Barker, FRBRizing Learning Materials
Phil Barker, who works for the Joint Information Systems Committee’s Centre for Educational Technology and Interoperability Standards (or, more briefly, JISC CETIS), posted FRBRizing Learning Materials on his blog last week.
I may have bitten off more than I can chew. I wanted an example for showing how the Functional Requirements for Bibliographic Records (FRBR) might be applied to a typical learning resource. I’m not entirely sure that there is such a thing as a typical learning resource, but the OpenYale online lectures seemed seemed like reasonable candidates. I chose one on Newton’s Laws of Motion as my example because it’s a subject I like.
Check out his diagram! Whew. Applying FRBR to a Complex Learning Resource: A Lecture, Transcript, Recordings and Supporting Materials (53 KB PDF) explains it all in more detail. Complex is right.
I’m not sure where you see the complexity. To put this into a relational database, you’d need eleven simple tables (or relations, if you like): four tables for work, expression, manifestation and item, four tables for HasPart, and three for realized, embodied and exemplified. It’s true that the diagram has a lot of lines, but it contains a lot of information.
I think Phil Barker has done a great job in putting this together. It raises some interesting questions about FRBR, such as whether aggregate expressions always have manifestations. It’s not clear to me that M makes sense as a real entity.
The diagram is also a great starting point for discussions of the user interface. For example, if a user is looking at E1.2, you can make an argument that E1.1 and E1.3 should be displayed in a different way (e.g., in a different part of the page) to E2 and E3.x. I think it would be a real challenge to display the several relevant parts of the diagram in a sensible way.
Comment by Graeme Williams — 31 October 2008 @ 1:56 pmIt is an interesting exercise, but it worries me slightly. FRBR was developed on the basis of ‘bibliographic’ records. In this context the ideas of manifestations, expressions and items have reasonably specific (if disputed) meaning.
I’m not saying that you can’t apply it to Learning Objects or other things of course, or that it isn’t an interesting exercise, but surely there is a question about what an entity/relationship model for Learning Objects looks like, rather than what does a Learning Object look like when you model it in FRBR?
Comment by Owen Stephens — 7 November 2008 @ 7:31 am