Re: [idn] Re: idn-uri document

[Martin Duerst <duerst@w3.org>]

Hello Steve,
(B
(BIn an URI, *.example.org (e.g. http://*.example.org) is very
(Bclearly illegal.
(B
(BAlso, I don't think it's a bad idea that a group that decides on the
(Bsyntax of a new protocol element give some thoughts about where the
(Brules should be checked, and where not, in order to assure consistent
(Bdeployment. The restrictions on domain names given by IDNA/nameprep/
(Bstringprep are not in any sense policy issues, they are very clearly
(Btechnical issues.
(B
(BRegards,    Martin.
(B
(BAt 09:40 02/11/01 +0000, Steve Dyer wrote:
(B>At 23:40 31/10/2002 +0100, Erik Nordmark wrote:
(B>> > >The defined syntax rules for declare certain ASCII domain names illegal
(B>> > >(such as *.example.org). Where is the check for illedgal names assumed to
(B>> > >be performed? For IDNA it probably makes sense to only apply this types
(B>> > >of checks (setting the UseSTD3ASCIIRules flag) when verifying domain name
(B>> > >registrations and not do such checks in the clients.
(B>> >
(B>> > This is an IDNA question, not a idn-uri question. As far as I remember,
(B>> > the idea was to have the checks done on the clients, too (with some
(B>> > leeway for unassigned characters to stay forward-compatible with
(B>> > new character assignements). The reason for this was to create
(B>> > pressure on registries to follow the rules.
(B>
(B>Hi,
(B>I am surprised that you indicate that certain ASCII domains will be 
(B>illegal. I can see why unix people may be unhappy about "*.example.org".
(B>I trust that most high ASCII will be allowed - "$Bq(B (e-acute) for example 
(B>since software such as BIND9 handles them.
(B>
(B>I am also concerned that a group discussing a protocol is suggesting that 
(B>the protocol is designed to "create pressure on registries". A protocol is 
(B>a stand-alone technical facility, not an instrument of policy.
(B>
(B>Regards
(B>
(B>Steve Dyer
(B>CentralNic Ltd
(B>
(B>