Wikidata:Requests for comment/Population data model
The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.
- stale --Emu (talk) 12:27, 3 February 2024 (UTC)
| An editor has requested the community to provide input on "Population data model" via the Requests for comment (RFC) process. This is the discussion page regarding the issue.
If you have an opinion regarding this issue, feel free to comment below. Thank you! |
THIS RFC IS CLOSED. Please do NOT vote nor add comments.
This is a proposed data model to document populations/demographics/census data. Below is an example structure of the data model showing items and their statements.
Here is another current proposed data model that I find could create an absurd number of statements on a single item: Wikidata:Property proposal/subpopulation 2. This data model is much more easy to browse and use.
United States of America
has populationhumans in the United States
humans in the United States
instance of (P31)group of humans (Q16334295)
number of parts of this work (P2635)x
has groupCatholics in the United States
Catholics in the United States
instance of (P31)group of humans (Q16334295)
religion or worldview (P140)Catholicism (Q1841)
number of parts of this work (P2635)x
has groupBlack Catholics in the United States
Blacks in the United States
instance of (P31)group of humans (Q16334295)
ethnic group (P172)Black people (Q817393)
number of parts of this work (P2635)x
has groupBlack Catholics in the United States
Black Catholics in the United States
instance of (P31)group of humans (Q16334295)
ethnic group (P172)Black people (Q817393)
religion or worldview (P140)Catholicism (Q1841)
number of parts of this work (P2635)x
This data model would involve:
- Creating a "has population" property to relate a region with the group of humans (Q16334295) that live there.
- Changing the name of number of parts of this work (P2635) to "number of parts" and expanding its scope beyond works to all types of groups.
- Creating a "has group" property to relate a group of things with the possible groupings of its parts. This property should only be used to relate a subgroup with its closest-possible higher-classified group. For example, "humans in the United States" item should not have has groupBlack Catholics in the United States but "Catholics in the United States" and "Blacks in the United States" should. The exception for this is if no item for a higher class group has not been created yet.
- Groups should only be created when they are notable and sourced. They should also not combine other population groups together (queries should).
- All population groups will be instance of (P31)group of humans (Q16334295) to ensure data model consistency and they will use describing properties on the groups to differentiate them.
Let me know your thoughts. Lectrician1 (talk) 20:10, 17 January 2022 (UTC)
- I have revised the proposal (use population (P1082) instead of number of parts of this work (P2635)) at Wikidata:Property proposal/subpopulation 3.--GZWDer (talk) 21:37, 17 January 2022 (UTC)
- As someone editing pages for higher education institutions, the subpopulations I would be creating under this model, if I understand correctly, would be things like "Black students at Georgetown University". Creating an item for that seems not ideal. {{u|Sdkb}} talk 08:04, 18 January 2022 (UTC)
Oppose as my comment has not been addressed. {{u|Sdkb}} talk 23:55, 26 January 2022 (UTC)- @Sdkb I don't think that's notable so it wouldn't be created... Lectrician1 (talk) 11:35, 27 January 2022 (UTC)
- An FA university article will discuss the racial composition of the student body; we should be able to model that information here, too. A scheme which requires massive creations of new items or which can't handle tasks like that is not a sufficient system. {{u|Sdkb}} talk 16:40, 27 January 2022 (UTC)
- @Sdkb Well, I'm not sure how else you're going to model it then. Having items for the groups and then defining their populations over time using statements is the best and only way we can really do this. Lectrician1 (talk) 16:10, 6 February 2022 (UTC)
- An FA university article will discuss the racial composition of the student body; we should be able to model that information here, too. A scheme which requires massive creations of new items or which can't handle tasks like that is not a sufficient system. {{u|Sdkb}} talk 16:40, 27 January 2022 (UTC)
Info I propose to close this RfC as stale after 31 January 2024. Please comment if you don’t agree. --Emu (talk) 20:35, 13 January 2024 (UTC)