- From: Karen Coyle <kcoyle@kcoyle.net>
- Date: Fri, 21 Jan 2011 07:57:25 -0800
- To: Thomas Baker <tbaker@tbaker.de>
- Cc: "gordon@gordondunsire.com" <gordon@gordondunsire.com>, public-xg-lld@w3.org
Quoting Thomas Baker <tbaker@tbaker.de>: > > I'm not sure I follow you... Are you saying that all of the > information in a library record can always be expressed with > triples with the same subject? Even, for example, if FRBR > distinctions are introduced? First, the FRBR entities of Group 1 are modeled as separate records (unfortunately). That's something I see as problematic, but that's how it is. It is my impression that in each such record, all of the triples will have the same subject. Maybe we need to try out some examples and see if this is true. If you look at FRAD, you see that the intention there (if I read it correctly) is that a separate name authority record carries the information about the named entity. The bibliographic record (one of the FRBR Group 1 entities) only contains the identifier for the named entity as an object, something like: <http://mydata/bibrec1><has author><http://mydata/author7> Now, in today's world, the identifier is a string like "Baker, Tom", but that is considered the identifier for the name. In my mind, the DCAM represents a full data model, not a record. The library world also has a data model, with 3 entity types, the three FRBR groups (and all groups are actually multiple entities). But each entity is a separate record in the instance data. Let me make it clear that I am NOT saying that this is the right way to do it. I'm trying to explain current thinking, as I read it, in library cataloging. > > Jeff introduced the unimarc notion of "primary entity" which, > if I correctly infer, nicely captures the notion that a record > is focused on a particular resource without implying that all > of the information in that record may be expressed directly > as an attribute of the primary entity. I don't know how primary entity is defined in terms of the record structure, so I can't comment on this. The idea makes sense to me as defined, but it would take more analysis to figure out exactly what the records actually express. It sounds much like what I am trying to explain, but I'd need examples. Note that library records often contain administrative data about the record or the creation of the record, and this isn't distinguished from data about the primary entity. Other than that I do believe that each record has a single focus today. What freaks librarians out is the idea that this structure will come apart in the linked data world. Diane Hillmann has some slides she uses to show that we can have more open data without losing the coherency that is so comforting to us today. kc > > Tom > > -- > Tom Baker <tbaker@tbaker.de> > -- Karen Coyle kcoyle@kcoyle.net http://kcoyle.net ph: 1-510-540-7596 m: 1-510-435-8234 skype: kcoylenet
Received on Friday, 21 January 2011 15:58:05 UTC