[Lake] EDHOC Review

[Kathleen Moriarty <[e-mail]>]

Greetings!

I had offered to contribute a review at the last interim and am very glad
to see this document come to this part of the process after the large
efforts that went into its development and demonstrating it's value for use
with constrained devices.

Here are a few nits to consider:

Section 1.1 Nit

OLD:

EDHOC does

   currently not support pre-shared key (PSK) authentication as

   authentication with static Diffie-Hellman public keys by reference

   produces equally small message sizes but with much simpler key

   distribution and identity protection.


NEW:

EDHOC does not

   currently support pre-shared key (PSK) authentication as

   authentication with static Diffie-Hellman public keys by reference

   produces equally small message sizes but with much simpler key

   distribution and identity protection.


Section 1.2:


The intent of the following sentence is to convey that these libraries are
already in use for OSCORE, but the wording of the following sentence could
be a bit more clear:

OLD:

By reusing existing libraries, the additional code size can be kept very

   low.

PROPOSED:

In using libraries already in the code base for OSCORE, the additional code
size can be kept very

   low.



Section 3.8

S/enrolment/enrollment/


Section 4.3

S/kan/can/

In the following sentence: in most encryption algorithms the same key kan be


IANA Registries


I see for the registries created that Expert review [RFC8126] is required.
What documentation is required? Is it also Specification required or is
there other guidance for the experts when considering updates? I see this
is discussed in 9.14, but perhaps adding specification recommended in each
of the places a registry is created would be helpful.


Thank you for your work on this document and protocol!
-- 

Best regards,
Kathleen