One of my first jobs was to attend a meeting of the KIP project. In short: Knowledge integration between various VLE's by building a respository in the library of (digital) learning objects. Naturally these learning objects will be catalogued by adding lots and lots of metadata.

Of course this has to the extremely sensible way to go. All learning objects must be as well catalogued and accessible as possible.

However: Daily life is different (it certainly was at Geosciences and does not appear any different at Social Sciences): A teacher builds his or her (more ladies at Social Sciences) course around lectures and activities. Any digital component is added to Blackboard (or WebCT) to help improve this learning process. Therefore: why bother to document everything you are adding to your VLE because it is not your primary worry?

I do think it is valuable to catalogue all learning objects in a context in which these digital learning objects are central to the learning process. This is for example the case if you start designing distance education or if you decide to take parts of the learning process and bring these in a digital medium (for example CBT courses).

For all those interested in metadata and how far you can go: please do visit the imsproject site and do not start thinking up your own sets on your own.