Norwegian languages in openehr terminology

Reference Implementation: Java (archive)
1

Dear all,

in the Java openEHR terminology file ( mini-termserv\src\main\resources\openehr_terminology_en.xml)

there are the following entries for the Norwegian language:

I believe that this needs to be:

rather.
“no” is the Norwegian “macro-language”, which is not really relevant here I believe.

An archetype would either be in “nb” or “nn” language but never in “no”

Currently the Java parser would not parse nb or nn archetypes.

Can someone please confirm this or contradict me?

Regards
Sebastian
2

Hi Sebastian,

I think you are right. The ISO standard does have "nb" and "nn" there.
And "no" means just Norwegian.

http://www.loc.gov/standards/iso639-2/php/code_list.php

Cheers,
Rong

3

Thanks Rong! I have committed the updated terminology file.
Sebastian

(attachments)

oceanlogo.png

4

Thanks, Sebastian!
Just a thought, maybe we could still keep a row for “no”, “Norwegian” in the XML or?
/Rong

(attachments)

oceanlogo.png

5

Hi Rong,

Yes, you could add that one as well, I suppose it doesn’t do any harm.
The only thing is that as far as I understand “Norwegian” is not a real language, but just a “macro-language”.
So you’d never have an archetype ontology that is in Norwegian, but only one in Bokmal and/or Nynorsk.
This is unlike a German (de) archetype ontology and potentially an ontology with the Swiss variant of it. Or “en”, with “en-us”, “en-gb”, “en-au” variants.

Sebastian

(attachments)

oceanlogo.png

6

Hi again,

OK, let’s keep it as is now until some Norwegian-speaking person complains :wink:

/Rong

(attachments)

oceanlogo.png

7

I was vaguely aware of this ‘macro-language’ thing, but now it occurs to me that anyone doing development with archetypes has to know what the formal model is. My current model in the ADL 1.5 compiler is probably wrong because it assumes ‘en’ and then ‘en-GB’, ‘en-US’ etc, and does language matches on the most specific then tries to find a more general one. So what are the correct language variant terms for Norwegian? Is ‘nb’ at the same level as ‘en’? And ‘no-nb’ can’t exist can it?

  • thomas
8

My understanding is that the approach in your compiler is correct because the second part (GB; US; etc) is for a region where a derived and usually only slightly modified version of the language is spoken.
nb and nn are at the same (top-level) level as en - because they are two separate languages and have no real connection to each other other than that they appear to be spoken/written in the same region, i.e. Norway.

What I am unsure about is a general approach for ISO-639 language codes that are classified as Macro-language.
For “no” there seems to not be a language at all…however Farsi (fa) is also classified as macrolanguage - whereas I believe that this would be the usual Farsi language and no more detailed codes are included in ISO-639-1 alone.

There is a bit of explanation here: http://en.wikipedia.org/wiki/Macrolanguage at least partly explains it - in summary based on ISO-639-1 alone you cannot have a general approach as to whether a macro language should be included or not.

Sebastian

(attachments)

oceanlogo.png