The initial congestion window is a size in bytes, so I think it has to be a size.
This paragraph is largely repeating text in RFC9002. We could also just remove it? On Mon, Mar 27, 2023 at 8:44 PM Gorry Fairhurst <[email protected]> wrote: > On 27/03/2023 04:06, ianswett wrote: > > Closed #167 <https://github.com/quicwg/ack-frequency/issues/167> as > completed. > > — > Reply to this email directly, view it on GitHub > <https://github.com/quicwg/ack-frequency/issues/167#event-8849156643>, or > unsubscribe > <https://github.com/notifications/unsubscribe-auth/ABYLLEWEC4NCKSNFSOGPG5LW6D73DANCNFSM6AAAAAAVYZMFAA> > . > You are receiving this because you authored the thread.Message ID: > <quicwg/ack-frequency/issue/167/issue_event/[email protected]> > > This seems to have closed, without answering the comment, so my suggestion > was to change: > > OLD: > > "In keeping with Section 7.7 of [QUIC-RECOVERY], a sender > can either employ pacing or limit bursts to the initial congestion > window." > > When I re-read Section 7.7 of [QUIC-RECOVERY], and found an issue in that > the text does not precisely say what I thought was intended, it says: > > "Senders SHOULD limit bursts to the initial congestion window", > > **I ** think the above existing text in RFC9002 refers to limiting bursts > to the size of IW, but I received a question whether this could be read as > during the time when cwnd<=IW. What do we intend? > > Can we write NEW: > > "In keeping with Section 7.7 of [QUIC-RECOVERY], a sender > can either employ pacing or limit bursts to the size of the initial > congestion > window." > > Gorry > >
