An addition to my previous advocacy of event-based processing.
It may be worth defining/supplying classes for a standard tree
representation, and a standard builder class that can create those -- so we
define the stream, but provide a ready-cooked abstract sytnax tree
representation for those who want it.
Personally I see groves as semantically equivalent to the API, but as so
obscure that they should not be a priority for our explanations of the API.
We probably need to refer to groves, for those who already know about them.
However, very few of our target audience of implementors will already know
about groves, nor will they care, if they can understand the API without
them.
I see XML-groves and XML-API as parallel and needing to be in synch. I
don't see either as having to depend on the other, though, and frankly,
given the relative penetration of groves and Java into the "global
developer consciousness", I don't see groves as that high a priority.
-- 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://dynamicDiagrams.com/
MAPA: mapping for the WWW \__________________________