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

Re: Possible issue: Accept-language priority based on language order

From: Martin J. Dürst <duerst@it.aoyama.ac.jp>
Date: Thu, 24 Nov 2011 15:41:13 +0900
Message-ID: <4ECDE709.7040803@it.aoyama.ac.jp>
To: Julian Reschke <julian.reschke@gmx.de>
CC: Amos Jeffries <squid3@treenet.co.nz>, ietf-http-wg@w3.org
Hello Julian,

On 2011/11/24 7:41, Julian Reschke wrote:
> On 2011-11-23 23:29, Amos Jeffries wrote:
>> ...
>> I've been keeping an eye on this since implementing language negotiation
>> in Squid.
>> It appears that nearly all agents are sending the language codes sorted
>> by q value anyway. Whether they send the q value or not it is still
>> possible to optimize by using the left-most wins assumption.
>> If anyone is interested in doing a deeper analysis I have a dataset
>> available covering the last year on several networks linking the
>> Accept-Language and User-Agent header pair.
>> ...
> Analysis would be good.

I agree with Harald's analysis. I think it's up to people who want to 
claim the contrary to do some footwork.

I would definitely NOT go as far as Dale and say "ignore the q= values, 
they will be in order". That would explicitly be against the current 
spec. But saying that if there are no q values, then the leftmost 
matching should win will definitely bring the spec and reality closer 

> I'm skeptical because we're not really allowed to make changes breaking
> previously compliant implementations without *very* good reasons.

Would the fact that widespread practice is different from the spec, and 
changing the spec seriously increases interoperability, not be a *very* 
good reason?

> I'm also not too enthusiastic having to consider whether this would be
> *specific* to Accept-Language, or apply to all Accept headers.

I don't know much about the other headers. There is a strong difference 
between Accept-Language (which can be set by the user through 
Options/Preferences in most browsers) and the others (which are mostly 
just baked into the browser).

Regards,    Martin.
Received on Thursday, 24 November 2011 06:41:48 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:13:54 UTC