Re: My prod at IDN requirements

["Martin J. Duerst" <duerst@w3.org>]

At 08:30 00/01/04 +0100, Harald Tveit Alvestrand wrote:
> At 12:05 04.01.00 +0900, Martin J. Duerst wrote:
> 
> > >    i18c Cyrillic A must compare not equal to Latin A
> >
> >Follows from consistent server behaviour and the fact that
> >we don't want to require it to compare equal.
> >May look like a serious practical problem, but won't, because
> >DNS label components are words, not letters.
> 
> Is there then a requirement that all characters in a DNS label must relate 
> to each other in some fashion?

The 'relate' would basically mean that it is from the same script.
I'm not sure this can always be inforced; for Japanese, you
definitely want to be able to mix hiragana and katakana and kanji.

> How could we express this requirement?
> Yes, I'm worried that someone will register <ASCII C> <Greek Omicron> 
> <ASCII M>.

Who would want to register this? If put anywhere in print,
what would be the chance that somebody types it in correctly?
Would it be possible to do such a registration, then wait for
somebody else to develop COM (all ASCII) to a lot of value,
and then claim that it's the same, and <ASCII C> <Greek Omicron> 
<ASCII M> was earlier, and therefore COM belongs to the registrant
of the first one? Of course, the person registering COM would
never have checked for the Greek Omicron case.

Do we have to nail down all the rules for what can be a name
and what not explicitly, or can we deal with some problems
in a more roundabout fashion if we can convince ourselves
(and document) that there should be no actual use cases?

> > >    i18c A with Ring Above must compare equal to a with ring above
> > >    i18c A with Ring Above must compare not equal to a with ring above
> >
> >'must compare' is not clear enough. Is this for the user, or for
> >a DNS server? I would say that for the user, it indeed must, but
> >that that should be worded carefully so that it doesn't imply
> >that it does on the server. Also, mention that case behaviour
> >should be user-dependent (Turkish dotless i).
> 
> I was thinking of the server's deciding whether 2 names match or not in 
> this case. And that (by the requirement for consistency) should not be 
> locale dependent.

Which then means that if we want case folding for the user, case
folding has to be done on the client side. But for the moment, it
mainly means that for each such requirement, we have to make clear
where it applies (user or server). And probably, most such requirements
should at the moment be seen from the user side.



Regards,   Martin.


#-#-#  Martin J. Du"rst, World Wide Web Consortium
#-#-#  mailto:duerst@w3.org   http://www.w3.org