that the next great network or transport protocol gets a chance at
working here. Extension points for layer3 and layer4 are probably a
good idea. Fortunately, UPA isn't a problem, so you can make the
simple type a union of what you have and xs:token (or some restriction
thereof in order to restrict the valid set of characters to something
sensible).
I do think that your justification could be made clearer. You wish to
identify a host on the remote side of a NAT by identifying a flow that
you know that host is participating in. We all got this one wrong.
--Martin
On 29 February 2012 08:45, Ray Bellis <Ray.Bellis@nominet.org.uk> wrote:
> It's only the complete 6-tuple that uniquely identifies the flow, and hence
> the internal target behind that NAT.
_______________________________________________
Geopriv mailing list
Geopriv@ietf.org
https://www.ietf.org/mailman/listinfo/geopriv
