Wiktionary talk:Votes/2007-10/Lemma entries/words

Some "fixes":

• The etymology is for word, not for words: is this from an OE plural, or form an OE singular with a modern plural suffix appended, or what?
• "Zero, multiple, or an unknown or undisclosed number of distinct units" is too wordy: how about just "Units"?
• The inflection line (both parts of speech) has the wrong word: word. Instead perhaps have inflection lines (better wikified than I'm writing it, but similar to): words (plural form of (deprecated template usage) word) [[category:English nouns]]; and words (third-person singular simple present form of (deprecated template usage) word) [[category:English verbs]]. (Better yet, use {{en-noun-form|plural|word}}, which will categorize.
• And link to the translation table on the lemma's page, or something, per my comment on Wiktionary:Votes/2007-10/Lemma entries.

—msh210℠ 18:37, 31 October 2007 (UTC)

I don't think form-of entries need an etymology section, unless it's somehow different than that of the root word. Synonyms, antonyms, related terms, translations, see also, et cetera are basically the same. What we want is useable information, not regurgitations of the same stuff that's in the main entry. I'm going to put 2 examples after this message. These entries really don't need that much information, but I think to just list the "form of" and nothing else is heinous :-p — [ ric | opiaterein ] — 19:22, 31 October 2007 (UTC)

English Pronunciation IPA(key): /wɜːdz/ — etc. with whatever other pronunciations Noun words Plural form of word. — I don't think the definitions aren't really necessary: English Wiktionary, for speakers of English? In the case of English entries, I'm not that concerned about it either way. Actions speak louder than words Verb words Third-person singular form of word — which, by the way is inexplicably missing from the actual 'words' article. :-) Anagrams sword Romanian Pronunciation IPA(key): [ku'vin.te.le] Noun cuvintele n (plural definite form of cuvânt) the words

---

Translations

Some seem to like the idea of adding translations to non-lemma entries, but when it comes to verbs in heavily inflected languages, how would we handle this exactly? It's an interesting idea, but is it practical? I'm going to use "said" (past particple and simple past form of say) as an example here.

past particple of say

• Korean: 말한 (malhan), 말하신 (malhasin)
• Latin: not even going to try. :D
• Romanian: zis
• Spanish: dicho

That's easy enough, IF we don't go into numbers and genders, which some say they want to do. But then...

simple past form of say

• Korean: 말했다 (malhaetda), 말했어 (malhaesseo), 말했어요 (malhaesseoyo), 말했습니다 (malhaesseumnida), 말했느냐 (malhaenneunya), 말했어 (malhaesseo), 말했어요 (malhaesseoyo), 말했습니까 (malhaesseumnikka), 말하셨다 (malhasyeotda), 말하셨어 (malhasyeosseo), 말하셨어요 (malhasyeosseoyo), 말하셨습니다 (malhasyeosseumnida), 말하셨느냐 (malhasyeonneunya), 말하셨어 (malhasyeosseo), 말하셨어요 (malhasyeosseoyo), 말하셨습니까 (malhasyeosseumnikka)
• Latin: dīxī, dīxistī, dīxit, dīximus, dīxistis, dīxērunt
• Romanian:
  • Singular: 1st person: zisei, 2nd: ziseşi, 3rd: zise
  • Plural: 1st person: ziserăm, 2nd: ziserăţi, 3rd ziseră
• Spanish:
  • Singular: 1st person: dije, 2nd: dijiste, 3rd: dijo
  • Plural: 1st person: dijimos, 2nd: dijisteis, 3rd: dijeron

At least for Romance languages, the simple past is a personal form, meaning that there is no real simplest way to put it. And for every possible language? Notice that those examples only have 6 forms, but consider Slovenian which has 3 numbers rather than 2, and since technically past tense verbs are conjugated according not only to number, but gender.... that's a lot of forms to be stickin' in this table. Comments, suggestions? We can't just go doing this without some kind of plan of action. — [ ric | opiaterein ] — 20:32, 1 November 2007 (UTC)

Indeed. I added the past forms of just a single Korean translation to the above example. It really should be about four times larger, though, because there are multiple common translations of "said" into Korean. This would be a maintenance nightmare. Rod (A. Smith) 20:52, 1 November 2007 (UTC)

These should have person, number, etc. indicated. An alternative approach follows, but I don't know enough about languages to know whether it's at all feasible; it's not if there a large number of different number-person-etc. combinations that are not shared by many languages.

simple past form of say, third person masculine singular

• Foo: bar
• Baz: blicky

etc.—msh210℠ 21:00, 1 November 2007 (UTC)

I threw in those number and person indicators for Romanian and Spanish. Looks a little better, but still that's a lot of information to be sticking in a translation table. Multiple tables would be easier to organize information, but....

drop downs within the drop down, watch out!

simple past form of say

1st person singular

• Romanian: zisei, spusei
• Spanish: dije

2nd person singular

• Romanian: ziseşi, spuseşi
• Spanish: dijiste

3rd person singular

• Romanian: zise, spuse
• Spanish: dijo

1st person plural

• Romanian: ziserăm, spuserăm
• Spanish: dijimos

2nd person plural

• Romanian: ziserăţi, spuserăţi
• Spanish: dijisteis

3rd person plural

• Romanian: ziseră, spuseră
• Spanish: dijeron

That looks better and is more manageable, I admit, but it's still a lot to manage. I personally don't have a strong objection to it, if this is what we end up wanting to do, that's fine. I just wanted to point this stuff out :) — [ ric | opiaterein ] — 21:13, 1 November 2007 (UTC)

Ohsht. What about synonyms? *Adds to table* — [ ric | opiaterein ] — 21:21, 1 November 2007 (UTC)

I'm amazed that some people think this approach is reasonable. If you're going to split it up for person, number, gender, then why not also by formality, deference, and mood? The reason is that the grammars of various languages differ so drastically that the only reasonable place to describe the grammatical relationships between inflected forms of a word are in the entry for the inflectable lexeme itself, supplemented by appendices for that lexeme's language. If we try to divide foreign term inflections by their approximate mapping to the semantics of non-lemma English entries, we will end up with an inaccurate, unmaintainable mess. Rod (A. Smith) 21:15, 1 November 2007 (UTC)

I didn't say reasonable. lol. Not all languages inflect. For chinese, this would actually be really easy. "wo shuole", "ni shuole", "ta shuole", "women shuole", "nimen shuole", "tamen shuole". It's completely possible to do this, but my qualm with it is that it's so much to do. It won't be inaccurate if people know what they're doing when they're adding, and it won't be a mess if they do it properly and we have a decent format to use before jumping into it. — [ ric | opiaterein ] — 21:21, 1 November 2007 (UTC)