Re: [ANN] Kludgey workarounds for IE and Netscape
James Clark (jjc@jclark.com)
Fri, 11 Sep 1998 17:40:33 +0700
Andy Dent wrote:
>
> At 5:50 PM +0800 11/9/98, James Clark wrote:
> >There needs to be a
> >XML representation specified because an XSL stylesheet specifies its
> >result as XML.
> Umm, I'm not sure I can agree.
>
> The XSL parsers that exist, and the standard, specify the result as XML.
>
> As far as I can see there's nothing saying the output of XSL+XML can't be
> an invisible set of c++ objects.
How are we disagreeing? I said "specifies its result as XML" not
"specifies that its result is XML". In other words it describes its
result in terms of an XML document. That XML document doesn't have to
be created. An XSL processor is perfectly entitled to create C++
rendering objects that are described by the XML instead of objects
representing XML elements.
James