I've got a newbie question, apologies if it's obvious.

Is anybody in this situation: Several eprints instances within an institutiton (e.g. at departmental level), with an overall institutional level repository (populated entirely through OAI Harvester or otherwise)?

Is this essentially supported out of the box by installing an institution-level eprints instance, and letting it harvest the departmental level instances? 

From a users perspective, is the functionality is the same (i.e. equally smooth, same look and feel, etc)? I.e. is there a difference between "native" records (say departmental level) and "harvested" records (institutional level), from the users perspective? By "user" I mean the person browsing the repository ("anonymously") on the web - obviously there's a difference from the contributors perspective.

Does it make sense for the institutional repository to harvest departmental repositories? Or would it be preferable to have a single institutional repository and generate listings for departments from it (which should be viewable within the sub-domain owned by the department)? If the latter is preferable, what would be the simplest way of doing that? (Set up departmental repository, and let it harvest the institutional one? Or are there other e.g. API-based options?)

Hope this makes sense!

Thanks!

Bjoern