On May 30, 2006, at 6:30 AM, Ian Davis wrote:
I'd just like to point out that eRDF can express metadata about
embedded objects and fragments, so this fragment:
<div id="ian"><span class="foaf-name">Ian Davis</span></div>
embeds the following triple:
<#ian> foaf:name "Ian Davis" .
Cool, thanks for pointing this out. How does this let one make a
statement about an embedded image? Or is it that you wrap a DIV
around the image and then make a statement about that DIV?
I'm also curious about what you regard as copy-and-paste metadata
and why eRDF doesn't support it.
Yes, I need to be clearer about this. Benjamin Nowack also asks the
same question [1].
Clearly, you can't copy and paste just any piece of HTML w/ RDFa, as
you and Benjamin point out. But you *can* do what CC needs to do:
prepare a self-contained small chunk of HTML w/ RDFa that *is* copy-
and-paste-able. This is because you can declare schemas right there,
not just in the head of the document. So, in fact, I'm not sure that
most documents will declare namespaces in the head. I foresee a bunch
of "wizards" like the CC one for generating chunks of HTML with self-
contained RDFa metadata.
In other words, the point is not that you can adversarially choose
any subcomponent of the HTML and copy and paste it, of course. The
point is hat, if you're a bit careful, you can create a self-
contained chunk of HTML that is copy-and-pasteable.
That's important for more than just wizards, by the way. It's
important when you want to aggregate a number of content components
into a single page. Or when you write a single blog entry at
something like blogspot but you want to add your own namespaces (and
you have no control over the head). In a number of cases, users only
control a small chunk of HTML on a given page. It's important that
this chunk can be self-contained with all of its metadata.
-Ben
[1] http://lists.w3.org/Archives/Public/public-rdf-in-xhtml-tf/
2006May/0048
