- From: Willy Tarreau <w@1wt.eu>
- Date: Thu, 25 Oct 2012 09:03:59 +0200
- To: Takeshi Yoshino <tyoshino@google.com>
- Cc: Mark Nottingham <mnot@mnot.net>, "William Chan (?????????)" <willchan@chromium.org>, "ietf-http-wg@w3.org Group" <ietf-http-wg@w3.org>
Hi Takeshi, On Wed, Oct 24, 2012 at 02:12:44PM +0900, Takeshi Yoshino wrote: > > > Do you mean the Chromium HTTP upgrade experiment agl referred to in > > > http://www.ietf.org/mail-archive/web/tls/current/msg05593.html? > > > > No, IIRC there was also some broader experimentation using ads; I'll dig > > around a bit more. > > > > > Maybe this work http://w2spconf.com/2011/papers/websocket.pdf which > resulted in introduction of masking to the WebSocket protocol. Yes I think this is the work Mark was talking about, but warning, there were some issues with this test due to the inability to analyse intermediaries behaviour. While very useful, we must keep in mind that several of the issues detected were assigned a possible cause that justified masking (eg: connection resets supposed to be caused by interpretation of payload contents as requests), while a number of other causes were much more likely true for a number of us implemntors (eg: classic close upon response receipt). I don't have all the details in mind anymore, this was discussed to such great lengths on the hybi list :-) Today I'd much rather trust real-world tests from websocket clients running in browsers ! Willy
Received on Thursday, 25 October 2012 07:04:35 UTC