Some CSL Lithuanian locale translation problems

maras

Here is new Lithuanian locale for CSL: https://gist.github.com/3318667

And some problems.
1. Is it better translate Latin terms (as circa, et. al., ibid.) to Lithuanian or not? There is no tradition Lithuania, so I don't know. Some Lithuanian styles uses Latin terms ("op. cit." is widely used), but not all. What is most widespread practice in translation? I personally like Latin terms, so... Even "no date" is presented as "s[ine] a[nno]" in some cases.

2. Language issues.
a. Gender of ordinals and long role forms. Endings are: mask. "-as" and fem. "-a" in all of them. How to overcome this? There is similar issue with French: http://forums.zotero.org/discussion/15085/1/french-localization-csl-10/
b. Four ordinals are equal. Maybe we don't need them at all? :)
c. There is confusion with months. Names of them must be in genitive in textual form of date in Lithuanian. I think that is suitable for dates, so I wrote them in genitive. Is it OK?

3. Terrible mess with editors in Lithuanian. "Redaktorius" means technical work with text mainly, more as stylist in Lithuanian. Unless it is editorial director of newspaper. I suggest "sudarytojas" as more convenient. BUT "vyriausias redaktorius" is rather similar to "editor" or "editor in chief". BUT verb "redagavo" as "edited" means mainly technical work in lithuanian.
What does term "editorial-director" mean exactly in Zotero context? I can suggest both editor and "editorial-director" translate as "sudarytojas" or as "sudarytojas" and " vyriausiasis redaktorius" accordingly. But I suggest one "sudarė" for verb role forms of both.

4. Miscellanea.
What does mean term "recipient"? Interviewee?
Figure is something like illustration?
Political sciences is broader than politology, I think?
"Literature" is one of humanities, not belletristic?
"Retrieved" is something similar to field "accessed" in Zotero?
What does mean term "reference"? Is it reference to other work in footnote?
Paper "presented at" conference?
"Cited" is for "cited after/from"?
What does mean "container-author" and "forthcoming"?

Thank you for patience to read all that and answers.

Ir gerai būtų, kad kas lietuviškai mokantis permestų lokalę, ar nesąmonių neprirašiau bei kokių pastabų bei pasiūlymų pateiktų. [That is invocatory in Lithuanian. :)]

Rintze

1) Keeping the Latin terms is fine.

2a) and 2b) CSL 1.0.1 will have better support for ordinal gender. We do need to know the gender of "edition" and "volume", and of the months.
2c) Don't know.

3) "editor" is more technical than "editorial-director". See e.g. http://forums.zotero.org/discussion/8565/ for a discussion about the differences.

4) "recipient" --> e.g. the recipient of a letter
"figure" --> I'm not sure if this is ever used, but my guess is that this is e.g. a figure or chart in a scientific paper.
"political_science"/"literature" --> don't worry about these (they don't have to be translated)
"retrieved" --> I think this can be either spatial or temporal (where or when was the item metadata retrieved)
"reference" --> I think this refers to a particular bibliographic entry
"presented at" --> yes, what you wrote
"cited" --> I'm not sure if this is ever used, and I'm not sure why you'd use it
"container-author" --> e.g. the author of the larger work (if you're citing a chapter, this would be the author of the book)
"forthcoming" --> e.g. the status of a paper that has been accepted but yet has to appear in print or online

aurimas

This looks great. I didn't see any translation errors and I completely feel your pain when trying to translate technical jargon into Lithuanian. I tried tackling some of the terms that you didn't translate, but I didn't have anything to add that I would feel very comfortable with.

I made some very slight modifications (closing single quotation mark and removed leading zero from text date). I also reformatted the text date definition slightly, which (on a more technical note) I think is more resilient way of doing it. Also fixed a couple open tags. It now validates against the CSL validator.

https://gist.github.com/3319907

Rintze

I think the "text" date format should be:

<date form="text" delimiter=" "> <!-- "2011 m. lapkričio 1 d." -->
<date-part name="year"/>
<date-part name="month" prefix="m. "/>
<date-part name="day" form="numeric" suffix=" d."/>
</date>

The "m." only gets printed when there is a month in the date, right? With my version, you should get "2011" (year), "2011 m. lapkričio" (month-year), and "2011 m. lapkričio 1 d." (day-month-year).

aurimas

The "m." actually refers to the year ("metai")

Edit: Also, your example is great for pointing out the difference in grammatical cases, which I don't believe are supported. In the case of year-month-day, the case for the month should be genitive (which is correct). More literal translation of "2011 m. lapkričio 1 d." would be "1st day of November of year 2011". In the case of year-month, the translation should be "2011 m. lapkritis" -> "November of year 2011" (nominative case).

maras

Thank you both. Your comments were very useful.

And I adopted Rintze's version of "text" date format as it seems more logic to me.

And yes, there will be problems with year-month form of text date. Russian style author used nominative, not genitive form of months (this language have the same "problem"). I'm confused. Well, manual edition of reference is inevitable in both cases, so I chose genitive.
Russian is widespread language, so maybe support for nominative-genitive will be introduced someday. :)

My concept of "editor" and "editorial-director" is different. Editor is more alike to English conception, not to French (so it resembles english "editorial-director"). "Editorial-director" is for collective editing (senior of editorial board) or for person, who was in charge of preparing text to publishing (manuscript etc.). This difference exists in Lithuanian.

"Cited" is for reference to citation, that is cited not from original, but from retell, I think.

I am historian, so some nuances are familiar to me.

What is difference between "online" and "internet"? First is publication type, second points to the source from which retrieved? That is the only question remaining.

maras

No, one more question. What is context of names "from" and "at"?

maras

And about language issues. The gender of "edition" and "volume", and of the months is the same in Lithuanian all the time. Changes only gender of ordinals and long role forms

Rintze

1)

And I adopted Rintze's version of "text" date format as it seems more logic to me.

You probably shouldn't, if, as aurimas indicated, the "m." is included when you only print a year.

2) if the nominative/genitive case is handled consistently among the languages that make this distinction (Russian, Lithuanian, Polish?), we could add CSL support for it. (it might be enough to define nominative and genitive variants of the month terms, and automatically use nominative for year-month, and genitive for year-month-day)

3) "online", "internet", "from" and "at" are just English words. Except for "online" they're probably barely used in CSL styles, though.

4)

The gender of "edition" and "volume", and of the months is the same in Lithuanian all the time.

Yes, but I need to know what that gender is. In CSL 1.0.1 it will be possible to define masculine (e.g. "pirmas") and feminine gender-variants (e.g. "pirma") of the ordinal terms. When rendering something like "1st edition" in Lithuanian, the gender of "edition" then determines which gender-variant of the ordinal term is used. So I need to know for "edition", "volume", and the months, which are feminine, and which are masculine.

maras

1. I'll consult with stylist of Lithuanian. But I don't remember inclusion of "m." into the references with only a year number.

2. And also Latvian too. Yes, I think that you described precisely.

3. I saw some of them in Polish and Russian locales which I've read as examples. But Lithuanian specific couldn't handle them.

4. Oh, now I understand, sorry :). Edition and volume - masculine. May and July - feminine, other months - masculine. But long ordinals can be both gender.

maras

There is almost final draft of Lithuanian locale translation: https://gist.github.com/3342275 I changed some terms to the more appropriate and translated some new.

I found resource of quotation marks thanks to link in the text of Polish locale: http://en.wikipedia.org/wiki/Non-English_usage_of_quotation_marks That solved one problem for me.

Where put "m." in text form of the date? I checked several versions of Lithuanian bibliographic records. There were no one occurrence of year number with "m.". So, I'll stay with Rintze's version.

No "date must" be in brackets in some styles. I think, that is quite reasonable and I did that - "[s.a.]".

"Forthcoming" and "in press" relations. As for me, forthcoming is for presented, but not yet accepted paper, and "in press" - accepted but yet has to appear in journal. Such is interpretations of Russian and Polish locales but that contradicts Rintze's interpretation.
This is the only remaining question for me.

So I am waiting for a critique :).

Rintze

No "date must" be in brackets in some styles. I think, that is quite reasonable and I did that - "[s.a.]".

Don't. When a style needs the brackets, it can add them. Terms shouldn't be translated with surplus punctuation.

forthcoming is for presented, but not yet accepted paper, and "in press" - accepted but yet has to appear in journal.

That's probably more accurate than what I wrote above.

aurimas

Where put "m." in text form of the date? I checked several versions of Lithuanian bibliographic records. There were no one occurrence of year number with "m.". So, I'll stay with Rintze's version.

I might be completely wrong on this, as I do not have any source to point to, but...

I think it's not surprising that you did not find any occurrences of year followed by "m.". This would require a style that uses the text-date format with no available day and month for that date, which is probably quite rare. Or perhaps it's actually not used that way. But consider the following example of where I think a text date might be used.

"Published January 1, 2001"
"Išleista 2001 m. sausio 1"

Without the available month and day, this would then become

"Published 2001"
"Išleista 2001" or "Išleista 2001 m."

In my opinion the second option is more correct Lithuanian.

Again, this is just something to consider and it may be completely wrong. I'll try to look for some sources.

aurimas

Edit: Also, your example is great for pointing out the difference in grammatical cases, which I don't believe are supported. In the case of year-month-day, the case for the month should be genitive (which is correct). More literal translation of "2011 m. lapkričio 1 d." would be "1st day of November of year 2011". In the case of year-month, the translation should be "2011 m. lapkritis" -> "November of year 2011" (nominative case).

The example in my last post made me reconsider the post quoted above.

"2011 m. lapkritis" (missing day) is correct, however it would only make sense when used alone. If placed in a sentence like "Published ...", "Išleista 2011 m. lapkritis" would make no sense. This would require the accusative case "Išleista 2011 m. laprkitį". Luckily, we can hide the accusative case in the abbreviated word for "month" and get by with using the genitive case for the actual name of the month. It would look like "Išleista 2011 lapkričio mėn.". Note the "mėn.", which, in this case, is short for "mėnesį" (accusative case of "month"), but could be short for any other case.

Even without the "Published" part, if the date does appear on it's own and is without a day, I would say that "2011 lapkričio mėn." would look more natural, because the intent is to say that something was published, accessed. etc. on that date.

Technically the full date could be written as "2011 m. lapkričio mėn. 1 d.", which would make it very easy to account for missing days, but I don't think it's a very common case.

So in conclusion, at least in the example given above, the genitive form would suffice, but when used without a day, we would have to include the suffix "mėn."

I guess this is more to give Rintze some more ideas about how to improve the CSL processor. I'm sure my post is a bit confusing, perhaps maras can give a better explanation.

maras

@ Rintze,
Brackets removed.

@aurimas,
you may be right. Most styles are based on Lithuanian ISO standarts with variations, often quite considerable. They are described here: http://www.lsd.lt/standards/catalog.php?ics=0&pid=638332 but I don't have this book. Particularly these instructions are written after ISO standarts: http://www.mb.vu.lt/ELBI/janonis.pdf I didn't found any form about you've wrote ("Išleista 2011 m. laprkitį"). There is only "žiūrėta", but not "išleista". Bibliographic entry must be understandable for foreigners, so dates should speak without words :). But I might be wrong too.
Situation that you described can exist, but in very rare occasions, I think.
There is nominative of months for magazines and genitive - for newspapers in text form of date (both without "m." for "years"!) and full text date for "accessed" with genitive form of month and with "m." in book of Janonis. I don't like such a system - too complicated and not elegant, even chaotic. And it is in use quite rarely, as I know.

(Edit. But if style of date is "2005, lapkričio 5" (without "d."!, that is from book of Janonis), then "m." as prefix of month isn't relevant. And what about "m." as suffix of year in such a case? Year and day must be from numeric, but month - from text form of date, as I can understand. Is it possible? If so, than version of aurimas is better.)

maras

After some considerations I adopted aurimas version of text dates.
There is last version of locale: https://gist.github.com/3351765 I removed all my comments. I think, this is final variant, unless there will be some remarks.

Rintze

I added the Lithuanian locale to the locales repository. Thanks!

https://github.com/citation-style-language/locales/commit/cd5b0d102466a52184d42dccdbf455fd0676948b

maras

It is good to hear that. Thank you!

maras

After some testing of Lithuanian locale I found several defects of it: there weren't spaces after year and month (I included incomplete changes made by aurimas) and a dash before short ordinals that is required.
There is corrected version: https://gist.github.com/3465838
Sorry for troubles.

Rintze

Updated, thanks!

https://github.com/citation-style-language/locales/commit/ae202fce7c69ccffeee0ed3a15315b072ab357b6

Gracile

@Rintze

"cited" --> I'm not sure if this is ever used, and I'm not sure why you'd use it

Some (french) styles uses it for "op. cit.", is it correct?

Rintze

Reading up at http://en.wikipedia.org/wiki/Op._cit. , I think so :). My background is in the life sciences, and I've never seen it in print.

Gracile

I wasn't clear: do they have the same function? I am using "op. cit." but with <text value="op. cit"/> ; I wonder whether it should be added as the correct translation of "cited".

Rintze

"cited" was added before my time, and I'm not sure about its intended use. (see http://xbiblio-devel.2463403.n2.nabble.com/schema-terms-and-labels-td5927364.html )

Seeing that "op. cit." is Latin, it might make sense to have a dedicated term for it, as we have for "ibid.".

adamsmith

I think the reason that we don't have an op cit term is that Bruce hates it. A lot. :-). But now that we support op cit type citations, yes, we should have a term.
"Cited" is used as a label for accessed dates in a large number of styles (edit: grep gives me a list of about 100 styles using term="cited" right now) - it doesn't make a lot of sense to me, but that's what the style guides require, so no, we definitely shouldn't change that.

Rintze

But now that we support op cit type citations

How exactly? Using the "position" conditional?

Gracile

ATM "cited" is used in 105 styles.
In most cases, it seems to be an alternative to "accessed". But in a few cases (10 French, 1 German and 1 Italian styles), it is redefined as "op. cit.":

annales.csl
geneses.csl
histoire-at-politique.csl
histoire-et-mesure.csl
invisu.csl
le-mouvement-social.csl
presses-universitaires-de-rennes.csl
revue-dhistoire-moderne-et-contemporaine.csl
universite-de-liege-histoire.csl
vingtieme-siecle.csl

universitat-heidelberg-historisches-seminar.csl

universita-di-bologna-lettere.csl

[Note that the 11 first styles have been created by the same author]

In addition, 4 others styles use <text value="op. cit"/> or <text value="op.&#160;cit." :

clio-medica.csl
journal-of-applied-philosophy.csl
lettres-et-sciences-humaines-fr.csl
medical-history.csl

Gracile

Cross-posting obviously.

I think the reason that we don't have an op cit term is that Bruce hates it. A lot. :-).

Sure... I remember that it was one of the first post I read on this forum!

How exactly? Using the "position" conditional?

I've always used "op. cit." (with a short title to avoid confusions) with position="subsequent".

adamsmith

But now that we support op cit type citations

How exactly? Using the "position" conditional?

I was thinking that op cit. often gets used with
first-reference-note-number - but that may have been an illusion.
Otherwise yes, with position="subsequent" - but we've always supported that.

Gracile

Using "first-reference-note-number" is better in some cases (it addresses the main criticism against "op. cit"). But, for the styles which are in the repo, op. cit. is used with position="subsequent".

fbennett

Where op. cit is used, the potential for reader frustration could be reduced by triggering it on position="near-note", to assure that the full citation is somewhere within reach.