Thanks much for the input. You are right about the schemas. There are
several issues with them and we are waiting (anxiously) for them to be
resolved. For now, my position is that we report what is in the schemas
(the version that the NS resolve to), rather than make up yet another
I did not address the first several items in your post because I believe
they are schema-related. I have tried to address the other issues below.
1. The drop-down list on the "display" box on the search page offers
"Qualifiers" as an option: I think what are being returned here are
"Element Refinements". According to
"Qualifiers" covers both "Element refinements" and "Encoding schemes".
So I think the term in your drop-down list should be "Element
refinements" not "Qualifiers".
Let me look at this. Should be a simple correction.
2. If I select, the option "Namespaces" from that dropdown list (without
making any other entries on the form), I only see 4 items found:
Summary of Registered Namespaces Items Found:4
But there are clearly two other namespaces indexed (dcmitype, dcterms)
and I (correctly) see 6 "Schemas", so looks like a glitch somewhere in
No, this is correct. The schemas that have been imported only include
references to the 4 namespaces listed. It is not enough to simply list the
NS in the schema. It has to be referenced (this may be particular to the
parser we are using?). Please let me know if this explaination is not
clear. I had to look at this a while myself.
3. References to documents like http://dublincore.org/documents/dces/
are rendered as hyperlinks like:
i.e. queries to the registry which return no results.
I guess the desired rendition would be a straight hyperlink to
http://dublincore.org/documents/dces/ but I can see that this might be
hard to do.
You are right, and also right about it being difficult to do. This is one
of the problems of using rdfs for our input data, and will need to be
addressed in a subsequent phase.
Thanks again for the input!