Re: [idn] case preservation

["Soobok Lee" <lsb@postel.co.kr>]

----- Original Message ----- 
From: "Patrik Fältström" <paf@cisco.com>
To: <idn@ops.ietf.org>
Sent: Tuesday, October 09, 2001 2:54 PM
Subject: Re: [idn] case preservation


> --On 01-10-09 04.37 +0000 "D. J. Bernstein" <djb@cr.yp.to> wrote:
> 
> >> prohibiting greek 'o' does not make sense
> > 
> > If someone is allowed to register aol.com with a Greek lowercase o, then
> > nameprep is a bad joke.
> 
> Already today, people can register domainnames with characters which look
> "similar" like 'o' and 0, or '1' and 'l'. I.e. this is not a new problem.
> It is well-known for the people which work with domain name dispute
> resolution.
> 
But, the real problem lies in the increased number of such ambiguities.
For LDH domains, we need to know about only 'O','o','1','l' .
But for IDN, almost _ALL_ of 'a' ~ 'z' have their look-alike characters in
greek,cyrillic and cherokee script. Seeing them printed  on papers,
most people can't know whether they are LDH domains or IDN ones.

ICANN/DNSO/ccSO should make some recommendations/enforcement 
on IDN registration and dispute resolution policies 
before IDN's being deployed and owned.

Soobok Lee


> Further, this wg has chosen to use the Unicode Character Set and
> Normalization according to UTC TR#15, normalization form KC. If you are
> unhappy with how individual characters are normalized, please to go the
> Unicode Consortium after reading the criteria for the normalization process
> chosen.
> 
> If you are unhappy with the choice of refering to the work in the Unicode
> Consortium, that is a valid concern, but should have been brought up some
> 1.5-2 years ago.
> 
> That horse is already dead.
> 
>    paf
> 
>