Difference between revisions of "Talk:Vocabulary"

From Learn Na'vi Wiki
Jump to navigation Jump to search
Line 113: Line 113:
:* English number words: "thirty-seven"
:* English number words: "thirty-seven"
:That would avoid confusion over whether 10 is ten or eight or vol or vomun or what.  --[[User:Erimeyz|Erimeyz]] 20:40, 21 February 2010 (UTC)
:That would avoid confusion over whether 10 is ten or eight or vol or vomun or what.  --[[User:Erimeyz|Erimeyz]] 20:40, 21 February 2010 (UTC)
:Even if you don't adopt that suggestion, I do agree with William about "octal ten".  Also "octal eleven", "octal twelve", "decimal eight", "decimal nine", and "decimal ten".  The words "octal" or "decimal" should always be followed by digits, never by number words.  Thus: "octal 10" "octal 11" "octal 12" "decimal 8" "decimal 9" "decimal 10". --[[User:Erimeyz|Erimeyz]] 20:46, 21 February 2010 (UTC)

Revision as of 20:46, 21 February 2010


This is excellent! But before you get too much further, I'm a little worried about using tables for everything. They frankly a pain in the butt to work with, and it's easy for even experienced Wiki editors to have trouble with them. Templates can take arguments... perhaps we could agree on some layout here, and then use that in the main dictionary (unless of course everyone else adores tables)? Wm.annis 20:50, 22 January 2010 (UTC)

I'm glad, that you like it. About the tables: I'm not a very experienced Wiki-editor, so I choosed tables, because I thought they would do it. If there are other ways to produce alike results, I'm keen on learning about them. So, if you have got some advice or suggestions about better ways for the layout, Id like to hear them. Na'rìghawnu 08:28, 23 January 2010 (UTC) 22:38, 22 January 2010 (UTC)

Well, I suggest removing the tables and doing the vocabulary as plain text, like the following example. Any comments? Na'rìghawnu 08:28, 23 January 2010 (UTC)

'ekong [Ɂɛkoŋ] (N). beat (rhythmic), e. g. heart~. ► ekong telanä - the beat of the hearts (ASG, Weaving Song).

That looks awesome. Wm.annis 15:07, 23 January 2010 (UTC)
It might be good to put the translations in quotes, so we can plain, unquoted text to make quick comments about the grammar of a word, etc. Also, do we really want to enter plurals as separate entries? Wm.annis 15:12, 23 January 2010 (UTC)
I agree - it looks awesome. I also agree that the translations of the sourced examples should be in quotation marks; I don't think the English definitions of the entries themselves should be, though (I'm not sure if that's what you meant or not, Wm.) -- Erimeyz 16:06, 23 January 2010 (UTC)
Right, I mean in the examples, not the definition. Wm.annis 16:15, 23 January 2010 (UTC)
Ok. I will do this. Na'rìghawnu 16:30, 23 January 2010 (UTC)
My next question — why use red to indicate the accent, when Frommer has already given us a way to do that, which is just as fussy in wiki markup as making the font red?
Because underling often causes problems, e. g. to separate "y" and "v" (as it did in frommers message text too). I see the red color as a much better way not to interfere with the letters. Na'rìghawnu 16:29, 23 January 2010 (UTC)
Ah! That makes sense. Can we choose a less alarming color, though? ;) Actually, we could write a very simple template to wrap a stressed syllable in, so if we don't like the look, we can easily change all of them by updating the template. Wm.annis 16:35, 23 January 2010 (UTC)
Very well! Since I'm not very expercienced in writing templates, I would appreciate any help. (If you could write such a template, I'd surely use it. Na'rìghawnu 16:44, 23 January 2010 (UTC)
Done. Take a look at the T section for examples. Wm.annis 16:53, 23 January 2010 (UTC)
Fine! Thanks. But how about a bit darker green? (Maybe it's just my monitor, but letters seem to be a bit too light.)
How about purple? Wm.annis 17:01, 23 January 2010 (UTC)
 :-) Perfect! Na'rìghawnu 08:31, 24 January 2010 (UTC)

Na'rìghawnu 16:58, 23 January 2010 (UTC) And about the separate plurals ... well, I thought it would be helpfull (exspecially for beginners), because it's not so easy to guess, that the word is "koren", when in the text, where you found the word, it is "horen" for instance. I allways liked e. g. my Greek dictionary, which gave me a hint, when I tried to find irregular verbs or so. Of course, the plurals are not absolutely necessary, but do they cause problems? Na'rìghawnu 16:44, 23 January 2010 (UTC)

That makes sense, too, I guess. I'd make them simpler entries, I guess, just give the word they're a plural of and maybe a brief definition, but no citation data or examples. Does that make sense? Wm.annis 16:53, 23 January 2010 (UTC)
Sounds OK to me. Will change it after dinner. ;-) Na'rìghawnu 16:58, 23 January 2010 (UTC)

Outstanding work y'all are doing here, guys. It really looks great, and contains exactly the information we'd want in a lexicon (and no more!). I love the format. Keep up the good work! -- Erimeyz 03:24, 24 January 2010 (UTC)

You are very welcome, Erimeyz. Let's make this a useful tool for the Na'vi community. If there are further suggestions or comments don't hesitate to speak about it. Na'rìghawnu 07:22, 24 January 2010 (UTC)

Division of Labor

Ma Na'rìghawnu, if you want to point out sections in need of work, please feel free to make a list. I'm happy to fill in parts of the dictionary, but I don't want to accidentally start an edit in the same sections you might be working on. Wm.annis 20:25, 24 January 2010 (UTC)

Ma William, thanks for the support. Well, it's a bit difficult for me to point out sections, because I don't do the list alphabetically or so. The way I do it is, that (1.) I have done a complete analysis of all the corpus-texts we have (without Jake's script for now), (2.) I go through the analized corpus texts/sentences step by step and add new vocabulary into the wiki as they appear in the texts/sentences (intentionally omitting words, where I'm still not sure how to handle them). Using this method I until now added the vocabs of all the interviews and the complete Weaving Song. Now I'm working on the Hunt Song. Next will be A Message From Paul. I hope, that I can complete this until today's evening (well, German evening). When this is done, there is surely still left work and things to discuss:
1. Will we also add the vocabs from the ASG-dictionary (even without having any sentences, where these words are used)? Na'rìghawnu 07:44, 25 January 2010 (UTC)
2. How to discuss vocabs, which seem problematic (at least to me)? Do we do this here or do we start a thread in the forum? Na'rìghawnu 07:44, 25 January 2010 (UTC)
3. I didn't deal with the stressed syllables very much until now. (This is a thing you may add at any time, since I'm surely not working on it in the next days.) Was the underlining in AMFP done by Frommer himself or did other people do it according what they heard in the spoken message? Na'rìghawnu 07:44, 25 January 2010 (UTC)
I suggest using the forums for discussion. The wiki talk pages are okay for short exchanges about wiki pages themselves (like this!), but language analysis will often need more extensive conversations, and the forum has better mechanisms to support that. Also, by talking on the forum, more people will be likely to see it, which leads to a better discussion (and greater awareness of the wiki's existence, but that's a separate matter).
The stress marking in AMFP was provided by Frommer, according to Prrton. -- Erimeyz 14:26, 25 January 2010 (UTC)

ASG Words

How should we indicate the source of the words from ASG, given that they don't have examples? -- Erimeyz 19:46, 29 January 2010 (UTC)

The lack of examples seems like an awfully stark signal that it's from that source alone. Wm.annis 19:58, 29 January 2010 (UTC)
Yes, I agree. As long as there is no example to add out of the corpus, the word is knowable as a word from the ASG list. Maybe we should include an explanation about this in the preface? Na'rìghawnu 19:16, 30 January 2010 (UTC)

Where to put derived lemmata

Gosh. Writing a dictionary for Na'vi should not cause my brain so many questions. I had put uvan si as a sub-entry of uvan. Na'rìghawnu, you moved it out to a separate entry, which is probably the correct way to go. But then, under Na'vi, the derived adjective and adverb are put there. On the one hand, it sort of makes sense to put derived forms next to each other, to get a feel for the semantic range. On the other hand, that'll probably be confusing to beginners. What's the best way to deal with this? Go double, put the derived forms and their examples all together, but have an entry out for the derived form which points to the main article, like we do for plurals? — Wm.annis 15:06, 2 February 2010 (UTC)

Yes, indeed, ma William, this thing isn't an easy task. I made the "uvan si" a separate entry, because of two points: (1.) "uvan si" is - of course - derived from "uvan", but it's a verb on it's own, consisting of two words, but nevertheless ONE single verb (with it's own semantics, pragmatics, syntax ... pretty the same like other compound-verbs, e. g. "yomtìng" (which I'd not like to see put just under "yom"); (2.) i've handled other verbs of this type the same way before (e. g. irayo si, teya si etc.) and wanted to maintain consistency.
About "Na'vi" ... I actually don't see "leNa'vi" and "nìNa'vi" as single words on their own rights. They are - at least to me - much less independent words than the "X si"-verbs, so that I would even hesitate to put all "nìX" words into the dictionary at all, "nìÌnglìsì" for instance. I added them in the case of "Na'vi" because I wanted to make clear in this case, that in Na'vi "Na'vi" isn't the language, but just the People, and since I suspect, that users could search after the language, I wanted to give them a hint, how these things are done in Na'vi. It's a related thing about "leNa'vi".
As I said: About "X si"-verbs, I would really suggest to keep them as separate entries, because they aren't just "derived", but independent verbs with their own rules (e. g. "eltu si" is another excellent example, it has an entirely new meaning and can't be seen just as a derivation of eltu (or si) the way one can derive "leNa'vi" from "Na'vi" ... it's a completely different quality of "derivation"; same about other compound-verbs, like "yomtìng" "tìng nari" and alike. If we'd delete the separate entry about "uvan si" and put it under "uvan", than we also had to put "eltu si" under "eltu" without a separate entry for this verb.) About "leNa'vi" and "nìNa'vi" I think, that these two are merely derivations and are in the dictionary just to illustrate some things (stricly spoken, they could even be deleted).
"What's the best way to deal with this?" ... I don't know. I would say, we should not hesitate to make single entries for all relevant words (and point to related words if necessary) [so I vote for "X si" as single entries here too, the same about "lefpom", which I wouldn't like to see just put under "fpom" ("happy" isn't just a dervation of "well-being" in my eyes)], but we should also not split it up too much (e. g. making "ayngaru" a single entry would be too much, I feel). But it's not a simple question. Should we make some rules for how to deal with alike cases? Would be fine to me. But how to let other users know about such rules? Na'rìghawnu 18:08, 3 February 2010 (UTC)
This isn't my area of expertise, but here are my thoughts:
  • Attested root words get their own entries.
  • Certain attested derived forms share entries with root words, but get their own gloss and cited examples.
    • part-of-speech: le- ne- tì- sä- -tu -yu
    • number: me- pxe- ay-
    • gender: -an -e
    • adpositional: mì, ne, hu, etc. (both prepositional and enclitic)
  • Attested forms using case suffixes (ergative, genitive, etc) or verb infixes (tense, aspect, affect, reflexive, etc) are included with the root word entry, but are only used in the cited examples. They don't get their own instance or gloss.
  • Attested compound forms using si or tìng get their own entries.
  • Unattested root words that are implied by attested derived forms get their own entries, but are marked as such; they serve as the anchor entry for the attested derived form(s).
  • Unattested derived forms are not included.
I think organizing things like this will help people develop their vocabulary and their grasp of grammar by seeing what the key conceptual vocabulary words are plus seeing how their forms and meanings are altered using the standard mechanisms. I'd even go so far as to say we don't need pointer entries for the derived forms, even the plurals. Someone trying to look up "lefpom" (for example) won't find it under L, but if they do a find in their browser they'll find it under "fpom", along with all the other (attested) words it's related to. -- Erimeyz 19:02, 3 February 2010 (UTC)
Well ... I don't have much time at the moment, so I just want to throw in one single thought about it: Of course, this is an online-dictionary and so the "search"-function of the browser is a thing one could use here, but I'd like to have the dictionary more like a analog one. My reason: In this special dictionary quotes of sentences are given to every single word. So e. g. "lefpom" doesn't only show up in it's own entry, but also (as part of a quote) in quite a lot of other entries too. If you use the browser-search you most likely have to click trough many entries, until you find the entry dedicated to "lefpom" itself. So I think, it would be much easier to find it under "L". Besides that: In any other serious dictionary you also will find "happy" as an entry of it's one and not just under "happiness". What I mean is, that words, which are really independent words on their own, should get their own entry, even if it is clear, that they are derivations of other words. Of course, you may see the method of derivation better, if there are derived words put together in one entry, but I see this as a "normal" dictionary, not an ethymological one. Just my two cents. Na'rìghawnu 08:48, 4 February 2010 (UTC)
Those are very good points. As I said, I'm not the expert here, so please take my suggestions as merely suggestions. The one solid piece of advice I have is to come up with a scheme that you and William both are happy with, then document it here on the talk page, then stick to it rigorously. - Erimeyz 13:36, 4 February 2010 (UTC)
I have had to use dictionaries for natural languages that are based around root, rather than derived words (Arabic, Sanskrit). I find them utterly maddening, and I have experience with these languages. I think with a liberal use of "see also" sections, we can both cover etymology and also make life a bit easier for students.
Putting the V si forms right after the V entry makes sense. They're right next to each other anyway. But for things like nìNa'vi it might be best to have a separate lemma for it, but just have a "see also" to Na'vi. —Wm.annis 17:33, 4 February 2010 (UTC)
Ok. Sounds fine to me too. Let's do it this way. Na'rìghawnu 17:36, 4 February 2010 (UTC)

How to mark compound verbs

Should we come up with some particular way to mark lemmata for compound verbs? This matters a lot in getting the infixes into the right place. Perhaps a dash in the citation, yom-tìng or sìl-pey? —Wm.annis 13:55, 5 February 2010 (UTC)

Ok. You are right, that this would be helpful. We could use e. g.
Most dictionaries I know, use the | to divide morphemes of lemma-words. But we can of course choose any other method too. Na'rìghawnu 17:50, 5 February 2010 (UTC)
I like the middot (of either size) or the dash. In the normal Mediawiki font size, the bar is easy to overlook or misinterpret. — Wm.annis 23:01, 5 February 2010 (UTC)
Very well ... then I'd like to vote for the smaller middot: yom∙tìng. About the dash, I think, that it looks maybe too "normal", so that users may misinterpret it as a formatting error or maybe a regular part of the word's spelling (like the "le-Na'vi" in the "Hunt Song", although this certainly is a phenomenon of the early stage of writing Na'vi). On the other hand, of course, to produce the middot is a bit more work than just the dash, so I'd like to hear, whether you were comfortable with it. Na'rìghawnu 07:21, 6 February 2010 (UTC)
Good point about the dash. Middot it is! — Wm.annis 15:56, 6 February 2010 (UTC)


Hi, I have a question concerning IPA with the symbol for ejectives. It must be the ʼ [U+02BC], right? Then the German Wikipedia article on IPA fails, because there they pretending ’ [U+2019] to be the right symbol, and there need some fixes here as well. (e.g. ’awkx [Ɂawk’] → ’awkx [Ɂawkʼ]) - Ochristi 07:22, 12 February 2010 (UTC)

According to the Unicode doc itself, it's U+02BC. U+2019 is a quote mark. — Wm.annis 16:33, 12 February 2010 (UTC)
Alright, thanks for the nice references. Then I will change 17 occurrences I found. Besides I know the U+2019 to be the symbol we use for the glottal stop and which should be preferred to U+0027 the normal apostrophe. - Ochristi 18:29, 12 February 2010 (UTC)

Good job, Ochristi! Thanks for participating. Na'rìghawnu 07:45, 13 February 2010 (UTC)

Octal Numbers

I'm not sure it's a good idea to notate octal 10 as the word ten. It's eight. If we're going to include anything ten-ish, it should be the written digits, 10, maybe with some notation about that not being base ten, such as 108. There's also the computer science way of doing this, o10 or 0o10. —Wm.annis 18:42, 21 February 2010 (UTC)

I would avoid using octal numerals. I would standardize on using the following conventions exclusively:
  • Na'vi number words: tsìvomrr
  • English transliterations: four-eights-five
  • Decimal numerals: 37
  • English number words: "thirty-seven"
That would avoid confusion over whether 10 is ten or eight or vol or vomun or what. --Erimeyz 20:40, 21 February 2010 (UTC)
Even if you don't adopt that suggestion, I do agree with William about "octal ten". Also "octal eleven", "octal twelve", "decimal eight", "decimal nine", and "decimal ten". The words "octal" or "decimal" should always be followed by digits, never by number words. Thus: "octal 10" "octal 11" "octal 12" "decimal 8" "decimal 9" "decimal 10". --Erimeyz 20:46, 21 February 2010 (UTC)