[80953] in North American Network Operators' Group
Re: Underscores in host names
daemon@ATHENA.MIT.EDU (Eric Brunner-Williams in Portland )
Thu May 19 12:48:45 2005
To: Edward Lewis <Ed.Lewis@neustar.biz>
Cc: Roger Marquis <marquis@roble.com>, nanog@trapdoor.merit.edu,
brunner@nic-naa.net
In-Reply-To: Your message of "Thu, 19 May 2005 12:09:20 EDT."
<a06200705beb267f93732@[192.168.1.101]>
Date: Thu, 19 May 2005 12:13:27 -0400
From: Eric Brunner-Williams in Portland Maine <brunner@nic-naa.net>
Errors-To: owner-nanog@merit.edu
> Supporting "IDN" is a necessary job. That's been made clear to the
> Internet community. If it "complicates" things, well, then that's
> what has to be done. If the Internet is to be global, it can't
> restrict the world to just a few convenient languages.
Not to quibble unnecessarily, but the folks I came to the dance with at
IETF-50, eventually went home fairly disapointed after -51, and -52,with
none of their proposed mechanisms drafts having obtained even working group
draft status.
You know what the constraints are -- no zone local semantics (e.g., case
folding rules, courtesy H.A.) for a glyph repetoire that in some ranges
is also a character set, no intermediate tables, no flag day(s) for apps,
and so on.
To describe that as "IDN", rather than "a way to represent, poorly for
some, not so poorly for others, character sets other than ASCII in apps",
leaves the later reader ignorant of the baroque design choices available
and discarded on the road to RACE II.
In Abenaki, "w", "ou" and "8" all collate to the same code point, and the
representation of the code point is application specific (modern, early,
and 17thrCa styles).
Eric
P.S. 17th century French lacked a "w" character, "8" is a "u" atop an "o".
