I’ve described the modelling and rationale in some more detail in a separate document [pdf].

I would warmly welcome any comments, suggestions and pointers to where I’ve gone wrong.

The background to the development of ORE is very much rooted in scholarly communication and specifically in open access repositories of research outputs, but hopefully the assumptions behind the abstract model isn’t so tied to this that the specification cannot be applied to education more generally. One idea that I know has come up is that ORE could be used to package together and present those resources a student/institution wanted to include in a particular view of an ePortfolio. ORE also seems to occupy the same space as IMS Content Packaging, but while ORE makes it simple to specify that you want to aggregate several independent resources, the means it offers for specifying an order or sequence for the resources that is not inherent in the resources themselves is not straightforward. So if you’re describing a book where chapter 3 follows chapter 2, which follows chapter 1, then you’re fine. But if you want to say that for this course students should read chapter 1, then use some other resource then read chapter three, then you’re no longer talking about properties inherent in Chapters 1 and 3 so you need to create a proxy which represents them in this aggregation. Which seems like a bit of a faff. Indeed this was described as an “edge case” during the meeting (though I’m not sure it is, reuse and exchange is going change relationships), so perhaps ORE should be seen as restricted to doing a similar job to the Resources section in a Content Package manifest and something else used for the Organizations. With that in mind it should be relatively simple to transform the ATOM serialization of an ORE resource map to and from an IMS Content Package manifest. Let me know when it’s ready ;-).