Re: Value of content negotiation? [was RE: content negotiation anti-principle]

At 9:25 AM -0600 1/9/03, Jeremy Dunck wrote:


>>In a roundabout way, that is what I was proposing as one solution: some means
>>to provide more information in the places that use URI's. Of course, things
>>like ENTITY declarations become somewhat tricky... but I guess this might be
>>one more reason for XInclude...


I think most parsers resolving URIs found in ENTITY declarations 
should be smart enough to ask for application/xml; text/xml. That's 
what they're always going to want anyway.

Unparsed entities are trickier, but nobody uses them anyway, and 
since they have to be resolved form the document rather than the DTD 
(that is, you don't load an unparsed entity until you find an ENTITY 
or ENTITIES type attribute) there's room in the document for an 
additional attribute that specifies the content type desired.
-- 

+-----------------------+------------------------+-------------------+
| Elliotte Rusty Harold | elharo@metalab.unc.edu | Writer/Programmer |
+-----------------------+------------------------+-------------------+
|           Processing XML with Java (Addison-Wesley, 2002)          |
|              http://www.cafeconleche.org/books/xmljava             |
| http://www.amazon.com/exec/obidos/ISBN%3D0201771861/cafeaulaitA  |
+----------------------------------+---------------------------------+
|  Read Cafe au Lait for Java News:  http://www.cafeaulait.org/      |
|  Read Cafe con Leche for XML News: http://www.cafeconleche.org/    |
+----------------------------------+---------------------------------+

Received on Thursday, 9 January 2003 10:50:15 UTC