There was broad agreement that a VLE should be an open framework of tools that could be added or removed as required by the learning design. From this consensus, the discussion centered around two poles:
- what is the core component of the VLE
- what is the right scale of aggregation/disaggregation.
There seemed to be agreement that there were two core components:
- a user environment
- an administration framework.
The administration framework would probably be implemented on a wider scale than the user environment but it need not be, and there is no reason why a user environment shouldn't "talk" to more than one administration framework.
Although he didn't say it explicitly, John Fontaine's position was that Blackboard would be able to provide most of the user-environment tools that plugged into the framework and the framework itself, while all the other discutants agreed that there should be a framework and an environment, but they needn't be vendor-specific.
The core component required of the admin framework is a very fine grained, federated access control system: particular people, in particular circumstances, can use particular tools to do particular things (discuss, collaborate, access content; do, submit, mark assignments; etc).
The right scale of aggregation for the user environment is below that of a whole institution, but above that of a single person or even course module. Recognising that all loci of implementation are different the right scale seems to be at the level of an academic unit: school, department. Environments have both instrumental instructional functions and affective functions in forming and sustaining cohort and group identities. The affective functions are best achieved by common look and feel, visual identity, shared task environment and so on. However it is also recognised that Business Schools will want theirs in a corporate livery while Schools of Education will want theirs in a well-worn tweed. Humour aside, different disciplines have different learning, teaching and assessment traditions. Departments of Psychology may require stats packages, while cultural theorists may need discourse-analysis tools.
It is generally asumed that the right scale of aggregation for the VLE admin framework would be the institution. But this was not really explored. I assert that there is no reason to assume this. A shared services agreement could see the implementation of an access federation and learning management system regionally serving several institutions. Or, conversely, admin implementation could be below the level of the whole institution. The question is as much - or more - to do with cultures of governance and patterns of employment than administrative necessity.