Jim,
Sorry I didn't answer your message before.
What I've just included in the specification is that dc:language in TAPIR metadata responses must follow RFC 4646 (http://www.rfc-editor.org/rfc/rfc4646.txt) and therefore use language codes specified by the IANA Language Subtag Registry (http://www.iana.org/assignments/language-subtag-registry). So yes, it's the same IANA that you were thinking.
I hope these are the correct and most up-to-date references for what we need.
Best Regards, -- Renato
Everything Renato is saying sounds right on and I think the invasive community will be able to agree on a language standard as long as it covers enough languages.
- The ISO 639-2 (3 letter codes) should work but the 2-letter codes miss
some language distinctions.
- The only IANA codes I am aware of are the "Internet Assigned Numbers
Authority" and these are country codes rather than language codes - did I miss something here?
- I'm not as familiar with "Ethnologue" and since it is not ISO, it may be
harder to sell
My only other comment is to restate that scientific names should be treated as language-independent. I see them as codes for taxons (they used to be Latin but that has changed over the years - see the Chinese dinosaur "Tsintaosaurus").
Thanks, Jim