Wikidata:Property proposal/homophone lexeme

homophone lexeme

Originally proposed at Wikidata:Property proposal/Lexemes

   Withdrawn

Description	lexeme with the same or very similar pronunciation as this one
Data type	Lexeme
Example 1	Laib (L493284) ⟷ Leib (L613407)
Example 2	eye (L3534) ⟷ Ei (L6682)
Example 3	交渉/こうしょう (L620305) ⟷ 工廠/こうしょう (L620306) (+~48 more)

Motivation

I would not restrict it to words of the same language since it could help to figure out how to pronounce it. should this better be used on forms? – Shisma (talk) 16:48, 27 November 2021 (UTC)

Discussion

•  Support The Japanese language has countless homophones. For example, the word "こうしょう" is said to have 48 different homophones. This property is very useful for a language like Japanese.--Afaz (talk) 05:34, 28 November 2021 (UTC)
  • Afaz: Excellent! I've added two lexemes as a last missing example: 交渉/こうしょう (L620305) ⟷ 工廠/こうしょう (L620306). Please review if they are alright 🙏😅 –Shisma (talk) 16:53, 28 November 2021 (UTC)
•  Comment Yes, this would be better to use on forms rather on the root of the lexeme. However, this has the same problem as the translation property in that if there are many homophones, all would be listed on all of them (sort of like the old interwiki system) and thus doesn't really leverage the power of linked data. I don't have a good solution for that though. Ainali (talk) 16:26, 29 November 2021 (UTC)
  • @Ainali, Afaz: It is possible, just as with translation networks, to conceive of "homophone networks", where not all one-to-one connections need to be present for the identical pronunciation between different words to be inferable. For example, the translation network of "mother" does not have any one-to-all linkages, yet every word for mother is still linked (in)directly to every other. Similarly, of the fifty-three kanji readings of wikt:こうしょう, perhaps "交渉" could just link to other lexemes containing "交" or "渉", rather than to all of the fifty-two others listed on that page. Mahir256 (talk) 05:28, 1 December 2021 (UTC)

  here's an example where forms are homophone but not the lemma: rasieren (L450732) (F4: verb, present tense: rasiert) rasiert (L622301) (F1: adjective, positive: rasiert) – Loominade (talk) 08:18, 8 December 2021 (UTC)

•  Support Could it be used on lexemes with the understanding that it means at least one of the forms of the subject lexeme has the same pronunciation as one of the forms on the value lexeme? ArthurPSmith (talk) 18:37, 29 November 2021 (UTC)
•  Comment: What do we do about words that are homophonous in some regional pronunciations but not others, e.g. the cot–caught merger (Q28401088)? Are there existing properties that could be used as qualifiers? ⁓ Pelagic ( messages ) 14:59, 8 December 2021 (UTC)

  pronunciation variety (P5237) could be used as a qualifier perhaps? ArthurPSmith (talk) 18:16, 8 December 2021 (UTC)

@Shisma, Afaz, Ainali, ArthurPSmith, Mahir256: Please consider supporting Wikidata:Property proposal/homophone form rather than this one. Or consider if it makes sense to have both properties. --Loominade (talk) 10:34, 10 December 2021 (UTC)

• You're right. It's allways the form only. Singular Laib and Leib are homophone. Plural Laibe and Leiber are not. --Vollbracht (talk) 21:32, 4 February 2022 (UTC)

•  Support, an important property for the language.--Arbnos (talk) 23:51, 23 April 2022 (UTC)
•  Comment I do not see why it should be used for lexemes. The property proposal for homofon from form would be more appropriate. I have support the other property proposal. — Finn Årup Nielsen (fnielsen) (talk) 13:39, 8 June 2022 (UTC)
  • I have now been using homophone form (P10822) a bit and I see that there are sometimes cases where a lexeme with all its inflections or conjugations may have a homophone relationship with another lexeme and all its inflections/conjugations. In Danish, an example is lege (L42896) and leje (L48302). — Finn Årup Nielsen (fnielsen) (talk) 14:42, 17 June 2022 (UTC)

@Afaz, ArthurPSmith, Arbnos: since homophone form (P10822) exists now, which is rather precise I would withdraw this proposal. any objections? -Shisma (talk) 16:45, 1 July 2022 (UTC)

@Shisma: No objection from me. However Finn Årup Nielsen (fnielsen) maybe was having second thoughts on this? ArthurPSmith (talk) 16:55, 1 July 2022 (UTC)

• @Fnielsen, Arbnos, Vollbracht, Pelagic, ArthurPSmith, Loominade:@Ainali, Afaz, Shisma: Given that Shisma said that they want to withdraw the property I marked it as withdrawn. ChristianKl ❪✉❫ 17:46, 4 December 2022 (UTC)