Re: [idn] NSI Multilingual Testbed Information (fwd)

[James Seng <James@Seng.cc>]

Header too long?

I suppose this is the one which causes the problem...

CC: Bill Manning <bmanning@ISI.EDU>, "Patrik
=?UTF-8?B?RsODxpLDhuKAmcOD4oCgw6LigqzihKLDg8aSw6LigqzFocOD4oCaw4LCpGx0c3Ryw4PGksOG4oCZw4PigKDDouKCrOKEosODxpLDouKCrMWhw4PigJrDgsK2bQ==?="
         <paf@cisco.com>,Hollenbeck Scott <shollenb@netsol.com>, 
        idn@ops.ietf.org

-James Seng

-------- Original Message --------
Date: Sat, 26 Aug 2000 14:55:56 -0700
From: Mark Davis <markdavis@ispchannel.com>
Reply-To: Mark Davis <mark@macchiato.com>
X-Mailer: Mozilla 4.7 [en] (WinNT; U)
X-Accept-Language: en,de-CH,fr-CH,it
MIME-Version: 1.0
To: Mark Davis <mark@macchiato.com>
Subject: Re: [idn] NSI Multilingual Testbed Information (fwd)
References: <200008261931.MAA08170@zed.isi.edu>
<39A836A5.42A49CC@ispchannel.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit

[Whoops. Hit the send key by mistake.]

Yes. I don't know how much you are familiar with Unicode, so I'll say a
few words.*

Unicode was designed from the start to be a universal character
encoding, not limited
to a particular set of countries or languages. To make sure of this, we
have worked
closely with a wide variety of people and organizations: members of the
consortium
itself*, the WG2 committee, the IRG (for Chinese, Japanese, and Korean
characters;
with membership from a number of Asian countries), and outside experts.

The recommendations that the consortium makes, the data that it collects
and presents,
and the specifications it approves, are designed to allow
implementations all over the
world to interwork effectively.


[NAMEPREP] in its final version has strong implications for registrars.
They should
not register any names that [NAMEPREP] would exclude, or register two
names that
[NAMEPREP] would identify. If they do something in advance that differs
from the final
specification, they will have some very disappointed customers.

Since the precise set of characters allowed in [NAMEPREP] has not yet
been decided, I
called the LC-NFKC-Identifiers file to people's attention, since it is
likely to be a
subset (or close to a subset) of whatever [NAMEPREP] ends up being, yet
includes the
required letters for the different languages. Since NSI has taken the
step of allowing
registration in advance of a final spec for IDN, my strong
recommendation is to
restrict registered characters to that set for safety's sake. If a
particular
registrar wants to restrict the set further, that is its own business.

Mark

* for more information, see http://www.unicode.org
* members include
http://www.unicode.org/unicode/consortium/memblogo.html

Mark Davis wrote:

> Yes.
>
> Bill Manning wrote:
>
> >  Would you hold the NU, CN, JP, KR registrars to the same criteria?
> >
> > %
> > % I would definitely agree with Patrik. At an absolute minimum, the provisional
> > % registrations should restrict themselves to characters that are distinguished
> > % under caseless NFKC identifier rules. While that may not be what the IDN group
> > % finally resolves, it has a far better chance of not causing collision problems
> > % than if arbitrary character choice is allowed.
> > %
> >
> > --
> > --bill