Re: WG last call on NAT-PT to Experimental

[Despres@psg.com, Remi <despres.remi@wanadoo.fr>]

Thanks for the comments.
Further remarks below.
Remi

> To the reachable only via IPv6, communication will be limited if there is
> not ability to translate. Companies are *very* likely to find translation
> at their DMZ a practical alternative to wide IPv6 deployment within
> themselves while the number of companies that one is forced to use that to
> get to is nominal.
ALG translations in DMZs are significantly more complex than just
some NAPT-6to4 at the entrance of v6&4 sites where there are v4-only
servers that have to be reached from outside.

> That said, as this draft says, tunnel encapsulation is a more
> end2end-freindly approach where it is applicable,
Agreed, but as you say "when it is applicable"
E2E encapsulation is not relevant to reachability of v4-only host.

> and dual stack
> implementation is obviously to be preferred for any currently-IPv4 network.
Agreed also.
But the subjet is making reachable some "servers" which, for some legitimate
reason(s), are not updated to dual satck (e.g. some egacy small devices like
sensors, or other automata, which today operate within v4 private sites).