At 07:53 PM 1/17/02 -0500, Simon St.Laurent wrote:
On Thu, 2002-01-17 at 18:32, Graham Klyne wrote:
> Well, we now have ways to express other (non-primary) content types.
>
> E.g., per RFC 2912, RFC 2913, ...
>
> Content-Type: multipart/mixed; boundary="break"
> Content-features:
> (& (Type="text/plain") (Type="image/jpeg") (Type="audio/wav") )
Can we simply say (for an XHTML document also containing SVG, MathML,
SMIL, and XLink):
Content-Type: application/xhtml+xml
Content-features: (&
(xmlns="http://www.w3.org/1999/xhtml";)
(xmlns="http://www.w3.org/2000/svg";)
(xmlns="http://www.w3.org/1998/Math/MathML";)
(xmlns="http://www.w3.org/2001/SMIL20/";)
(xmlns="http://www.w3.org/1999/xlink";)
)
Sure... if feature tag xmlns is registered (per RFC 2506). (Something like
this is almost anticipated by CC/PP, which mentions a "schema" attribute -
described in terms of schema identification rather than namespaces).
If so, then we may already be out of the brush, though my reading of RFC
2913 suggests that we would need to add xmlns to RFC 3023's existing
structure.
RFC 2913 and RFC 3023 are almost orthogonal ... there's arguably a very
small overlap of functionality (between RFC 2912+2913 and RFC 3023), but
nothing about using RFC 2913 that I can see requires any change to RFC
3023. Note that RFC 2913 specifically excludes use of content-type
parameters, so adding xmlns to RFC 3023 wouldn't change that, and would
create a more significant overlap.
#g
--------------------------
__
/\ \ Graham Klyne
/ \ \ (GK(_at_)ACM(_dot_)ORG)
/ /\ \ \
/ / /\ \ \
/ / /__\_\ \
/ / /________\
\/___________/