On Thu, 29 Nov 2007 17:14:10 +0100, <[EMAIL PROTECTED]> wrote:
Comment:At the end of the section on the send method, this para appears:
--
If the user agent implements server-driven content-negotiation it should
set Accept-Language, Accept-Encoding and Accept-Charset headers as
appropriate; it must not automatically set the Accept header. Responses
to such requests must have the content-encodings automatically decoded.
[RFC2616]
--
The Accept-Language header is currently the only mechanism available in
XHR for locale management. It may be important for locale-sensitive
interactions to convey a language or locale to the server. Thus, it
would be useful to separately mention the Accept-Language header and its
use in informing the server of language/locale preference. In addition,
we suggest you recommend the use of the BCP 47\'s Lookup algorithm(found
in the RFC 4647 portion of BCP 47) for matching the A-L header.
I think it actually makes sense to require user agents to not set the
Accept-Language as appropriate so that content authors can do that. Making
encoding as transparant as possible from the author point of view makes
sense, but language doesn't really.
I'm going to make a change that will give Accept-Language identical
treatment to Accept.
Note: the normative words "should", "must not", and "must" all appear in
non-normative form.
That's because you copy and paste in a way that does not preserve markup.
(I sense an i18n issue here! ;-))
--
Anne van Kesteren
<http://annevankesteren.nl/>
<http://www.opera.com/>
