (UTC)- Comment as TweetsFactsAndQueries, it's unclear what the use would be. Can we see some sample queries? We already have Q7566#P4316 so I doubt this would add anything. There are many other relationships with kinship equivalent in SPARQL at Wikidata (P4316) that lack a specific property. --- Jura 08:11, 26 August 2020 (UTC)
- Support –IagoQnsi (talk) 13:55, 29 October 2020 (UTC)
- Support --Tinker Bell ★ ♥ 20:26, 30 October 2020 (UTC)
