W3C home > Mailing lists > Public > ietf-http-wg@w3.org > July to September 2014

Re: #537: Remove segments (consensus call)

From: Greg Wilkins <gregw@intalio.com>
Date: Thu, 17 Jul 2014 10:21:40 +1000
Message-ID: <CAH_y2NGr4zxNJoCp1UKysGAX9xMeqYtKTvzK--1bze8X-JsCbA@mail.gmail.com>
To: HTTP Working Group <ietf-http-wg@w3.org>
On 17 July 2014 08:39, Martin Thomson <martin.thomson@gmail.com> wrote:

> Would you prefer an intermediary who knows what you are doing and has
> provided a positive indication of that,
>

In a perfect world the client, intermediaries and server would all embrace
in a group hug to celebrate their acceptance of the same semantics


> or an intermediary who is
> merely unwittingly passing your packets?
>

Failing living in a perfect world, then unwittingly passing on packets is
possibly better than discarding them.

However, I think the horse has bolted on this one and there are lots of
other impediments to intermediaries being able to mindlessly forward new
protocols.    So I am actually OK with removing segments on the basis that
generalising without use-cases is bad.

In some parallel universe we are chartered to make h2 carry both http and
websocket semantic and group hugs are more frequent in that dimension.

cheers

PS> I really fear that websockets over h2 is going to end up with a
long-polling style request stream that will carry push promises for each
message from server to client, just so that the traffic will look like http
and pass through http only intermediaries.

-- 
Greg Wilkins <gregw@intalio.com>
http://eclipse.org/jetty HTTP, SPDY, Websocket server and client that scales
http://www.webtide.com  advice and support for jetty and cometd.
Received on Thursday, 17 July 2014 00:22:08 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 30 March 2016 09:57:09 UTC