Wiktionary talk:Votes/2007-10/Translation into lemma only

Definition from Wiktionary, the free dictionary
Jump to: navigation, search


"If the target lemma is specific to one gender and the English noun is used...for a gender different from that of the target lemma, provide a translation for both genders. "

So you'd translate girlfriend as novio? DAVilla 00:29, 16 October 2007 (UTC)

The intent is to translate good as just bueno m; friend as amigo m, amiga f; boy as chico m; and girl as chica f. That wasn't clear from the original wording, so I reworded it. Better? Rod (A. Smith) 07:14, 16 October 2007 (UTC)
Ruakh improved the text, and as it stands covers the first two numbered situations below. Reviewing WT:BP#Translation into lemma only, I notice we're missing the request made by Lmaltier: If two different nouns exist in the target language depending on the gender (for the same English word), provide both. To cover that request, I think the wording needs expanded:
  1. For foreign nouns without grammatical gender, those whose grammatical gender is unrelated to the semantic gender, those whose form is gender invariant, and those whose lemma gender matches the only English semantic gender, only give the lemma translation:
  2. For foreign nouns with grammatical gender that varies with semantic gender and whose lemma form is different from the only semantic gender of the English noun, list the translation that varies from the lemma only by gender but link to the lemma entry:
  3. For foreign nouns whose grammatical gender varies with semantic gender and whose lemma form matches one of the two semantic genders of the English noun, only the lemma translation and the equivalent for the other semantic gender(s) should be given:
Is there a clear but brief way to specify the above? Should I delay the vote to allow wording refinement? Rod (A. Smith) 01:25, 17 October 2007 (UTC)
I don't much like the combination of numbers 2 and 3; either of them by itself makes sense, but taken together they seem somewhat contradictory in spirit. Also, by only talking about gender, the section as a whole makes number 1 seem pointless: obviously, if the foreign language has only one gender for a given noun, then that's what we translate to. I think I'd prefer something like this:
  • In translating nouns:
    • In many languages, nouns have different forms depending on number, case, definiteness, and/or other factors. Only one form, called the "lemma form", appears as headwords in dictionaries, and only this form should appear in translation tables in English noun entries. So, for example, the English translations at computer include Spanish computadora ("computer"), but not Spanish computadoras ("computers").
    • Where English has a single gender-neutral noun, many languages will have two nouns (or one noun with two forms, depending how you look at it), one for men/males and one for women/females. For example, English cousin corresponds to both Spanish primo ("[male] cousin") and Spanish prima ("[female] cousin"). In such cases, both the male/masculine lemma form and the female/feminine lemma form should be listed in the translation table.
    • Where English has a gender-specific noun, like boy or girl, the translations should correspond to the gender. In the event that the foreign-language entry is a soft redirect to an entry with a different gender (say, with a definition along the lines of "Feminine counterpart to noun."), then … ??
  • In translating adjectives:
    • In many languages, adjectives have different forms depending on number, gender, case, definiteness, and/or other factors. Only one form, called the "lemma form", appears as headwords in dictionaries, and only this form should appear in translation tables in English noun entries. So, for example, the English translations at good include Spanish bueno (the masculine singular form), but not Spanish buena (the feminine singular).
I don't know what to do in the last noun case. My preference would be for something like "chica f, see chico", but I'm not sure how other editors would feel about that.
RuakhTALK 03:48, 17 October 2007 (UTC)
My biggest motivation in this is to stop editors from filling adjective translations with the non-lemma genders, so I won't put up a strong fight against such variations as above. My opinion, though, regarding non-lemma entries and "see" is that hyperlinks are a terse way to say "see the entry linked here for details". Assuming that Spanish chica is to be considered a non-lemma form of Spanish chico in the same way as the plural noun form words (ug, I just noticed that "words" has translations), all of the details to be found at "chica#Spanish" and "es:chica#Español" should also be at "chico#Spanish" and "es:chico#Español". It seems, then, that "chica f" is the terse and reader-friendly way to say "chica f, see chico". I guess this makes most sense in the context of the other ELE lemma discussion, i.e. acknowledging in WT:ELE that some words are to be given full lemma entries but others should be soft redirects. Rod (A. Smith) 17:59, 17 October 2007 (UTC)
Personally, I don't really agree that chica is just a non-lemma form of chico, but I'm O.K. with treating it as one for Wiktionary purposes. The problem is, your proposal doesn't do that; #2 in your proposal treats chica as a non-lemma form of chico, while #3 treats chico and chica as separate nouns. —RuakhTALK 18:40, 17 October 2007 (UTC)
I agree that the proposal as it now stands is unfinished. It's an inadequate attempt to incorporate Lmatier's request to list m/f noun forms. I have delayed the start and end by a week so we can work out the kinks.
As for the lemma/non-lemma status of "chica", consider that a single member of "chicos" is either "chico" or "chica". Thus, all four forms ("-o", "-a", "os", "-as") are inflections of a single word. (Note that English plurals denote some semantics of quantity not denoted by the corresponding singular, but the semantics of quantity do not grant lemma status to plurals. Similarly, "chica" and similar Spanish feminine noun forms denote human gender without granting them lemma status.) Rod (A. Smith) 19:33, 17 October 2007 (UTC)
Chicos can mean "children" (generally) or "boys" (specifically); "chicos y chicas" gets 1.4 million Google hits, which wouldn't make sense if chicas were strictly a subset of chicos. However, like I said, I'm O.K. with taking chica as a form of chico for Wiktionary purposes, as I do see the argument for that approach; my concern is with the inconsistency, firstly because in general I dislike inconsistency, and secondly because I can only imagine that new editors find inconsistencies harder to learn. —RuakhTALK 21:46, 17 October 2007 (UTC)
(Strangely, the query you linked only gives me 289,000 hits. Did Google reindex after your search? Yes, "chicos y chicas" is common, but "chicos" by itself does not specify a gender, except to exclude a group of specifically all females.) I also prefer consistency, so I'll think about this, see whether more feedback comes, and attempt another revision. Rod (A. Smith) 22:24, 17 October 2007 (UTC)
Re: "Strangely, the query you linked only gives me 289,000 hits.": Even more strangely, I now clicked again, and it says there are 1.59 million hits! Dunno what's up with that. Regardless, it gets tons of hits, enough to make my point. :-P   —RuakhTALK 23:04, 17 October 2007 (UTC)

Some slight changes in wording I recommend:

  1. In the ===Lemma entries section===, say "kinds of words", not types of words. There is an important and often overlooked difference between these words. A type is a theoretical model, not a specific instance (kind). I strongly prefer the word "kind".
  2. "...and words of other inflected languages". I think it would be better to say "and in some cases, words in other inflected languages are handled differently." with the phrase "other inflected languages" linked to Wiktionary:Language considerations.
  3. What does "optinally" mean? I assume this is a typo.

The only additional thing I don't see here is a clear explanation ot what "lemma" means. I think rather than simply linking to the WT entry for that term, we need an explanation specific to ELE. --EncycloPetey 23:36, 25 October 2007 (UTC)

The above suggestions refer to Wiktionary:Votes/2007-10/Lemma entries. I have changed that text as recommended, linking to Wiktionary:Glossary#L for clarification of the lemma.[1] I did slightly alter the suggested change. Let me know if the reword conflicts with your intent. Rod (A. Smith) 16:27, 26 October 2007 (UTC)

Animate noun gender inflection[edit]

This section moved to Wiktionary talk:Translations#Animate noun gender inflection. Rod (A. Smith) 18:11, 19 October 2007 (UTC)

What's the point?[edit]

Sorry, but it is entirely unclear what this addresses. The wording in it, seems wrong, in that it prohibits glosses for non-lemma foreign language entries. But that aside, it is giving no new clarification that I can see, only adding verbosity to a policy that is already far too long. Am I missing something here? --Connel MacKenzie 17:07, 26 October 2007 (UTC)

Is the wrong vote linked from WT:VOTE? --Connel MacKenzie 17:08, 26 October 2007 (UTC)

I think the point is to have

*Romanian: produs
*Spanish: hecho

instead of

*Romanian: produs m, produsă f, produşi m pl, produse f pl
*Spanish: hecho m, hecha f, hechos m pl, hechas f pl

which seems to happen relatively frequently in translation tables. That help? — [ ric | opiaterein ] — 17:40, 26 October 2007 (UTC)

2010 update[edit]

this vote was withdrawn before it started. I'm not sure in such circumstances how honest it is to reopen it. It probably is. How much can we change the wording? Mglovesfun (talk) 11:49, 21 June 2010 (UTC)