David Wiley talks about something called the "Reusability Paradox" of learning objects. It's one of the fundamental issues in dealing with learning objects, and basically boils down to this (grossly oversimplified Coles Notes version of Wiley's paper):

"If a learning object is useful in a particular context, by definition it is not reusable in a different context. If a learning object is reusable in many contexts, it isn't particularly useful in any."

Not rocket surgery, but it's a tough problem to solve. The exciting thing (for me, anyway) about some of the projects we're getting involved with at the Learning Commons, is that we're finally starting to attempt to address this fundamental paradox. How can you have learning objects that are usable and reusable? We're looking at ways to assemble content from lesser bits, building context along the way. We're looking at dynamic assemblies of learning objects with varying degrees of contextualization, where the assembly itself provides the context to make it usable in a particular setting, while letting the lesser components of the thing remain reusable in other contexts...

Stay tuned, folks. This is going to be a fun ride!