Re: Can we revise RFC3023?

Bjoern Hoehrmann wrote:

>>1. Deprecate text/* for anything that's in XML.  That's because it 
>>forces the provider to provide a charset header, because in its absence 
>>the receiver is required to assume either ASCII or 8859 depending on the 
>>context,
> 
> No, the implementation is required to default to us-ascii, period.

Roy and TimBL tell me that in the case of HTTP it defaults to 8859-1.  I 
haven't checked but they're unlikely to be wrong.  In any case, it's 
still a problem.

>>2. Deprecate the charset parameter for application/xml and 
>>application/*+xml.
> 
> I disagree. XML 1.0 should be changed in a way that makes it a fatal
> error if the encoding specified in the XML declaration or encoding
> declaration is different from higher-level protocol encoding
> information.

It already is, I believe.  If the server provides charset=, it can only 
increase the chances of this error occurring, so the server in general 
SHOULD NOT do this.

-- 
Cheers, Tim Bray
         (ongoing fragmented essay: http://www.tbray.org/ongoing/)

Received on Wednesday, 17 September 2003 18:15:28 UTC