RE: RDF Media Type (rdaniel@interwoven.com)

>----- Forwarded message from Ron Daniel <rdaniel@interwoven.com> -----
>Should the media type document be extended to define
>text/rdf+xml, and the use of a +rdf+xml suffix, in addition
>to its current definition of application/rdf+xml?
>
>The main reason the XML media types
>document proposed the +xml suffix was because there are
>entirely different formats, such as SVG vs. XHTML, which
>are expressed in XML but need very different software
>to handle. I have assumed the same will happen with RDF,
>where generic RDF software could deal with things as triples,
>but different RDF Schemas would arise for vastly different
>applications, such as Calendaring vs. descriptive metadata.
>Those specific schemas would need their own software for
>dealing with their specific elements.

Just a warning from someone who had to fight very hard to get +xml through 
the IETF:

The MIME community was not very happy about the prospect of multilayered 
(beyond two) media types.  To some extent it's compatibility and politics, 
but it was a rather drawn-out battle to make +xml acceptable.  I suspect 
those same people would be even more unhappy about the prospect of +rdf+xml.

If you want to move forward on this, at least if the IETF's blessing is of 
interesting, explore:
http://www.imc.org/ietf-xml-mime/

I'd suggest putting on a flak jacket and bring it up on that list.  Better 
to follow Tommie Usdin's advice about "inviting the bad fairies" to the 
discussion early than to go a long ways and find out that your proposal 
will be on the margins of MIME types for eternity.

Then there's the tantalizing prospect of replacing MIME types with URIs 
pointing to RDDL...

Simon St.Laurent
Associate Editor
O'Reilly & Associates, Inc.

Received on Tuesday, 24 April 2001 12:37:14 UTC