Quoth Charles McCathieNevile at 06/05/2006 02:43 PM...
> Why are you using another namespace? To include the XML directly? You
> could use meta elements, and note the dc scheme.
What I'm trying to achieve is to be able to provide metadata about an
object in a document, rather than the whole document. I don't think
that meta elements are permitted within the <body> element.
So, yes, I wanted to include other XML within the XHTML.
> Otherwise I suggest using <link rel="meta"... (although I believe in
> XHTML 1.0 you still *should* use a profile attribute to explain what
> meta means, since it is undefined in the spec... on the other hand it is
> pretty common.
This is the approach that I use to include pointers to EARL metadata
about a document and FOAF metadata about the document's author.
> It would be nice to use <a rel="meta" ... but the only way to get
> browsers to note that is with an extension that we wrote (works in
> Opera, FF, and maybe even IE with the userJS extension stuff)...
...and this is what I'm planning to do to achieve the initial goal, due
to the problems with the namespace import (which I was sure was standard
XML and therefore inherited by XHTML). I'm not terribly fussed about
"human interface" user agents not understanding the significance - I
would have a title attribute in the link to take care of this. The
whole reason to have this metadata here is so that it can be undersood
by software agents. A human can (or should) be able to read a page, but
software, likely as not, won't be able to make sense of what is said.
The metadata is there so that the software agent can also benefit.
Does this make sense?
Cheers
M
--
Matthew Smith
IT Consultancy & Online Application Development
http://www.kbc.net.au
|