- From: Martin Thomson <martin.thomson@gmail.com>
- Date: Tue, 26 Aug 2014 16:22:13 -0700
- To: William Chow <wchow@mobolize.com>
- Cc: "Roy T. Fielding" <fielding@gbiv.com>, Mark Nottingham <mnot@mnot.net>, HTTP Working Group <ietf-http-wg@w3.org>
On 26 August 2014 16:07, William Chow <wchow@mobolize.com> wrote: > What I am hoping we'd be able to do is to include language that describes the relationship between the original request/response and the pushed responses, so that caches have guidance on when it is safe to serve a pushed response when the client-initiated request is actually received later from the UA. Any association between a request and any server pushes it generates is largely incidental from the perspective of what we are talking about right now (caching, validity, general usability). Each request-response pair can be considered completely independent in that regard, in fact, in virtually all respects. Caches can and should treat each independently. The only tricky part is to determine when it is best to push something when you don't have any cues from the server. You could, for instance, remember the set of pushed resources that came in response to a given request and push those too, or you could build new heuristics around push for origin servers that don't generate it. That's something that is up to you.
Received on Tuesday, 26 August 2014 23:22:40 UTC