Some confusion about policy application and its effects

[JFC Morfin]

At 00:15 05/08/2008, John C Klensin wrote:
>If there is a layering violation, it is there in both IDNA2003 and 
>IDNA2008 although it is certainly more likely with the latter.

We know from the very begining that IDNA is a measured risk 
architectural protocol violation.  It is not end to end. Its is not 
an RFC 1958 technology wide solution. You perfectly described the 
fuzzyness of IDNA in some cases. So, it is restricted to areas where 
it is not fuzzy : this was made fully clear by the answers James Seng 
gave to my questions about IDNA ambitions, that Vint approved. The 
purpose of this WG is to publish a fully consistent IDN200X along its Charter.

This WG and AD decided that it is up to others organization to 
discuss and document interoperable or non-operable IDNA replacements 
in the areas as Andrew considers. They decided that they had to 
consider and to discuss interoperability with IAB. I regreted that, 
but IETF does not govern the Internet, it only strives to influence 
those who design, use and manage it (RFC 3935) so it works better. So 
it is up to ICANN to document their cons and pros, and to refer to 
them or not in their contracts depending on the situations. It is up 
to the TLD Managers and to the users to use them or not.

The best we can do is to clearly, urgently and precisely document the 
IDNA protocol and its applicability limitations in the security section.
jfc