Yes, this is the potential problem if we alter the public
API, which seems to be the "value" as well as the "curse" of
a Java 5 jump. It's why I ponder an org.jdom2 package if we
do that, so the piles of existing code keep working and new
code can be written against the new package. That's not
ideal, but it's practical. What do you think, Michael?
Well, that has a downside too: I will have to fork the net.sf.saxon.jdom
package to create a net.sf.saxon.jdom2 package, and maintain and test both.
That's an extra 1000 or so tests to run each time I make a release. But it's
probably the lesser evil.