Re: 8 bit characters in DNS names (and URNs?)

Masataka Ohta (mohta@necom830.hpcl.titech.ac.jp)
Mon, 11 Mar 96 11:39:39 JST


From: Masataka Ohta <mohta@necom830.hpcl.titech.ac.jp>
Message-Id: <199603110239.LAA05204@necom830.hpcl.titech.ac.jp>
Subject: Re: 8 bit characters in DNS names (and URNs?)
To: sint@oeaw.ac.at (Peter Paul Sint)
Date: Mon, 11 Mar 96 11:39:39 JST
Cc: masinter@parc.xerox.com, keld@dkuug.dk, martin@terena.nl,
In-Reply-To: <v02130505ad668ac95a8d@[193.170.88.66]>; from "Peter Paul Sint" at Mar 9, 96 2:34 am

> >> While uppercase mapping is culturally sensitive, can we not make a
> >> culturally independent 'character matching' algorithm that is good
> >> enough for directory services.

> >Theoretically, it is a union of all the matching rules of all
> >the culture. But, in practice, it is hard especially because
> >the expected degree of matching differs service by service.

It's a union. OK?

> German has a lower case letter
> (looks like a beta

Of course.

> You would never write umlaut A as an A. (only aliens do so - and software).

Of course.

> The back transformation is not unique!

Of course.

But, it is not a problem if, with some internationalized non-strict
directory service, a pattern of umlaut 'A' matches both 'a', 'A',
umlaut 'a', umlaut 'A', 'ae' and 'AE'.

							Masataka Ohta