RE: [idn] host name vs. domain name

["McDonald, Ira" <imcdonald@sharplabs.com>]

Hi Paul,

Right - summaries are dangerous.  I omitted to say
that the RFC 2277 requirement for UTF-8 support is
for 'text' and NOT for 'protocol elements' (such
as 'STOR' in FTP or 'Content-Encoding' in HTTP).

Whether domain names are 'text' or 'protocol elements'
(by the definitions in RFC 2277) is an interesting
question.  Generally 'protocol elements' have a
fixed, finite set of values.  That's certainly not
true of domain names.  Of course, RFC 2277 also has
some interesting fuzziness about 'names' (without 
fully clarifying what it means by 'names').

Nonetheless, RFC 2277 is a BCP policy and not merely
a goal.  And it has already been construed to apply to
text in SNMP MIB objects by the IESG.  This precedent
is relevant in my opinion.

Certainly, the excellent work by Larry Masinter and
Martin Duerst on Internationalized URI is specifically
addressing 'names' in the relevant sense (see
<draft-masinter-url-i18n-05.txt>, March 2000).

Cheers,
- Ira McDonald (consulting architect at Sharp Labs America)
  High North Inc

-----Original Message-----
From: Paul Hoffman / IMC [mailto:phoffman@imc.org]
Sent: Thursday, March 16, 2000 4:22 PM
To: idn@ops.ietf.org
Subject: RE: [idn] host name vs. domain name


At 03:56 PM 3/16/00 -0800, McDonald, Ira wrote:
>It is rather more than a 'goal' of the IETF to move away
>from ASCII.  Per RFC 2277 'IETF Policy on Character Sets
>and Languages' (BCP 18, January 1998), ALL Internet protocols
>MUST support the use of UTF-8 in order to 'enter or remain
>on the Internet standards track'.  Newly developed Internet
>protocols MUST default to UTF-8 for text strings.  Existing
>Internet protocols MUST be updated to support UTF-8.

RFC 2277 is a nice, short document. It would be great if people read it. 
When they do, they'll see that much of what appears in the previous 
paragraph is not at all accurate.

--Paul Hoffman, Director
--Internet Mail Consortium