> This is actually pretty simple to do in XSLT, *if* you have a processor
> that supports an encoding like "US-ASCII" for serialization.
yes, but going through XSLT is a high price to pay:
* it expands entities
* it loses DTD subset and doctype
* it expands CDATA sections (unless you take ad hoc care of certain elements)
on a daily basis, I'd find these features an irritation.
Sebastian Rahtz OUCS Information Manager
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431