e-learning standards

The universe consists of objects having various qualities and standing in various relations.
-Alfred North Whitehead and Bertrand Russell, Principia Mathematica (1911-13).

The world is everything that is the case.
What is the case, the fact, is the existence of atomic facts.
The logical picture of the facts is the thought.
The thought is the significant proposition.
-Ludwig Wittgenstein, Tractatus Logico-Philosophicus (1921)

Derek Zoolander: Why do you hate models, Matilda?
Matilda Jeffries: Honestly?
Hansel: Yes.
Matilda: I think they’re vain, stupid, and incredibly self-centered.
Hansel: I totally agree with you. But how do you feel about male models?
-Ben Stiller, Christine Taylor, Owen Wilson, Zoolander (2001)

Models are hot right now in metadata circles.

Not the Miranda Kerr, Derek Zoolander, $10K-to-get-out-of-bed model, I mean the (much more exciting) conceptual, data-structuring type. Why? Here are two reasons:

1) Machine-readability

To create a linked web of data we need unambiguous metadata. We have to be clear about the things we’re describing and what we’re saying about them. Because metadata must now, above all, be machine-readable, and computers don’t handle ambiguity as well as we do. They don’t have our ability to make best guesses based on context and past experience. As I read recently in a Boston Globe article, a four-year old child can quickly fill in the blank in the sentence “The giraffe had a very long __” but a computer will struggle. Models help us put things in terms even a computer can understand.

2) Interoperability

Mikael Nilsson and his colleagues point out in their paper, Towards an interoperability framework for metadata standards, that unless the models of two metadata formats are compatible, it’s not possible to re-use components from one in the other. So to achieve interoperability between Dublin Core metadata and LOM, the DCMI/IEEE LTSC Taskforce decided that LOM instances would have to be expressed in terms of the DC Abstract Model (DCAM). (Incidentally, today Tom Baker has published an interesting paper about a potential review of DCAM:  DCAM: past, present, future.)

Metadata modelling is not new. Aristotle presented not one but two models of the world in his Categories. Unfortunately, as often with Aristotle, it’s not easy to work out exactly what he is saying, but his categories of ‘things that are said’ are: (1) substance; (2) quantity; (3) quality; (4) relatives; (5) somewhere; (6) sometime; (7) being in a position; (8) having; (9) acting; and (10) being acted upon.

Take the first category (substance) as equivalent to ‘resource’ or ‘entity’ and the others as attributes of a resource and you can reduce Aristotle’s model to the familiar entity-attribute-value (or in RDF terminology, subject-predicate-object) triple that is the core of today’s metadata models, eg:

• Australia (subject) has-capital (predicate) Canberra (object)
• The novel Moby Dick (subject) has-author (predicate) Hermann Melville (object)

These modern models are much like the logical atomism of early 20th century philosophers, Russell, Whitehead and Wittgenstein.

Once the foundation model is established, we can build domain models on top to model particular parts of the universe, eg educational resources in the case of DC-Education, bibliographic resources in the case of FRBR. The Dublin Core Singapore Framework describes how this all works.

So not all models are vain, stupid and self-centred. And there is a lot more to life than being really, really, ridiculously good-looking.