RE: [idn] stringprep comment 1

[Yves Arrouye <yves@realnames.com>]

> >But they *are* allowed because the Server S uses Nameprep-08!
> 
> I'm misunderstanding your scenario then. You said in nameprep-08,
> character U+XXXX has a "deleting mapping". I understood that as
> "character U+XXXX is now assigned but is prohibited". If that's not
> what you meant, please help me.

I meant that Nameprep-08 maps the codepoint out (to reuse the language in
Nameprep). Let's say because this codepoint is a new variant selector,
another zero-width thingamagic, whatever. 

> >No this one is a specific critic of IDN breaking the existing DNS
> "matching
> >must be case insensitive rule." If it is not must (MUST) then maybe it's
> not
> >an issue.
> 
> There is no such rule: there is a strong attempt. The Turkish dotless
> I is an example of where the attempt will inherently fail in one case
> or the other.

OK. Forget the case argument then. (Man, I am glad I won't be at the end of
customer support for IDN. And that's not a reflection on the IDN work, just
the complexity of the issue and the user expectations.)

YA