W3C home > Mailing lists > Public > ietf-http-wg@w3.org > April to June 2011

Re: #285: Strength of requirements on Accept re: 406

From: Mark Nottingham <mnot@mnot.net>
Date: Thu, 23 Jun 2011 09:27:18 +1000
Cc: Adrien de Croy <adrien@qbik.com>, httpbis Group <ietf-http-wg@w3.org>
Message-Id: <3B675866-3A99-408D-B6DB-1C1E6F35908A@mnot.net>
To: Julian Reschke <julian.reschke@gmx.de>
I'll point out that this text was in 2616:
  http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.3
and also in 2068 (section 14.3).

So, we've gotten this far without the Web blowing up, and as the link below points out, people are relying on this behaviour; i.e., if we change the requirement, we'll make servers (e.g., google.com) non-conformant. 



On 22/06/2011, at 10:20 PM, Julian Reschke wrote:

> On 2011-06-22 11:18, Mark Nottingham wrote:
>> ...
>> Some background here:
>>   http://www.stevesouders.com/blog/2010/07/12/velocity-forcing-gzip-compression/
>> ...
> 
> If I read this correctly, this is a workaround for broken intermediaries (and maybe IE6 behind proxies).
> 
> Do we really need to address this in the spec? And does it affect anything beyond Accept-Encoding?
> 
> Are we sure that 406 is supposed to be about all "Accept-*" headers?
> 
> Maybe a compromise would be to mention that some intermediaries are known to eat "Accept-Encoding", and thus origin servers can pick Content-Codings of their choice if they are sure the recipient will be able to process them.
> 
> Best regards, Julian

--
Mark Nottingham   http://www.mnot.net/
Received on Wednesday, 22 June 2011 23:27:47 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 06:51:41 GMT