Re: [PANRG] RG Last Call for draft-irtf-panrg-questions

[Theresa Enghardt <[e-mail]>]

Dear PANRG,

I just read this document again and I think it's ready for publication.

In fact, I think this document nicely captures the key questions that
we're working on in PANRG, and I expect this to be interesting to people
outside of PANRG as well, so it really should be published.


Just a couple of nits that I found:

- Typo in abstract: "intetnetworks"

- Normative vs informative references: It's not clear to me why RFC 4302
(IP Authentication Header) and RFC 5246 (TLS 1.2) are Normative, while
RFC 7624 (on Passive Surveillance) and RFC 8170 (on Protocol Adoption
and Transition) are Informative. Maybe all of these should be Informative.

- For TLS, perhaps the reference should be updated to TLS 1.3.


Finally, one suggested addition:

Section 2.3, "Supporting Path Selection", states that "Path selection
   must, like path property information, be trustworthy, such that the
   result of a path selection at an endpoint is predictable."

I think it might be worth noting that results should not only be
predictable, but should also result in an outcome that is "better than
the default" or "better than random". Maybe surprisingly, this is not
trivial, as we also noted the Path Properties draft (Section 3.1) after
somebody (I forget who) brought up this point in a PANRG session.

I suggest to add something like the following phrase:

"Moreover, any path selection mechanism should aim to provide an outcome
that is not worse than using a single path, or selecting paths at
random, e.g., in terms of performance."


Best,
Theresa