Right now, J's xml support is based on sax rather than dom.

sax is simple in the way that sequential machine is simple: it is
fast, but it takes a bit of thought to use it.

But what really gets me is this statement in the J wiki:

  "This paradigm results in systematically faster XML processing than
DOM, as the SAX stream has a tiny memory footprint."

Presumably, most J programmers should be familiar with the character
of space/time tradeoffs, as well as the character of development time
vs execution time tradeoffs.  In other words: whether it is faster in
practice is really an issue for benchmarks, and even then, speed might
not be a pertinent issue.

That said, we do not have a dom implementation of xml support to work with.

But maybe we should?

-- 
Raul
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to