RE: [docbook-apps] Using XIncludes to include infosets

[Dave Pawson <dpawson@nildram.co.uk>]

At 11:45 02/02/2004, Peter Ring wrote:
>You have my sympathy. It's no fun these days to comply to the recs and
>working drafts.
>
>I think I can articulate the sorry state of xpointer() a bit more for those
>that perceive this as a mere implementation issue.
>
>XInclude depends on XPointer depends on XPath; more trouble to come with
>XPath 2.0, which relies on W3C XML Schema; i.e., if I understand this
>correctly, simple inclusion of document fragment will require a PSVI for the
>including document. It'll get hard to avoid W3C XML Schema processing even
>when you don't really need it ;(


 From my use of xslt 2.0, we should be able to use 95% of xpath
in 2 as we did in 1.0

they pretend backwards support with a mode,
but it all needs checking carefully.

I do my utmost to avoid xml schema, and (mostly) I've succeeded.
I think xpath is probably the cleanest 1.0 to 2.0 'port'.

HTH DaveP