Robert Sayre wrote:
Danny Ayers wrote:
I think Robert's (unfinished I think) PacePropertyDesign is promising,
I don't think so. I pretty much hit the wall with that one.
What you have there will work wrt RDF. The entry as data dictionary is a good fit:
<atom:entry rdf:about="http://www.dehora.net/journal/2004/05/a?id">
<atom:feed rdf:resource="http://www.dehora.net/journal/" />
<atom:title>Thus sprach metadata</atom:title>
<atom:link>
<rdf:Description
rdf:about="http://www.dehora.net/journal/2004/05/a.html"
atom:rel="alternate"
atom:type="text/html"
atom:href="http://www.dehora.net/journal/2004/05/a.html" />
</atom:link>
<atom:id
rdf:resource="http://www.dehora.net/journal/2004/05/a?id" />
<atom:created>2004-05-23T01:45:36Z</atom:created>
<atom:summary>Seairth Jacobs</atom:summary>
<atom:author>
<rdf:Description rdf:about="mailto:[EMAIL PROTECTED]">
<atom:name>dehora</atom:name>
<atom:url rdf:resource="http://www.dehora.net/journal" />
<atom:email>[EMAIL PROTECTED]</atom:email>
</rdf:Description>
</atom:author>
<dc:subject xmlns="http://purl.org/atom/ns#">SemanticWeb</dc:subject>
... more stuff here ...
</atom:entry>
IOW it's mostly transform happy.
Things like Person constructs are trickier - those wind up in their own rdf:Description blocks as above, making the markup ugly. Otoh, if you make these XML literals you hide a lot of good data from the RDF - it's a bit like using CDATA in XML.
Last time I looke at this, I hadn't figured out a clean way to express atom:content
cheers Bill
[LOL. I'm currently shipping a system that has a model for 'events' that does exactly what PacePropertyDesign does with the value of the event - literals, xml or resources. It works well enough, even though the types are disjoint.]
