> >> >but I don't see why it would fail to interoperate if the comparison >> >against different spellings were done on the server. >> >> Which DNS server ? > >the authoritative servers for the zone. > >caches and resolvers wouldn't be able to return an answer unless >they had an exact match (which might be a CNAME record returned >from an earlier query). without an exact match in the cache, >they'd have to forward or refer the query to an authoritative server. That is just like I defined it in my draft. It will allow the possibility of implementing different matching rules for .fr and .us, if we do not do the matching normalisation in the resolver. Dan
- Re: [idn] UTC feedback Keith Moore
- Re: [idn] UTC feedback Mark Davis
- Re: [idn] UTC feedback Andrea Vine
- Re: [idn] UTC feedback Keith Moore
- Re: [idn] UTC feedback Harald Alvestrand
- Re: [idn] UTC feedback RJ Atkinson
- Re: [idn] UTC feedback Keith Moore
- Re: [idn] UTC feedback Mark Davis
- Re: [idn] UTC feedback RJ Atkinson
- Re: [idn] UTC feedback Keith Moore
- Re: [idn] UTC feedback Dan Oscarsson
- Re: [idn] UTC feedback RJ Atkinson
- Re: [idn] UTC feedback Dan Oscarsson
- Re: [idn] UTC feedback Mark Davis
- Re: [idn] UTC feedback RJ Atkinson
- Re: [idn] UTC feedback Keith Moore
- RE: [idn] UTC feedback Jonathan Rosenne
- Re: [idn] UTC feedback Mark Davis
- Re: [idn] UTC feedback Marc Blanchet
- RE: [idn] UTC feedback Fran�ois Yergeau
- Re: [idn] UTC feedback Keith Moore
