Stupid U-label question [correction]

John C Klensin klensin at jck.com
Wed Aug 20 15:47:14 CEST 2008



--On Wednesday, 20 August, 2008 13:46 +0200 Frank Ellermann
<hmdmhdfmhdjmzdtjmzdtzktdkztdjz at gmail.com> wrote:

> Tina Dam wrote:
> 
>> A domain name cannot begin with "xn--" unless it is an
>> A-label. Your "u2" below is not an A-label as far as I can
>> tell.
> 
> Yes, based on an IDNA2003 rule and the "--" rule in John's
> draft.
> 
> If he'd remove the "--" rule instead of replacing it by
> something better xn--4caä could be an U-label, after all it's
> LDH plus one perfectly PVALID umlauted a.

But I didn't remove the simple "--" without replacing it with
something else.   So I don't understand what you think is the
issue here, other than than we always need to be careful about
out definitions.

> And if xn--4caä *would* be an U-label, then it *would* follow
> that xn--xn--4ca-cxa is an A-label (and vice versa).  Of
> course nobody here wants this.  An obscure corner case, maybe
> useful as example.

If pigs could fly, they might be both halal and kosher, at least
if they had enough feathers on their wings to make them birds
and not swine.

    john





More information about the Idna-update mailing list