If we take the view that nothing can change, then we will be correct.
Nothing will change.
At that point, the best possible outcome is NAT66, which is not much of
a "best" in my book.
LISP, and similar schemes, will provide better NAT support for limited
cases.
But we will not have an architecture. We will have an assemeblage of parts.
It will keep working. But it will make various things harder and
harder. Referrals are only the first thing to fall victim to such an
assemblage.
There are ways to change more things, with transition strategies that
enable communication in the intervening cases. (Saying that we must
change the host, without dealing with the intermediate / mixed cases
would indeed be a non-starter.)
But, as long as we assume we can not do so, then we won't.
Yours,
Joel Halpern
_______________________________________________
rrg mailing list
[email protected]
https://www.irtf.org/mailman/listinfo/rrg
