Wikidata:Property proposal/model lexeme

model lexeme

Originally proposed at Wikidata:Property proposal/Lexemes

   Done: model lexeme (P11464) (Talk and documentation)

Description	model lexeme entity for a paradigm or conjugation class item
Data type	Lexeme
Domain	item
Allowed values	lexeme ID
Example 1	روپاولی ۷(Q115730143)شبداولی دی مثال (L747194)
Example 2	پھٹوکِر صفت(Q115557831)شبداولی دی مثالپھِٹوکَر (L742540)
Example 3	ساگ گوشت ناؤںਸਾਗ/ساگ (L697738){{{3}}}
Planned use	Add model lexemes to paradigm class items.
See also	model item (P5869)
Single-value constraint	yes

Motivation

We have model item (P5869) to document examples of best practices within a class of items, and it would be useful to have a similar property to link to an example of a lexeme from an item used as a value in conjugation classes, paradigm classes, noun classes and so on. These items tend to be specific to a language or group of languages and have names that can be quite cryptic even if acquainted with the source material they come from. (Often, sources assign these alphanumeric names like paradigm 7 or conjugation class 2a.) While we can easily get a list of lexemes which link to an item, the intent of this property proposal is just to improve the documentation for people contributing data by linking to a model example. (One which is recognisable or typical, and has references and forms pertinent to the item can be chosen by users.) Feel free to add examples from other languages to those above. -عُثمان (talk) 15:32, 17 December 2022 (UTC)

Discussion

•  Support -wd-Ryan (Talk/Edits) 04:17, 20 December 2022 (UTC)
  • @عُثمان, Wd-Ryan:--GZWDer (talk) 02:16, 5 January 2023 (UTC)

    @Wd-Ryan, @GZWDer: Nice, you beat me to it!

    I had a slightly different proposal in mind; maybe it's not too late to integrate our ideas with each other? You may find my notes at User:SM5POR/Languages#Model property proposals. My key points:

    • While the typical use would be to attach the model lexeme property to a part of speech or other lexical category, I have also considered subject items such as phonetics, stylistic devices, etymology, language history etc, to illustrate how to document these focal aspects.
    • I would suggest language of work or name (P407) as a recommended (or even mandatory) qualifier to the model lexeme statement, rather than assigning it to a language item only. Lexemes for different parts of speech will have very different characteristics, so it will be hard to pick out a "typical Russian lexeme" for instance.
    • I have considered also model sense and model form with partly similar item types, in order to locate sophisticated usage of individual senses or forms. Finding entire lexemes that are fully developed in every detail (for every language) may be difficult. However, we may put those proposals on hold until we have experimented a bit with this model lexeme and found out how it will be most efficiently used.

    As a result of your effort, I can now focus on a few other things, which serves me fine. I thus have yet to put forward my first property proposal, and it may even become unnecessary... --SM5POR (talk) 18:57, 5 January 2023 (UTC)