Re: [idn] Re: REORDERING: stability issues and UTC solutions

[Patrik Fältström <paf@cisco.com>]

--On 2001-10-22 11.45 +0900 Martin Duerst <duerst@w3.org> wrote:

>> > My  question was that:
>> >   1) newly-approved TAGALOG characters X,Y  have   NFC X -> Y,
>> >   2) old Nameprep/ACE encodes X.com and Y.com as two distinct ACE
>> >   labels,
>> 
>> unassingned code points are allowed in query from old version nameprep.
>> X.com Y.com are passed through into DNS servers which have no stored
>> tagalog labels before TAGALOG is added, but begin to accept it after
>> it's approved  & added.
> 
> That makes a lot of sense. Or do you want to say it doesn't.

I would appreciate if discussions on versioning uses the terminology etc
which is already established in earlier versions of the nameprep document,
and now moved to section 6 of draft-hoffman-stringprep-00.txt.

I especially see an embryo of reinvention of an already in this wg
established way of handling unassigned codepoints.

That this is handled the correct way is especially important as that
section of the stringprep/nameprep algorithm is in accordance with
agreements between IETF and UTC on how additional codepoints are assigned.

  paf