Sorry I should've elaborated a little more...
I'm using a xslt 1.0 stylesheet specifying xsl:output method="xhtml"
with saxon 7.8
This allows me to post-process the result of my transform with xml
tools, whilst ensuring I don't have any minimised elements so that IE
copes.
I suppose I could just make the final phase output html, but I would
prefer to use xhtml throughout (saves times when debugging).
The reason I use ascii is so that character references (no entities
involved) remain as character references through to the output, so at no
point do I have multi-byte characters (which caused me some problems in
java :).
All of that aside, I wondered why a processor must include a declaration
when the output encoding is ascii, as the default input encoding is
utf-8 (of which ascii is a subset)?
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list