Martha,
No, I hadn't seen the RDF model, so thanks for pointing that out. It
will take a while to look it over. Each data element is defined and
given a URI. That's good. I'm a bit perplexed by the combination of
relationships in the properties, but I don't know if this is normal RDF
practice. As an example:
Property: Principal Creator--Artist
URI: http://myee.bol.ucla.edu/ycrschema#pcartist
Label: principal creator, artist
Domain: ycrschema:Work
Range: Union of ycrschema:Person or ycrschema:Corporatebody
Subproperty of: ycrschema:principalcreator
It seems that you want a Principal Creator property and an Artist
property, since they are different kinds of relationships. And that
those would be expressed as a relationship, like "creatorOf." Have you
looked at the FRBR in RDF at http://vocab.org/frbr/core ? It only has
the FRBR relationships, of course, so it has less detail in some areas
than a full-fledged set of cataloging data elements.
I also wonder about data types. Few of the entries have a data type (I
saw one with "date"), yet every piece of data must be of a type. I
suppose anything that isn't specified is assumed to be a string or
"CDATA." How are fixed lists handled? They seem to be treated as
separate properties:
Property: Key Identifier for Genre/Form
URI: http://myee.bol.ucla.edu/ycrschema#keyidgenr
Label: key identifier for genre/form
Domain: ycrschema:Genre/form
Cardinality OWL 3.4.2: A genre/form has exactly one unique key identifier
I'm having trouble envisioning how this would work, but I'll think about
it some more.
kc
Martha Yee wrote:
> Have either of you looked at the RDF model at my web site
> (http://myee.bol.ucla.edu) yet? Unless I am misunderstanding you, I believe
> these rules and this RDF model are trying to do what you are asking us to
> do... The question for me is whether our current deprofessionalized
> staffing is capable of implementing such a complex set of rules and such a
> complex model...
>
> Martha
>
> -----Original Message-----
> From: Next generation catalogs for libraries
> [mailto:NGC4LIB_at_listserv.nd.edu]On Behalf Of Riley, Jenn
> Sent: Friday, November 30, 2007 9:14 AM
> To: NGC4LIB_at_listserv.nd.edu
> Subject: Re: [NGC4LIB] Martha Yee's cataloging rules for a more
> FRBR-ized catalog, with an RDF model
>
>
>> Martha Yee wrote:
>>> I have written elsewhere about the fact that our rules and our
>> cataloging
>>> data are already considerably FRBR-ized and that what is lacking for
>> the
>>> creation of true FRBR-ized catalogs is adequate software support.
>> Martha, you and I have discussed this at length, so you know that I
>> disagree that the problem lies with systems. It is true that
>> bibliographic records are very rich and contain a lot of important
>> data.
>> However, as long as bib data continues to be expressed as text strings
>> that require human interpretation, systems will NOT be able to make use
>> of the underlying concepts. This is one of the great errors in the RDA
>> drafts that we have seen: the bibliographic description continues to be
>> textual in nature, with relationships left as implicit in that text. We
>> need rules that can make explicit what today is implicit. And we need a
>> bibliographic record carrier that can carry those explicit expressions.
>
> Surely, now, the problem is to some degree both with the data and the
> systems. There is a great deal more systems could do with our existing data,
> but our current data structures also in some cases make it significantly
> more difficult than it should be (and sometimes impossible) for systems to
> do more advanced things imagined by this group.
>
> I see much of the current debate about why our catalogs don't function
> better as finger-pointing -- "if only *they* (some group other than mine)
> would do it better..." I think to move forward we need to accept that all of
> us have something to contribute, and take responsibility for making that
> contribution. I have every hope that the work that has been done to improve
> systems will demonstrate some of the possibilities, and in turn both inspire
> more innovative system development and expose areas in which our data could
> be better-structured in order to enable more robust discovery and use
> services.
>
> Jenn
>
>
> ========================
> Jenn Riley
> Metadata Librarian
> Digital Library Program
> Indiana University - Bloomington
> Wells Library W501
> (812) 856-5759
> www.dlib.indiana.edu
>
> Inquiring Librarian blog: www.inquiringlibrarian.blogspot.com
>
>
--
-----------------------------------
Karen Coyle / Digital Library Consultant
kcoyle@kcoyle.net http://www.kcoyle.net
ph.: 510-540-7596 skype: kcoylenet
fx.: 510-848-3913
mo.: 510-435-8234
------------------------------------
Received on Fri Nov 30 2007 - 14:56:13 EST