Re: XSLT 2.0 HTML output of namespace nodes

> As a matter of historical fact, I don't know what the original rationale
> was. I have assumed in the past, with no particular evidence, that it was
> done for the convenience of the users of products that allow HTML documents
> to contain XML data islands.

I just think that's such an edge case that it makes more sense to
have the default behavior of the HTML output method to output something
that is more HTML-like than XML-like. If users need to override this
for their wacky 'data islands'.

> I can't see any particular reason for doing anything else with elements that
> are not part of the HTML specification, unless we adopt the draconian step
> of saying that everything output by the HTML output method must be valid
> HTML.

Taking it to that extreme wasn't part of my argument.

I just don't think that an XSLT processor should only be expected to go beyond
the definition of HTML when serializing a result tree as "HTML". To require
that it do XML-like things, rather than making this an option that is not
chosen by default, is getting into the territory of guessing what pseudo-HTML
a minority of users might want to abuse this output method for. Most XSLT
users don't want to produce pseudo-HTML with prefixed element names and/or
xmlns attributes. So this is the reason that I see for doing something else.

   - Mike
____________________________________________________________________________
  mike j. brown                   |  xml/xslt: http://skew.org/xml/
  denver/boulder, colorado, usa   |  personal: http://hyperreal.org/~mike/


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list