RTL labels and numbers?

[Andrew Sullivan]

On Tue, Oct 13, 2009 at 09:05:54AM -0400, Vint Cerf wrote:
> for the most part, the IDNA2008 specification does place a great deal of 
> responsibility on
> the registry but for some cases, it was considered important to bar  
> particularly confusing
> situations at the protocol level. This was debated substantially during 
> the course of the
> development of IDNA2008.
>

In addition, the argument in this case was that the numbers had a
tendency to jump across the label separator, so that _no_ registry
could actually exert proper control over how this would work (the
parent could redelegate via DNAME without telling the child, and the
child could do something surprising that would affect display from an
otherwise-properly delegated name from the parent).  Because of this
natural tendency to involve other labels, and becauset he only
practical tests are intra-label and not inter-label, at least some of
us argued that the best guide was prudence.  Therefore, the stricter
rule was put in place.  (Otherwise, I also would have argued "registry
policy".)

A

-- 
Andrew Sullivan
ajs at shinkuro.com
Shinkuro, Inc.