Question on ISO-639:1988

Peter Constable petercon at microsoft.com
Wed May 26 01:00:45 CEST 2004


>> If ISO639 isn't a good way to tag content language, then why is it there at 
>> all?  What else does it tag?

> ISO 639 is good at identifying languages, but there are many cases in which it is not 
> sufficient enough to identify content narrowly.

At issue here are two different senses of the term “language”: ISO 639 uses language in a narrow sense, meaning a linguistic variety. Possibly spoken, possibly written, but all that is distinguished is the linguistic variety. 

RFC 3066 uses “language” in a derivative sense that encompasses both dialectal variations and also representations of linguistic information, in particular, written forms. Thus, for purposes of RFC 3066, Chinese in simplified characters and Chinese in traditional characters are two different things; and English with US spelling and English with UK spelling are two different things. For XML content, such distinctions are important. But ISO 639 does not make such distinctions.

Is ISO 639 of any use? Absolutely. It is a building block on which RFC 3066 is built, and there are other usage contexts in which differences related to dialect or written form are not relevant.


Peter
 
Peter Constable
Globalization Infrastructure and Font Technologies
Microsoft Windows Division




More information about the Ietf-languages mailing list