Wikidata:Property proposal/terminology

From Wikidata
Jump to navigation Jump to search

field of usage

[edit]

Originally proposed at Wikidata:Property proposal/Lexemes

   Done: field of usage (P9488) (Talk and documentation)
Descriptionspecialized area or terminology where this sense is used
Representsterminology (Q8380731)
Data typeItem
Domainsense
Allowed valuesterminology (Q8380731)
Example 1ipso facto (L227969), sense 3 → jurisprudence (Q4932206)
Example 2window (L3327), sense 1 → computer science (Q21198)
Example 3note (L4316), sense 2 → music (Q638)
Planned useusing on many lexemes
See alsolanguage style (P6191)

Motivación

[edit]

I've suggested this on Wikidata:Property_proposal/part of terminology, but I didn't proposed it formally. --Tinker Bell ★ ♥ 05:27, 19 March 2020 (UTC)[reply]

Discussion

[edit]
 Support Yes I think lexemes is the place for this, and I guess we don't currently have a property that's quite right for it. ArthurPSmith (talk) 18:20, 19 March 2020 (UTC)[reply]
 Neutral I approve of the idea, but I don't like the requirement of a terminology item. If only since even within computing or legal terminology subfields may apply different definitions... Circeus (talk) 17:41, 24 March 2020 (UTC)[reply]
@Circeus: do you have an example? --Tinker Bell ★ ♥ 07:12, 29 March 2020 (UTC)[reply]
Leaning towards  Support Clearly  Support. I'm just wondering: are we sure we couldn't use other properties (maybe facet of (P1269)?) and if terminology (Q1725664) is probably not the right item, here it's "terminology" as in « The set of terms actually used in any business, art, science, or the like; nomenclature; technical terms. », the second sense en:wikt:terminology while terminology (Q1725664) is more about the first sens of the same entry « The doctrine of terms ». A new item is needed (and would useful for other items like nomenclature (Q863247)). Cheers, VIGNERON (talk) 17:39, 30 March 2020 (UTC)[reply]
@VIGNERON: I've corrected the item. --Tinker Bell ★ ♥ 23:34, 30 March 2020 (UTC)[reply]
Thanks (somehow I missed this item, no need for a new item then). And what about my first question? Cheers, VIGNERON (talk) 07:35, 31 March 2020 (UTC)[reply]
@VIGNERON: Before opening this proposal, I considered using part of (P361). But I realized a sense doesn't belongs to a terminology, but its concept, so I'm pretty convinced this new property is necessary. I don't see how facet of (P1269) can be used here. --Tinker Bell ★ ♥ 08:41, 1 April 2020 (UTC)[reply]
TBH, I'm not sure either, I just wanted to be sure you consider other options. Thanks. Cheers, VIGNERON (talk) 08:51, 1 April 2020 (UTC)[reply]
 Comment I support adding information about field of use. Just a question - is it necesary to limit the values to terminology (Q8380731)? Would be linking to mathematics (Q395), architecture (Q12271) or card game (Q142714) not sufficient enough? --Lexicolover (talk) 21:27, 31 March 2020 (UTC)[reply]
@Lexicolover: I think it's useful to restrict the allowed values to a set of topics that can be identified and referred easily. For example, the sense about computer mouse (Q7987) could use computing (Q179310) or computer (Q68)? I would prefer using computing terminology (Q3457057) because it's clearer: the topic isn't too specific, nor too broad. --Tinker Bell ★ ♥ 07:18, 1 April 2020 (UTC)[reply]
@Tinker Bell:  Disagree It seems you have a need to have an identifier for "lists of words used in a domain". This list can be generated at any time with SPARQL. There are tens of thousands of domains, and so with your opinion, there would be a need to maintain tens of thousands of such lists or identifiers for such topics (really just compound concepts. computing + terminology), like "bicycle terminology" "poker terminology" "browser terminology"... Since "lexeme" actually means "a lexical unit of language", then we are already in the realm of "terminology" and "words". There's no need to create compound concepts.
I understand this proposed property to be the equivalent of symbols such as ⚒, ⚓, ⚔, ⚕, ⚖, ⚗, ⚘, ⚙, or corresponding abbreviated terms "phil", "polit" or whatever when no symbol is available, that are typically used in conventional printed dictionaries to indicate field of usage. I agree that it shouldn't be limited to fields that have a "terminology" item, but I also wonder whether it's meaningful to allow thousands of different fields and subfields. To take an example, would one sense of the lexeme "check" be marked as a term in chess (Q718) specifically, or would it be sufficient to mark it as game (Q11410) term? I see nothing wrong with having multiple senses belonging to the same field, say for entirely different games. And a "mechanical device" mouse would still be distinct from a "zoological" mouse, even if somewhat broader in scope than a "computer" mouse. --SM5POR (talk) 13:49, 9 August 2020 (UTC)[reply]

 Comment I was looking around for this today, to mark the domains "business" "finance" on S2 sense for Lexeme acceptance while reviewing the same sense on Wiktionary. "field of study" is much more appropriate and allows both broad as well as precise data enrichment. "bow" on a violin -> domain: music, bowed string instrument, violin and similarly a "bow" on a boat -> domain: vessel, marine. "domain" within Wikidata seems to be conceptually wrapped up in "field of study" which seems reasonable, since any "class" can be a field of study. But I would be OK with usage of facet of (P1269) statements on Senses as others mention which seems to be a common alternative way for many Wikidata editors to say "field of study" and where it is typically used as "a domain or field of study" predicate. It seems we already have this ability now to apply any statement to an individual Sense. So I can already apply "facet of" or "field of study" to any Sense. So, I don't see a need for this property to be introduced. Instead, we should just improve the Lexeme documentation with tips on how to apply "field of study" or "facet of" Thadguidry (talk) 16:19, 22 April 2020 (UTC)[reply]

  • @Thadguidry: Using studied in (P2579) would be fine for items, but not for lexemes. For example, music terminology (Q77655668) has statement studied in (P2579) music theory (Q193544). We can say that a terminology itself is studied by some field or discipline, but a lexeme by itself is not studied by the field it belongs, but by lexicology (Q178433). About using facet of (P1269), its description says "item that offers a broader perspective on the same topic". I can't see how a lexeme fullfills that description. --Tinker Bell ★ ♥ 01:25, 25 April 2020 (UTC)[reply]
  • @Tinker Bell: Hi! Sure, there are lots of relationships that can be made between "words" (lexemes) and "concepts" (items). I guess what I'm trying to lead up to is this... What does the community feel are the most important relationships between the two? Do you think its more important to have links between the two that describe a relationship that "word senses" are studied, used, or both in a domain ? About the facet of (P1269) , that could be broaden to be used on both items and lexemes, Correct? I feel that sensible reuse of properties between the two, i.e. I don't think there is anything inherently wrong with saying that a "word sense" can be a facet of some domain. And I think that is the core of this proposal in trying to offer a way that a "word sense" has some relationship to some domain. And the question that everyone has I think is this... What is the best way to say that some "word sense" is used in some domain? Can we reuse existing properties and make small changes to allow that? I think so, for instance, it might be decided and easier to expand used by (P1535) to be "used by/in" to allow a triple pattern of "bow" -> "used by/in" -> "music" ? or would that really confuse things and we need to create new properties? Thadguidry (talk) 02:36, 25 April 2020 (UTC)[reply]
  • I like the idea behind the property, but I don't like the way it's currently proposed. The name is not intuitive to me and it also feels strange to me to link to a terminology item rather than to the field itself. And if we're not linking directly to the field, why not just have "instance of computing term" and not even bother with a new property? My suggestion would be "field of usage". - Nikki (talk) 12:19, 4 January 2021 (UTC)[reply]
  • I agree now with Nikki, in that we do need this new property, but that it could best be labeled as "field of usage" and linked to the field instead of a terminology item. I've changed my previous Disagree with Comment. Once we change the label and link to the field, then I'd be happy to support this. --Thadguidry (talk) 16:01, 23 April 2021 (UTC)[reply]
  • @Thadguidry, Tinker Bell, Circeus, VIGNERON: @Lexicolover, SM5POR, Nikki: I have modified the proposal as suggested above (I think) - please express your support or otherwise for this revised version, or revert this change if that's not what is wanted! Note the Spanish label and description should be adjusted also. ArthurPSmith (talk) 17:35, 23 April 2021 (UTC)[reply]
  • @Thadguidry, Tinker Bell, Circeus, VIGNERON: @Lexicolover, SM5POR, ArthurPSmith: ✓ Done - Nikki (talk) 19:24, 26 April 2021 (UTC)[reply]
Retrieved from "https://www.wikidata.org/w/index.php?title=Wikidata:Property_proposal/terminology&oldid=1409167289"