- From: Martin Thomson <martin.thomson@gmail.com>
- Date: Thu, 6 Mar 2014 14:04:09 +0000
- To: Julian Reschke <julian.reschke@gmx.de>
- Cc: William Chan (ιζΊζ) <willchan@chromium.org>, Roberto Peon <grmocg@gmail.com>, Mark Nottingham <mnot@mnot.net>, HTTP Working Group <ietf-http-wg@w3.org>
On 6 March 2014 14:00, Julian Reschke <julian.reschke@gmx.de> wrote:
>> When we say we want to do this, does that mean we have to put in Section
>> 3.6 [1] too in the HTTP/2 spec or something normatively referenced?
>>
>> [1]:
>> http://tools.ietf.org/html/draft-nottingham-httpbis-alt-svc-03#section-3.6
>> ("Proposal: Discovery of TLS Support for http:// URIs")
>
>
> Would you be ok with it if there was no requirement to implement it?
I think that this would absolutely have to be optional. I don't think
that we can realistically force implementations to do things with
Alt-Svc.
The big question on Section 3.6, to me, is whether we need to actively
*forbid* or discourage its use. Otherwise, I think that you end up
with it happening as a byproduct of the other functions that Alt-Svc
defines.
Received on Thursday, 6 March 2014 14:04:37 UTC