----- Original Message ----- From: "Edmon" <[EMAIL PROTECTED]> To: "Harald Alvestrand" <[EMAIL PROTECTED]> Sent: Wednesday, August 16, 2000 10:26 AM Subject: Re: [idn] Adding "optional" characters in draft-ietf-idn-nameprep > > ----- Original Message ----- > From: "Harald Alvestrand" <[EMAIL PROTECTED]> > To: "Edmon" <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]> > Sent: Tuesday, August 15, 2000 6:38 PM > Subject: Re: [idn] Adding "optional" characters in draft-ietf-idn-nameprep > > > > At 16:03 15/08/2000 -0400, Edmon wrote: > > > > machines need to know how to translate IDNs from whatever > > > > encoding they use into unicode/10646 (for queries) and back > > > > (for address lookups). > > > > > > > > > >That is very reasonable... why then is it not a good idea to tag the > > >encoding as we have suggested in a standard and easily recognizable way? > > > > it is not a good idea to move the nonstandard encodings over the Internet. > > As I told you before: if you use an encoding with someone, you have to > have > > an agreement to use this encoding. > > If you use an encoding without private agreement on the Internet, you > > require the whole Internet to understand this encoding. > > > > Don't send multiple encodings across the Internet. > > But in theory, UTF-8, is in itself a multiple encoding scheme including the > ASCII set... > Your suggestion mean that we should prob go straight to UCS-4 and have > everything uniform byte length without any "multiple" encoding issue. > That is in fact not a bad idea... > > Edmon > > > > > Harald > >
