- From: Mark Nottingham <mnot@mnot.net>
- Date: Tue, 15 Oct 2013 12:22:15 -0700
- To: Martin Thomson <martin.thomson@gmail.com>
- Cc: "ietf-http-wg@w3.org Group" <ietf-http-wg@w3.org>
On 15/10/2013, at 10:30 AM, Martin Thomson <martin.thomson@gmail.com> wrote: > On 15 October 2013 00:57, Mark Nottingham <mnot@mnot.net> wrote: >> We talked about using a response header for negotiation last week, a la Alternate-Protocol, and one of my action items was to isolate the Alt-Svc proposal. > > Brief comment: You should identify protocols, not URI schemes. > Preferably, that identification should use the same identifiers that > are used in other places. That's the intent. The strings in the draft are just straw men; happy to clarify. > ALPN (and NPN) deployments currently identify "http/1.1", and there > are strings in the ALPN draft that are also (apparently) in use. Even > if the "spdy/1" string might not be relevant, I think that "spdy/3" is > widely used. Yes. The upcoming revision to http2-encryption will explore this a bit more. I haven't got hung up on the syntactic form yet; focus is on the semantics of the protocol identifier. I.e., shed colour advice gladly taken. Cheers, -- Mark Nottingham http://www.mnot.net/
Received on Tuesday, 15 October 2013 19:22:38 UTC