- From: Mike Bishop <Michael.Bishop@microsoft.com>
- Date: Wed, 3 Feb 2016 19:01:48 +0000
- To: Martin Thomson <martin.thomson@gmail.com>, Amos Jeffries <squid3@treenet.co.nz>
- CC: HTTP Working Group <ietf-http-wg@w3.org>
We had talked at one point about intentionally pushing very specific conditional requests which would always have a 304 response. The client, upon receipt, would either know that the resource in the cache was fresh (and could update cache lifetimes) or realize that an object they will need is not in the cache, and request it. Not something that we've implemented yet, but it's a valid scenario. -----Original Message----- From: Martin Thomson [mailto:martin.thomson@gmail.com] Sent: Tuesday, February 2, 2016 8:19 PM To: Amos Jeffries <squid3@treenet.co.nz> Cc: HTTP Working Group <ietf-http-wg@w3.org> Subject: Re: Submitted new I-D: Cache Digests for HTTP/2 On 3 February 2016 at 07:26, Amos Jeffries <squid3@treenet.co.nz> wrote: > That is because PUSH by itself cannot do conditional requests. > Consider it to be the equivalent for a non-conditional request that > always gets the 200 status response with full new object and new expiry details. This isn't strictly true. You can server push a conditional request, and in fact it can be advantageous to do so. Of course, the server doesn't always know enough about a client to make the *right* conditional request.
Received on Wednesday, 3 February 2016 19:02:21 UTC