Re: My prod at IDN requirements

["Martin J. Duerst" <duerst@xxxxxx>]

At 09:34 00/01/04 +0100, Patrik F$BgM(Jtstr$B‹N(J wrote:
(B> --On 2000-01-04 09.23 +0100, Patrik F$BgM(Jtstr$B‹N(J  wrote:
(B> 
(B> > Regarding what Martin wrote, the question whether the server or the user
(B> > should define equality, I must say that it must be the same.
(B> 
(B> Let me immediately correct myself here.
(B> 
(B> What I was trying to say is that we have to think about what the server 
(B> treats as equality. As the example Harald brought up:
(B> 
(B> <ASCII C> <039F;GREEK CAPITAL LETTER OMICRON> <ASCII M>
(B> <ASCII C> <ASCII O> <ASCII M>.
(B> 
(B> Are those domainnames equal?
(B> 
(B> The server _must_ descide, and depending on this descision, the user might 
(B> either in his user interface see a difference or not, all depending on the 
(B> presentation format of the domainname.
(B> 
(B> For example, on many terminals and in many fonts, 0 (digit zero) and O 
(B> (capital o) looks equal, and the same with 1 (digit one) and l (lower case 
(B> L). Is that a problem in the DNS that a user might see these as equal? No.
(B
(BWhy not? Couldn't somebody create something like WWW.1BM.c0m, and try
(Bto confuse people?
(B
(B
(B> It is up to the presentation system to see that the user sees a difference 
(B> based on the rules in the DNS server.
(B
(BThere is no guarantee that at all places where
(BDNS names are displayed, a font with the relevant distictions is
(Bused. Or is there?
(B
(B
(BRegards,   Martin.
(B
(B
(B#-#-#  Martin J. Du"rst, World Wide Web Consortium
(B#-#-#  mailto:duerst@xxxxxx   http://www.w3.org