Re: [idn] IDNA: is the specification proper, adequate, and complete? (was: Re: I-D ACTION:draft-ietf-idn-idna-08.txt)

["Soobok Lee" <lsb@postel.co.kr>]

----- Original Message ----- 
From: "Dave Crocker" <dhc@dcrocker.net>
 > The working group needs to focus more narrowly:  we have exactly ONE DNS 
> PROBLEM that must have immediate solution:
> 
>          Permitting an internationalized form of domain name for business
>          cards -- that is, for URLs and email.  These are usually called
>          hostnames.
> 
>          All other IDN requirements can wait.
> 
> So far, the IDN working group has produced nothing that is used.  If the 
> working group issues a first, narrow specification, we can make immediate 
> progress and offer the community something it desperately needs.  We can 
> also learn more about how this all works and then apply it to the next, 
> ummmmmmm,,, domain.

Do you suggest that the new narrower specification should be about

  1)  transitional directory solution that may become the precursor to the real IDN with "NEW CLASS" ?  or
  2)  the real IDN limited to hostnames  and its supporting RRs (NS,A,MX,CNAME) ?  
 
If future "new class" should have type A and type NS records like "IN" class,  approach 1) is more 
appropriate in this stage. some commercial IDN TLD registries already implemented this transitional
directory by wildcard label (*) resolutions for native 8bit A-type queries. They redirect such incoming
native 8bit A-type queries into a forwarding webserver which maintains registrants-supplied
ASCII URL address with no ACE hostname. Moreover, the forwarding webserver can extract some default
locale information from the http header information from the brower client and make most correct
guess about what the local charset was and transcode the 8bit hostname into UCS one which is then 
compared against UCS-to-URL databases . I think this directory feature is enough for most desperate
consumers for a while  and IMO, forever.

Soobok Lee