Ryan,
Many thanks. It's a relief to know I'm not crazy or stupid (or any more
than I usually am anyway).
On the other hand, the bug (if it is a bug) has to do with exactly the
observation you made on my misuse of dc:creator as an rdf type. When I
replaced the offending code
<dc:creator
rdf:about="
http://content.lib.washington.edu/contentcollections/creator#Gehry,_Frank_O.,_1929-">
<rdfs:label>Gehry, Frank O., 1929-</rdfs:label>
<owl:sameAs
rdf:resource="
http://errol.oclc.org/laf/n82-13991.html"/>
</dc:creator>
with
<uwdl:creator
rdf:about="
http://content.lib.washington.edu/contentcollections/creator#Gehry,_Frank_O.,_1929-">
<rdfs:label>Gehry, Frank O., 1929-</rdfs:label>
</uwdl:creator>
The problem simply evaporated. (I left the owl:sameAs statement only in
the authorities.n3 file. Jena drew the correct inference, and longwell
correctly displayed it as two equivalent resources.) It would seem that
the longwell 'referring resources' code expects dc:creator to be used as a
property. Not a bug, a *feature*, yes?
On Mon, 13 Jun 2005, Ryan Lee wrote:
> Hi Jon,
>
> Sometimes wiser heads make mistakes; this looks like a bug. Doing some
> tinkering of my own reveals that commenting out the RDF (only in the code
> section you listed first) about Gehry will get your referring resources back.
> The Longwell 1.1 code dealing with finding referring resources seems to be
> slightly wrong.
>
> Sorry for the puzzler, there's nothing really wrong with your data in this
> regard.
>
> I did note that dc:creator is used as the object of an rdf:type, which it
> shouldn't be because it's a property; you should find a suitable rdf:type for
> your Gehry resource in the RDF.
__________
J.J. Crump
Dept. of History 353560
University of Washington
Seattle, WA. 98195
Received on Mon Jun 13 2005 - 22:03:07 EDT