W3C home > Mailing lists > Public > public-html@w3.org > July 2008

Re: websocket HTTP response parsing

From: Ian Hickson <ian@hixie.ch>
Date: Thu, 3 Jul 2008 16:19:38 +0000 (UTC)
To: Julian Reschke <julian.reschke@gmx.de>
Cc: "public-html@w3.org" <public-html@w3.org>
Message-ID: <Pine.LNX.4.62.0807031617050.13974@hixie.dreamhostps.com>

On Thu, 3 Jul 2008, Julian Reschke wrote:
> Ian Hickson wrote:
> > ...
> > This isn't HTTP. It's WSP (Web Socket Protocol), which while it may vaguely
> > look like HTTP in the handshake, really isn't HTTP and shouldn't be treated
> > like HTTP.
> > ...
> 
> So why then use a message format that looks like HTTP/1.1 in the first place?

So that it is possible to run Web Sockets over port 80 (or 443) and have 
the HTTP server defer to a Web Socket server on a URL-by-URL basis. 
Basically the protocol is intentionally HTTP-like to make it possible to 
use it in environments that are currently expecting HTTP, without in fact 
requiring people implement HTTP if they don't actually need it.


> And also, why not require an empty reason phrase (which would get rid of 
> having plain text messages in English hardwired into the protocol)?

Given that the handshake blob is basically opaque, what's wrong with 
having hardwired? The previous version had the handshake be just "Hello" 
"Welcome", which is just as hardwired to English.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
Received on Thursday, 3 July 2008 16:20:14 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:38:56 UTC