Some of us think this was a significant reduction in the power of XML to
represent useful information (the difference between an element that marks
a point phenomenon, and one that is empty because it just doesn't have any
content).
>That begs the question of the processor's information set, however --
>a processor designed for use with repositories or with editors, for
>example, needs to preserve lexical as well as structural information
>about the XML document, such as comments, general entity references
>(even within attribute values), specified vs. defaulted attribute
>values, CDATA sections, whitespace within tags, etc.
It's not possible to write _valid_ SGML document instances without this
information, something not true of comments, DTD, info, or the other
lexical information.
I think the EMPTY declaration status, and the lexical form of the element
occurrence are useful for that practical reason alone.
>SAX as it currently stands is not designed to preserve most lexical
>information; in the future, we may devise a SAX level-2 to return this
>information, but since most applications that need it will probably
>use a DOM anyway, the demand may not be strong enough.
This information is more than purely lexical, which is why it should be in
there...
-- David
_________________________________________
David Durand dgd@cs.bu.edu \ david@dynamicDiagrams.com
Boston University Computer Science \ Sr. Analyst
http://www.cs.bu.edu/students/grads/dgd/ \ Dynamic Diagrams
--------------------------------------------\ http://www.dynamicDiagrams.com/
MAPA: mapping for the WWW \__________________________