Re: Language Identifiers in IDNA tables and 3066bis

At 06:37 27/12/2004, Tex Texin wrote:
>JFC,
>hi. I can certainly agree that language tags are used in multiple ways, causing
>confusion.
>
>However, restricing their use to language tags identifying IDNA tables is going
>too far in the other direction.
>Language tags are mentioned in several RFCs for other purposes, such as
>content-language in headers,
>the RFCs for HTML, etc.

We are not discussing language tags in here, but language tags related to 
country codes. These are the local languages as accepted by the Local 
Internet Community. This is a sovereign decision by the language legal 
authority as the ccTLD are more and more and consistently accepted as 
public services under the Gov authority.

>So it seems fair to say that RFCs should serve the purposes of IETF network
>interoperability, but that should include more than IDNA tables, unless I
>misunderstand the meaning of "IDNA tables".

IDNA tables define the character set accepted for a given language by a 
ccTLD. First reactions to the publications of Arabic, Hebrew, Russian 
tables by Poland, shown how much this issue is important to language authorities.

This does not prevent any other tag - but to related to a country code. 
This is why I disfavor the 2 letters codes which confuse and overlap ccTLD 
and I prefer a more precise grid.

To smooth the things, I have asked comments on the ccTLD mailing list. If 
they respond you will get elements. If they don't we will be able to tell 
we asked.

All the best.

jfc

Received on Monday, 27 December 2004 06:18:02 UTC