Learning objects don't need to be aggregated in one place, or be 'locked in' to one virtual learning environment. And how do we know when a learning object is being reused, which after all is meant to be the raison d'être for their existence? Although it's been available from last October if you want a really good audio-visual overview of a distributed learning object repository implementation then spend 22 minutes listening to what Alan Levine (Maricopa Community Colleges), Brian Lamb (University of British Columbia), and D'Arcy Norman (University of Calgary) have to say. The content is interesting by itself but I also admire how polished this audio-visual presentation is, e.g. smooth handovers between different speakers providing a sense of pace and changes of activity. There is also the opportunity to branch from the linear nature of presentations such as this via the context sensitive links provided on most slides (just remember to press the pause button first).
It might be worth playing the slides talking about 'trackback' a couple of times because this is a concept now implemented in many weblogs which has incredible potential for providing information about how a learning object (or package, or resource) is being used, sometimes in ways or in contexts that the original author never envisaged.
Then follow-up with a visit to the wiki site TheFuss