Been pondering this for a while now:
In the EU4ALL project we are developing a framework at a service definition level and at a technical infrastructure level that supports that. We are in the process of selecting what are the services we will implement at the two major pilots – one at the OU and the other at UNED in Spain. A key criteria in selecting these services has been what would best support the validation/evaluation work at these pilots. This has surfaced the question – how do we validate/evaluate a framework as distinct from a given instantiation of it.
One partner has suggested implementing a broad range of possible services is key. However I think this is not sufficient because the key point of a framework is that it is valid in a wide range of circumstances. Hence my view is becoming that alongside the detailed review of a wide range of services implemented in the two major pilots we need to construct a methodology that exposes what the issues would be if those services were implemented in a diversity of contexts. It is likely that this will have to be done without implementation in those contexts. However the view of the implementations at the 2 major test sites by key stakeholders at other contexts would be of great value.
Any comments or pointers to literature welcome!