ISO language codes versus country codes

[Caoimhin O Donnaile]

Cord said:

> Language and country are two different, although often related, concepts.
> They require different codin systems.
>
> The best thing to do is to keep these two data elements apart. I personally
> feel that having identical codes for the language (e.g Japanese) and the
> country name (e.g. Japan) is not necessary and may even be a disadvantage
> sometimes.
>
> Keep your data elements apart and you are fine.

I would agree and go further.  It seems that the confusion which
Sean Burke has found between country codes and language codes is
an argument for using 3-letter (ISO 639-2, Ethnologue) language
codes, or 4-letter language codes?, and abandoning the 2-letter
ISO 639-1 codes.  Except that a lot of the 2-letter codes are so
well established that they would be a very long time in going.

Caoimhín