[idn] Re: [MultiLingualList] Changes to RACE

["James Seng" <James@Seng.cc>]

Just a quickie note. Any correct implementation of RACE _should_ already have
this feature since if all input streams comes from ROW 0, then there shouldnt
be any RACE transformation.

IMHO, this requirement should be done at the nameprep level, and not the RACE
transformation. Supposing we decided to move away from RACE to something else
(just example sake), we can ensure this behavior is consistent.

-James Seng

----- Original Message -----
From: "Paul Hoffman / IMC" <phoffman@imc.org>
To: <mltestbed@nsiregistry.net>
Sent: Wednesday, October 18, 2000 12:30 AM
Subject: [MultiLingualList] Changes to RACE


> Greetings again. The -02 version of the Internet Draft for RACE was
> posted this morning by the IETF. This draft has an important change
> in section 2.2:
>
> =====
> If a name part consists exclusively of characters that conform to the
> host name requirements in [STD13], the name MUST NOT be converted to
> RACE. That is, a name part that can be represented without RACE MUST NOT
> be encoded using RACE. This absolute requirement prevents there from
> being two different encodings for a single DNS host name.
>
> If any checking for prohibited name parts (such as ones that are
> prohibited characters, case-folding, or canonicalization) is to be done,
> it MUST be done before doing the conversion to an ACE name part.
> =====
>
> Words to this effect had been in earlier versions of the document,
> then taken out because we wanted to make the document more "generic".
> However, some folks have been reading the RACE document in the
> absence of other protocol documents, so it is important to state the
> prohibition here.
>
> --Paul Hoffman, Director
> --Internet Mail Consortium
> ---------
> Participants on the NSI mltestbed list are requested to not cross-post
messages.