W3C home > Mailing lists > Public > ietf-discuss@w3.org > June 2001

Re: Are international characters allowed in email addresses?

From: Eric Brunner-Williams in Portland Maine <brunner@nic-naa.net>
Date: Thu, 28 Jun 2001 12:47:00 -0400
Message-Id: <200106281647.f5SGl0X25428@nic-naa.net>
To: Keith Moore <moore@cs.utk.edu>
cc: Eric Brunner-Williams in Portland Maine <brunner@nic-naa.net>, Eric Brunner-Williams <wampum@maine.rr.com>, discuss@apps.ietf.org, brunner@nic-naa.net
> and you've been in the IDN discussion so you should also be familiar 
> with those issues.

The IDN discussion has produced a requirements memo that states, inter
alia, that

	The intended scope of this document is to explore requirements
	for the internationalization of domain names ...

This isn't header processing, and it isn't 8-bit clean processing, and
the use of "internationalization" here differs fundamentally from industry
and academic i18n usages.

In the same statement of scope, the requirements memo even constains
a recommendation, viz,

	It is recommended that solutions not necessarily be within the
	DNS itself, but could be a layer interjected between the
	application and the DNS.

This is somewhat unique to the IDN WG's Requirements Draft editors, a
recommendation in lieu of a, and before, a requirement.

It is within reason that Apps area ADs, former and current, could want to
solve a problem which could be stated as infrastructural, or as application
specific, in the latter form.

It is within reason that the application-specific problem statement is not
the unique means to identify the mechanism or set of mechanisms which form
a solvable problem.

The majority of the IDN poll participants have expressed some form of
interest in an application solution for extending the set of 63 (only
37 unique) values available for composing dns labels. These all involve
encoding a larger set of values on the unchanged (though tagged) set of
63 (or 37) values, and some heuristics about the larger set of values.
There are the predicable distractions when someone mentions "language"
or "script" or "search" or "friendly" or ... but these are distractions,
the only dns issue is the label space, 4-bits-plus-5, or greater.

That doesn't make the proposal correct or best or anything but one with
some interest (mine included, but not in an exhaustive "ACE ONLY" context).
It does suggest that the discourse on the issue could be improved by moving
the WG into the Apps Area, since its co-chairs, requirements editors, and
more popular proposals favor application-specific mechanisms.

In a nutshell, the discussion of 8-bit transparency, in SMTP, or in any
other part of the problem space, in the IDN list, has been a little less
than professional.

This isn't about Eric's time or Keith's, availability or valuation.

Received on Thursday, 28 June 2001 12:48:33 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:08:12 UTC