This paper discusses the rationale for the representation of user feedback in a structured and reusable format so that it can be reused by different recommender systems. We emphasize how information about the context can be included in such a representation. This work-in-progress takes place in the context of two large European initiatives that set up collections of digital educational resources in distributed repositories to serve the needs of different user communities, and to collect user feedback such as ratings, bookmarks and tags related to the resources. The overall aim is to facilitate the exchange and reuse of their data sets in order to support recommendation of appropriate resources to the end users.