About the model: Question 10
10. Is there any way to define types of data in an RDF model? For example, would it be possible to define a particular data element as being transcribed (copied from the resource), as opposed to composed by the cataloger? I have discovered the “built-in datatypes” which seem to provide a way to define certain data as “date” or “integer,” for example, but I haven’t quite figured out whether these datatypes can be placed right into the model or whether they have to be placed in the resource records created according to the model.
“I don’t think there’s much correlation between a type and a datatype. Datatypes are very closely linked to the exchange format they’re expressed in (so you can say that an RDF field’s value must be integers, dates, strings of certain kinds, and so forth), so I don’t think you need to worry about that. A type is a straight “is_a” relationship in the taxonomical hierarchy, so it’s enough to just add ; some_book has_subject fish whales element_content_source transcribedjust like ; fish element_content_source copied_from_resourceis the default in your model (but you wouldn’t need to express this)”–September 11, 2007 email from Alexander Johannesen.
Addition to question 10: Could “literal value surrogate” be used to mean a particular element is transcribed, and “non-literal value surrogate” be used to mean that a particular element is selected from a list, including lists such as LCSH (for subject headings) and the LC/NACO authority file (for name and work identifiers)?