On Fri, Sep 2, 2016 at 5:35 AM, Martin J. Dürst <duerst@it.aoyama.ac.jp>
wrote:
That's where the language code comes into play. Language codes like "ja-JP"
>> uses the national addressing format in Japan: big endian. Language codes
>> like "en" or "ja-Latn" use international addressing format in Japan:
>> little
>> endian.
>>
>
> This makes sense at first sight, but is quite ad-hoc. It's totally unclear
> what "language code" other conventions would use. Also, "ja-Latn" says
> "Japanese language, written with Latin script", but what you really want to
> identify is "Japanese (country!) format, when using Latin script". So you
> would need a country and a script, but not a language. That's not exactly
> what language codes provide.
Exactly so. What is more, no tagging is needed to determine if an address
in Japan is in Latin script or not: that can be determined by inspection of
the content. I'm all for BCP 47 language tags, but they are meant for
specifying languages, not national conventions independent of language.
Similarly, if I had a field in a contract document for "governing law", I
wouldn't use a language tag there, but an ISO 3166 country code, with ISO
3166-2 subtags in the case of federations with multiple legal systems.
--
John Cowan http://vrici.lojban.org/~cowan cowan@ccil.org
Dievas dave dantis; Dievas duos duonos --Lithuanian proverb
Deus dedit dentes; deus dabit panem --Latin version thereof
Deity donated dentition;
deity'll donate doughnuts --English version by Muke
Tever
God gave gums; God'll give granary --Version by Mat McVeagh