Wikidata:Property proposal/Basic Formal Ontology ID

Basic Formal Ontology ID

Originally proposed at Wikidata:Property proposal/Authority control

   Not done

Description	identifier in Basic Formal Ontology (BFO), a top-level ontology developed by Barry Smith and his associates for the purposes of promoting interoperability among domain ontologies
Represents	Basic Formal Ontology (Q4866972)
Data type	External identifier
Example 1	independent continuant (Q53617489) → BFO_0000004
Example 2	entity (Q35120) → BFO_0000001
Example 3	continuant (Q103940464) → BFO_0000002
Formatter URL	http://purl.obolibrary.org/obo/BFO_$1

Motivation

Basic Formal Ontology gets used a lot in biology, so it's worth being able to reference it directly even when at the moment there aren't that many entries. ChristianKl ❪✉❫ 20:07, 6 December 2020 (UTC)

WikiProject Molecular biology has more than 50 participants and couldn't be pinged. Please post on the WikiProject's talk page instead. ChristianKl ❪✉❫ 20:09, 6 December 2020 (UTC)

Discussion

 Support +  Comment It seems reasonable. I wonder if we want all OBO Foundry ontologies to have specific properties on Wikidata. I think they are generally very useful. As a side note, one of the examples says independent continuant (Q53617489) → 0000002, but BFO_0000002 is Class: continuant, I think it was a typo or something. Best, TiagoLubiana (talk) 02:40, 7 December 2020 (UTC)

• I fixed the value. I copy paste and forgot to fill in the proper values. I do think that we likely want all the OBO Foundry ontologies to have external ID properties in Wikdiata. ChristianKl ❪✉❫ 10:45, 7 December 2020 (UTC)
•  Support --Tinker Bell ★ ♥ 01:44, 8 December 2020 (UTC)

•  Oppose Basic Formal Ontology has only 35 class terms. Linking with equivalent class (P1709) would be sufficient. Simon Cobb (User:Sic19 ; talk page) 09:48, 21 December 2020 (UTC)
  • While it has few items, having the ability to look up the definition quickly is valuable. The fact that the items are on top of the ontology also means that they are likely to be seen much more often then a lot of other items. ChristianKl ❪✉❫ 20:33, 22 December 2020 (UTC)

@ChristianKl: Yes, the ability to look up the definition is useful and it is possible to do that without creating a property. Not sure I properly understand your point about items on top of the ontology being seen more often but it does not seem to justify the creation of this property, especially as we already have a property to store exactly the same information and connect to these terms.

Also, can this proposal be placed alongside your other OBO property proposals so they can be discussed together? Succu has correctly noted that the prefix, in this instance BFO_, is part of the ID and this should not be overlooked. Although there are extant OBO ontology properties, e.g. Environment Ontology ID (P3859) and OBO Gazetteer ID (P6778), they are seldom utilised and I think it would be sensible to discuss the option of a single property (with the prefix included to identify the ontology) as an alternative to a property for each ontology. Or, if we can establish a consensus to (potentially) create properties for each ontology, I will withdraw my objections based on the number of terms. Simon Cobb (User:Sic19 ; talk page) 08:29, 23 December 2020 (UTC)

@Sic19: There are entities that have identifiers in multiple OBO ontologies. That means giving our semantics around external identifiers, a property that takes all IDs wouldn't be an external ID. When a user looks at a property it's valuable when all the external identifiers are listed together and that doesn't happen with a string property.

I did switch the ID name to BFO_.

I have created a list for all OBO ontologies at https://www.wikidata.org/wiki/Wikidata:WikiProject_Bioinformatics/OBO_Foundary_Ontologies . Most of them don't have Wikidata properties at present. While I do believe that it would be benefitial to eventually have properties for all of them, I currently don't propose to add Wikidata properties without planned usages. The ontologies I did propose were ontologies that touch items with whom I worked. ChristianKl ❪✉❫ 15:27, 23 December 2020 (UTC)

@ChristianKl: regarding identifiers in multiple ontologies, see the RDA value vocabularies ID proposal below, which implies it is possible to have an external identifier datatype property in this scenario. It also deals with the issue of some of the vocabularies being too small to require a specific property. Simon Cobb (User:Sic19 ; talk page) 16:32, 23 December 2020 (UTC)

@Cmungall: you proposed OBO Gazetteer ID (P6778), do you have plans to fill it with more data? ChristianKl ❪✉❫ 15:27, 23 December 2020 (UTC)

• @ChristianKl, TiagoLubiana, Tinker Bell, Sic19:  Not done Stale proposal. ArthurPSmith (talk) 17:54, 2 August 2021 (UTC)
• @ArthurPSmith: Since when do we set proposal with net positive votes to not done? ChristianKl ❪✉❫ 18:15, 2 August 2021 (UTC)
  • @ChristianKl: I look at proposals with more than 200 days since the last comment; if it's one I haven't been involved in discussion on and it seems to have been inconclusive I close it like this. Feel free to reopen if there's something new to say here. ArthurPSmith (talk) 18:16, 2 August 2021 (UTC)
    • @ArthurPSmith: How is a clear majority ( 3 people including the creator in favor and one against) inconclusive? ChristianKl ❪✉❫ 13:37, 3 August 2021 (UTC)
      • It's not clear the other two supporting comments were aware of Simon Cobb's concern about the small number of likely uses, which is a concern frequently raised with other similar property proposals. I pinged all involved when I set the "not done" so they should be able to contribute further to clarify if needed. ArthurPSmith (talk) 14:59, 3 August 2021 (UTC)