Updating TCP to Support Rate-Limited Traffic
Note: This ballot was opened for revision 11 and is now closed.
(Spencer Dawkins) Yes
(Martin Stiemerling) Yes
(Jari Arkko) No Objection
(Alia Atlas) No Objection
Deborah Brungard No Objection
(Ben Campbell) No Objection
Comment (2015-06-23 for -12)
It appears that this draft obsoletes, makes historic, and according to section 1.2, updates 2861 :-) I guess it's okay to obsolete and make it historic in one fell swoop, but it might be worth removing "updates" from the first sentence of 1.2. My inner pedant concurs with Barry's.
(Brian Haberman) No Objection
(Joel Jaeggli) No Objection
Barry Leiba No Objection
Comment (2015-06-22 for -12)
Just a few very minor comments -- nothing that needs any discussion. -- Section 1 -- RFC 2616 is an obsolete reference for HTTP. The current reference is RFC 7230 o To incentivise the use of long-lived connections Will you please appease my inner pedant and not say "incentivise" (which, by the way, Chrome doesn't think is a word either)? You can take advantage of the previous bullet's use of "To remove the incentive for" by using this parallel construction: "To provide an incentive for the use of long-lived connections". -- Section 4.2 -- The method RECOMMENDS that the TCP SACK option [RFC2018] is enabled and the method defined in [RFC6675] is used to recover missing segments. Even more ridiculously pedantic than the other: subjunctive mood with "recommends", please. Make both "is" into "be". -- Section 4.4 -- A TCP sender implementing this specification MUST enter the non- validated phase when the pipeACK is less than (1/2)*cwnd. Given that there are "MAY"s and a "SHOULD" involved in how pipeACK is computed, it seems rather odd to have a MUST that relates to its value. You couldn't possibly determine whether an implementation was doing this "right" because there's so much variability in the value of pipeACK anyway. No need to discuss this, but I suggest that you just do this: NEW A TCP sender implementing this specification enters the non- validated phase when the pipeACK is less than (1/2)*cwnd. END
(Kathleen Moriarty) No Objection
Comment (2015-06-24 for -12)
It does not appear that the SecDir review got a response, maybe the editor and shepherd didn't see it? https://www.ietf.org/mail-archive/web/secdir/current/msg05697.html He included a bunch of nits that might be helpful.