Wikidata:Property proposal/HAL institution ID
HAL structure ID edit
Originally proposed at Wikidata:Property proposal/Organization
Description | numerical code for a research institution as stored in HAL author affiliations |
---|---|
Represents | HAL (Q3144107) |
Data type | External identifier |
Domain | organization (Q43229) |
Allowed values | [1-9]\d+ |
Example 1 | Institute of Physics of 2 Infinites Lyon (Q3152062) → 2 |
Example 2 | Aimé Cotton Laboratory (Q3214360) → 8 |
Example 3 | Center of Theoretical Physics (Q16008925) → 10 |
Source | https://api.archives-ouvertes.fr/docs/ref/resource/structure |
Number of IDs in source | 35992 |
Expected completeness | eventually complete (Q21873974) |
Formatter URL | https://aurehal.archives-ouvertes.fr/structure/$1 |
See also | HAL author ID (P4450), French national research structure ID (P3016) |
Motivation edit
Yet another research institution identifier. It is linked to other identifiers: French national research structure ID (P3016), ISNI (P213) and IdRef ID (P269), so it should not be too hard to import this ID using these connections.
Note that records in this database have a valid_s
field indicating if the record is VALID
, INCOMING
or OLD
. This is because records in this database are added on the fly when users input institutions that do not exist in the database, which therefore contains a fair amount of duplicates. I propose to restrict this property to the VALID
records (35,992 out of the 315,208 records stored), which are curated by the service and should therefore be more reliable. The proposed formatter URL only works for such records. − Pintoch (talk) 07:29, 3 May 2019 (UTC)
Notified participants of WikiProject Universities
Discussion edit
- Support ArthurPSmith (talk) 16:44, 3 May 2019 (UTC)
- Support (disclaimer: involved in the same project as Pintoch). That said, I wouldn't restrict to valid records, as OLD records may refer to ancient institutions which may well exist as Wikidata entities --A3nm (talk) 18:33, 3 May 2019 (UTC)
- My understanding was that records were marked as OLD when they were found to be duplicates of others - this is at least the case for a few I have been looking at. To be honest I have not found any documentation about this, so I welcome any insights! − Pintoch (talk) 19:04, 3 May 2019 (UTC)
- Could we add a "end time" ("p582") as a qualifier or something like that ? MaximeBUSHS (talk) 12:23, 6 July 2023 (UTC)
- My understanding was that records were marked as OLD when they were found to be duplicates of others - this is at least the case for a few I have been looking at. To be honest I have not found any documentation about this, so I welcome any insights! − Pintoch (talk) 19:04, 3 May 2019 (UTC)
- Comment I'm not sure that an API call with a JSON output is a good catch for the formatter URL. Linking to something like AURéHAL would be better don't you think? @Pintoch: --Nono314 (talk) 17:42, 8 May 2019 (UTC)
- @Nono314: that's much better of course, thanks! I did not know this interface existed. I have changed the formatter URL accordingly. − Pintoch (talk) 08:02, 9 May 2019 (UTC)
- Support Arpyia (talk) 08:37, 11 May 2019 (UTC)
Done @Arpyia, Pintoch, Nono314, A3nm, ArthurPSmith: --Kristbaum (talk) 12:17, 23 May 2019 (UTC)