CLDR data (Re: Comments on the IDNA2008 document)

John C Klensin klensin at jck.com
Mon Jan 12 20:54:03 CET 2009



--On Monday, January 12, 2009 20:39 +0100 Harald Alvestrand
<harald at alvestrand.no> wrote:

> I'd strongly object to placing a dependency on CLDR's
> character lists in  the standard itself.
> 
> Two reasons:
> 
> 1) Enforcing this requirement would require (not just
> recommend) that  the intended locale for each and every domain
> name be known at  registration-checking time. Otherwise,
> there's no way to know what rules  to enforce.

And, of course, that was more or less tried.  The experience was
that one could not get people to do it and could not enforce it
in any way.  Making that sort of information a requirement of
the protocol would require that the local be somehow captured
and stored as part of the protocol (e.g., by requiring that
every IDN label be accompanied by locale information, stored in
some specific non-address R, at the same note).  Horribly
complex and still hard to enforce in any meaningful way.  

For those registries at the top level for which global rules are
possible, one might be able to figure out a way to do this in
the registrar-registry protocol and associated databases, but
they are a tiny fraction of the registries (zone
administrations) on the Internet and, even there, it appears
that very strong recommendations to keep language or locale
information for each label has not been successful.

>...
> The question of making registrations match with locales has
> been pushed  off to the registries, and I think it should stay
> pushed.

Concur.

     john





More information about the Idna-update mailing list