On Tue, 2011-07-05 at 11:26 -0400, John Cowan wrote:
> The first substantive change is that a subset of XML processing
> instructions are now part of MicroXML. PIs are not in the data model,
> but processors SHOULD pass them to applications anyhow. A MicroXML PI
> MUST look like a start-tag, except for beginning "<?" and ending "?>",
> so PIs similar to xml-style work, but random PIs like "<?foo bar?>" do
> not.
As I understand it html5 is deprecating processing instructions.
I'm also not keen on allowing error recovery without specifying how it
works - HTML taught us that this does not lead to interoperability. THe
approach taken by HTML 5 is probably a good one if you've decided to
break backwards compatibility and have things labeled XML that no
existing (conformant) XML processor can handle.
Liam
--
Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/
Pictures from old books: http://fromoldbooks.org/
