- From: Martin Thomson <martin.thomson@gmail.com>
- Date: Wed, 4 Nov 2015 09:10:51 +0900
- To: James M Snell <jasnell@gmail.com>
- Cc: "ietf-http-wg@w3.org" <ietf-http-wg@w3.org>
On 4 November 2015 at 00:36, James M Snell <jasnell@gmail.com> wrote: > Has a recommended best practice emerged among > implementations for ensuring that unsafe requests are not sent or > handled out of order? I don't think that we're in any different a situation to where we were in the past. There are plenty of cases in HTTP/1.1 where non-safe methods are multiplexed, they just happen to use connections rather than streams. I can infer from this pattern that the defacto approach is that applications that care about ordering need to provide their own ordering guarantees. If pipelining were more widely used, or if it were the case that people didn't use multiple concurrent connections to a server, I think that this would more of an issue.
Received on Wednesday, 4 November 2015 00:11:19 UTC