[
Lists Home |
Date Index |
Thread Index
]
At 11:43 AM -0800 1/23/02, Tim Bray wrote:
>><?rddl-doctype href="http://example.org/some-rddl-description.html"; ?>
>
>Blecch. Assuming you believe that this is a good idea, stick it
>in a namespaced tribute required to be on the root element. Then
>you can get at it through all the existing APIs and address it
>with XPath and so on.
>
You can do all these things with a Procesing instruction in the
prolog too. SAX supports processing instructions. So does DOM. So
does XPath.
>PIs are for application-specific processing IMHO. This is not
>application-specific at all.
I think it is application specific. It just depends on how you define
application.
The issue is that if we add an attribute then that attribute must be
declared in valid documents. We can't just go adding this willy-nilly
to XHTML or SVG or all the other documents we might want to stick it
in. We can so add a PI.
This strikes me as very analogous to the existing uses of PIs for
things like robots, cocoon preprocessing, and stylesheets. It is an
extra piece of meta-info we want to attach to documents outside of
the document's normal element structure.
--
+-----------------------+------------------------+-------------------+
| Elliotte Rusty Harold | elharo@metalab.unc.edu | Writer/Programmer |
+-----------------------+------------------------+-------------------+
| The XML Bible, 2nd Edition (Hungry Minds, 2001) |
| http://www.ibiblio.org/xml/books/bible2/ |
| http://www.amazon.com/exec/obidos/ISBN=0764547607/cafeaulaitA/ |
+----------------------------------+---------------------------------+
| Read Cafe au Lait for Java News: http://www.cafeaulait.org/ |
| Read Cafe con Leche for XML News: http://www.ibiblio.org/xml/ |
+----------------------------------+---------------------------------+
|
