- From: Patrick McManus <mcmanus@ducksong.com>
- Date: Tue, 9 Dec 2014 12:42:27 -0500
- To: HTTP Working Group <ietf-http-wg@w3.org>
- Message-ID: <CAOdDvNqe1vPgTR2M4XX5AFq6zYx7pq1YCk1ABXN+KDQM3UqvTg@mail.gmail.com>
These suggestions are editorial in nature only. 5.3.4 currently reads " Similarly, streams that are in the "idle" state can be assigned priority or become a parent of other streams. This allows for the creation of a grouping node in the dependency tree, which enables more flexible expressions of priority. *Idle streams that are made a parent of another stream are assigned a default priority* (Section 5.3.5)." I believe that last bolded bit isn't quite right. Idle-Unprioritized streams that are made the parent of another stream are assigned the default priority - but a group node can be in the idle state and explicitly have its priority set.. at which point it can be made a parent of another stream and should not be assigned the default priority as doing so rather undermines the usefulness of a group node. I propose: " Similarly, streams that are in the "idle" state can be assigned priority or become a parent of other streams. This allows for the creation of a grouping node in the dependency tree, which enables more flexible expressions of priority. Idle streams begin with a default priority (Section 5.3.5)." also in 6.3 "The PRIORITY frame is associated with an existing stream. If a PRIORITY frame is received with a stream identifier of 0x0, the recipient MUST respond with a connection error (Section 5.4.1) of type PROTOCOL_ERROR." To say associated with an existing stream begs the question of states qualify as existing (idle? closed?) I propose just editing out the first sentence: "If a PRIORITY frame is received with a stream identifier of 0x0, the recipient MUST respond with a connection error (Section 5.4.1) of type PROTOCOL_ERROR." -Patrick
Received on Tuesday, 9 December 2014 17:43:20 UTC