FSIs were discussed at the beginning. A decision was made that they
were better left for later, and I agree.
A decision was also made that all system identifiers would have an
implicit FSI identifier of <URL>, which I also think is usually a good
idea. This allows FSIs to be added later, and any unlabeled system ID
is implied to have <URL>.
What I was suggesting on the SIG was that for system identifiers in
notation declarations, the assumed FSI label would be <mimetype>. As
Rick pointed out, this is legal HyTime 2 FSI notation, and would be
very useful. However, the WG has made its decision.
I believe that XML authors are largely going to refer to images simply
by URLs instead of entities; in that case, file system associations or
HTTP headers can be used to ascertain the entity's type. In cases
where NDATA entities are used, I would recommend that XML implementors
ignore the system identifier of the notation, and make their decision
based on the entity itself.
-Chris
-- <!NOTATION SGML.Geek PUBLIC "-//Anonymous//NOTATION SGML Geek//EN"> <!ENTITY crism PUBLIC "-//O'Reilly//NONSGML Christopher R. Maden//EN" "<URL>http://www.oreilly.com/people/staff/crism/ <TEL>+1.617.499.7487 <USMAIL>90 Sherman Street, Cambridge, MA 02140 USA" NDATA SGML.Geek>