comments on draft-ietf-v6ops-nat64-pb-statement-req-00

[Dave Thaler <[e-mail]>]

Section 3.1:
The taxonomy is missing several unfortunately common cases.

a)      What if a human types in the address?

b)      What about a URL with an IP literal that you get in email or an instant message?

c)       What about apps that "hard code" addresses (e.g., in a config file)?

In the paragraph on referrals, I think "nodes" should be "applications" thoughout.

Section 3.2
Re "... it seems unlikely
that v4-only nodes will have a strong need to communicate with v6-
only nodes (at least at the initial stages of v6 deployment)."
It depends on what you mean by v6-only.  If it's a dual-stack node that can't get
a public IPv4 address, that doesn't seem unlikely.

Typo "or f other"
Typo "need to e modified"

Re "NAT64 box internally.  The last case, where the v4 node has public
address and the NAT64 box has a private address seems harder to
justify though."
Just put a normal (carrier grade) NAT between the NAT64 and the internet
and you have this case starting from the Public v4/Public v4 case.

Section 4.1
Re "unless explicitly stated in the particular requirement.  The
translation mechanism MAY require changes to v6-only nodes."
Clarify.  What does v6-only node mean?  Not IPv4-capable?
Connected to an IPv6-only network?

-Dave