Resource-Oriented Lightweight Information Exchange
Note: This ballot was opened for revision 10 and is now closed.
Alexey Melnikov (was Discuss) Yes
Comment (2017-11-16 for -14)
Thank you for addressing my DISCUSS points. I talked to Mark Nottingham about use of RFC5005 link relations and he explained that I was mistaken. However we agreed that slightly modifying link relation definitions from RFC5005 is not the best thing. I am still concerned that user authentication is under-specified, but I can live with existing text.
Kathleen Moriarty Yes
Alia Atlas No Objection
Ben Campbell (was Discuss) No Objection
Comment (2017-11-15 for -14)
Thanks for addressing my DISCUSS points and other comments!
Alissa Cooper No Objection
Comment (2017-10-25 for -11)
The schema registration in Section 8.1 says "See section A of this document." I think you mean Appendix A? Also, as with Ben I don't understand what it implies for the schema to be informative. Section 8.3 says "Designated Expert reviews should be routed through the MILE WG mailing list. Failing this, the Designated Expert will be assigned by the IESG." Might it be better to assign a back-up designated expert at the time of the document's approval, so that we don't end up in a situation of prolonging a registration delay by having to find and approve an expert after the WG mailing list has proved unresponsive?
Spencer Dawkins No Objection
Suresh Krishnan No Objection
Mirja Kühlewind No Objection
Eric Rescorla No Objection
Comment (2017-10-20 for -11)
OVERALL I share Martin Thomson's concerns about the restriction on 0-RTT. In the discussion, I saw two sets of concerns about 0-RTT: - Replay - Lack of FS As Martin says, the replay issue is an issue for the HTTP profile, so any concerns should be directed there. I agree that 0-RTT has inferior FS properties, but it's worth noting that TLS 1.2 session resumption with tickets has FS properties that are as bad or worse than those with TLS 1.3 0-RTT, and I don't see a prohibition here on session resumption. This leaves me a bit unclear on the security rationale here, and I think this needs to be consistent. INLINE COMMENTS or serialization. This approach allows the provider to support multiple, compatible formats allowing the consumer to select the most suitable version. What does "compatible" mean here. Do you mean isomorphic? supporting interactive user logins by members of the consortium SHOULD support client authentication via a federated identity scheme. Such as? Proper usage of TLS as described in Section 5.3 will in many cases aid in the mitigation of these issues. You should also note that TLS 1.2 and lower client auth leaks the user's identity to on-the-wire attackers. supported. TLS 1.2 SHOULD be implemented according to all recommendations and best practices present in [RFC7525]. You need a citation to 6125 about valiation, though I realize that 7525 cites it.
Alvaro Retana No Objection
Adam Roach (was Discuss) No Objection
Comment (2017-11-15 for -14)
Thanks for addressing my discuss points and comments.