From: Kai Henningsen (kaih@khms.westfalen.de)
Date: Sun Nov 02 1997 - 07:40:00 CST
mduerst@ifi.unizh.ch (Martin J. Dürst) wrote on 31.10.97 in <Pine.SUN.3.96.971031135515.245v-100000@enoshima.ifi.unizh.ch>:
> We can either say that we adopt the rules for equality given in the
> Unicode standard (which means that every program working with newsgroups
> and wanting to compare two names has to implement them), or we can say
> that:
>
> - Comparison of newsgroups is octet-per-octet.
>
> - For those cases where Unicode defines equivalences, only the "more
> popular" version should be used.
>
> - This group feels that it would be good to know in detail what "more
> popular" in the above sentence means, but thinks that it is
> beyond their scope and expertise to define such a thing, that
> other IETF standards may face similar problems, and that this
> should be dealt with separately.
I'd add a recommendation for reading agents to use the full comparision
rules for newsgroup names entered by the user, and converting any
representation to that in the group list, if it's equivalent according to
these rules, except that this might give problems when someone manages to
actually create two "equivalet, but spelled differently" newsgroup names -
reading agents might select the "wrong" one. Hmm ...
MfG Kai