- From: Martin Thomson <martin.thomson@gmail.com>
- Date: Thu, 17 Jul 2014 13:16:16 -0700
- To: Kari Hurtta <hurtta-ietf@elmme-mailer.org>
- Cc: Yutaka Hirano <yhirano@google.com>, HTTPBIS working group mailing list <ietf-http-wg@w3.org>
On 17 July 2014 13:06, Kari Hurtta <hurtta-ietf@elmme-mailer.org> wrote: > seems allow :scheme = ws[s] This is true, though you aren't really getting what you think at that point. See RFC 7230: HTTP is also designed for use as an intermediation protocol for translating communication to and from non-HTTP information systems. HTTP proxies and gateways can provide access to alternative information services by translating their diverse protocols into a hypertext format that can be viewed and manipulated by clients in the same way as HTTP services. The implication here being that the data from the non-HTTP information system can be put in an HTTP message. > So I guess that intermediaries does not need care about > :scheme. Is there anything why web server needs care > about :scheme request field ? Looks like that web > server may "safely" ignore it. That entirely depends on what the intermediary is doing. A tunnel can probably ignore the field. > In both cases intermediaries need to know that its is full-duplex > communications inside of HTTP/2 stream (that is not HTTP > request/reply exchange). Correct. > In other hand :method = CONNECT does not like You can use CONNECT to do websockets, but that's not the form of the protocol we've been discussing. We've been discussing integration, as opposed to tunneling.
Received on Thursday, 17 July 2014 20:16:43 UTC