[jdom-interest] Question about JDOM & memory usage

[Mike Hatalski]

In Jason Hunter & Brett McLaughlin's JavaWorld article "Easy Java/XML
integration w/ JDOM", there's a quote:
"(JDOM)... does not require the entire document to be in memory. The API
allows
for future flyweight implementations that load information only when needed"

I'm wondering what work has/is being done in that area, and what's that
status?

Alternatively, does anybody have any suggestions for using JDOM in working
with
large XML document that are relatively "flat" (i.e. the trees are wide, not
deep).  Given some element with a lot of children, it'd be nice to begin
processing the children before the entire document has been parsed into
memory,
and also to be able to release a child's memory once it has been processed.

Thanks,
--Mike