Wikidata:Property proposal/mode of derivation

From Wikidata
Jump to navigation Jump to search

mode of derivation[edit]

Originally proposed at Wikidata:Property proposal/Lexemes

Description(qualifier) The way that a term is derived
Data typeItem
Domainlexeme/form/sense
Example 1edit (L175)back-formation (Q989162)
Example 2Wikipedia (L6596)blend word (Q287903)
Example 3flea market (L6598)calque (Q204826)
Example 4黑客 (L6599)phono-semantic matching (Q1128781)
Example 5haha (L6600)onomatopoeia (Q170239) (Note: I don't know whether this should handled by this property)

Motivation

Note: I don't know whether this should be used as a claim or a qualifier. GZWDer (talk) 12:48, 14 July 2018 (UTC)

Discussion

  • Pictogram voting comment.svg Comment I think this could work as a qualifier and I would call it "method of formation". It could simplify the ethymology analysis for future tools if we replaced current ethymology properties with one property "word formed from" and "method of formation" qualifier. --Lexicolover (talk) 21:49, 13 September 2018 (UTC)
  • @GZWDer: I have updated proposition with "qualifier" keyword. KaMan (talk) 11:07, 14 September 2018 (UTC)
  • Symbol support vote.svg Support. Would "borrowing" and "inheritance" also be one of the possible options? This would be useful for derivations from a lexeme in another language, and allows to specify that French apprendre and appréhender both came from Latin apprehendere, but the former was inherited (passed in an unbroken chain from generation to generation) while the latter was borrowed. Rua (talk) 11:30, 14 September 2018 (UTC)
    @Rua: set of possible values is not limited. I usually use diminutive (Q108709). You can select whatever can be created as Q-item. I just created no label (Q56611977) and inheritance (Q56611986) for you and it can be selected. You can adjust their labels and descriptions in preffered way. KaMan (talk) 11:58, 14 September 2018 (UTC)
    @KaMan: Thank you! Is "diminutive" really a mode of derivation, though? I'd say it's more what you end up with after the derivation, not how you derive it. At least in the languages I've seen, creating diminutives involves some other derivational process like suffixation. Rua (talk) 12:10, 14 September 2018 (UTC)
    @Rua: I don't know your language but in my language "zdrobnienie" is both process and a result. But generally you are right, it should be splitted and distinguished in the future. KaMan (talk) 12:18, 14 September 2018 (UTC)
    @KaMan: I just saw there's already loanword (Q103808). That should probably be used rather than no label (Q56611977) right? Rua (talk) 12:25, 14 September 2018 (UTC)
    @Rua: You are right, for some unknown reason I missed this myself. I merged them. You can merge duplicates yourself following description on page Help:Merge. KaMan (talk) 12:35, 14 September 2018 (UTC)
  • Pictogram voting comment.svg Comment I noticed that blend word (Q287903) is given in one of the examples. Blends are always a combination of multiple lexemes, so isn't this a case for combines (P5238)? Rua (talk) 12:10, 14 September 2018 (UTC)
    I dont think so. Compounds consist of two or more stems while blended words have origin in multiple words but their stems are not all retained. --Lexicolover (talk) 18:21, 14 September 2018 (UTC)
    I agree with you there, but I don't think combines (P5238) refers specifically to compounding, it's used for affixation as well. And even if it does, the things we call "compounding" aren't always straightforward cases where the stems are retained. Just look at sámegiella (L558). Rua (talk) 19:08, 14 September 2018 (UTC)
    I definitely can't argue about anything in Sami language. :-) Different linguistic approaches can use different terminology, that's for sure. I follow the theory that complies with this Wikipedia article, where adding suffixes and preffixes is called derivation. At least some editors use combines (P5238) (e.g. renew (L235)) while some other use derived from (P5191) (e.g. kalendarzowy (L3408)). So let's agree on the fact that unless we all follow one policy in this matter blending can be understood as different method from compounding. And thats the whole point of this proposal, to have a property to distinguish between different methods instead of having them all as one. I think it would be for good to have common understanding of ethymological terms but thats issue for another discussion. --Lexicolover (talk) 21:53, 14 September 2018 (UTC)

@GZWDer, KaMan, Lexicolover, Jura1, Rua: ✓ Done: mode of derivation (P5886). − Pintoch (talk) 21:54, 20 September 2018 (UTC)

Do we want to make this required as a qualifier for derived from (P5191)? —Rua (mew) 22:15, 20 September 2018 (UTC)

@Rua: Not sure about this. I don't know dedicated name for mode of derivation in relations I enter. For example I know noun okazałość (L22779) is derived from adjective okazały (L22777) but I don't know how to best name this way of derivation. I only suppose it would be in Polish "wikt:pl:rzeczownik odprzymiotnikowy" and google translates it as "a rejective adjective" but I'm not specialist to judge. Many dictionaries I know which gives etymology, do not provide description of mode of derivation so adding this as requirement would make it harder to enter informations for average user. KaMan (talk) 07:34, 21 September 2018 (UTC)
@KaMan: I would use combines (P5238) with okazały (L22777) and -ość (L23224). It's not a separate mode of derivation, just suffixation. —Rua (mew) 08:46, 21 September 2018 (UTC)
@Rua: That makes sense, I've converted all such words to use suffixation. KaMan (talk) 10:05, 21 September 2018 (UTC)