Stephen Downes makes a valid point about my recent question about whether or not academics should manually create websites. I agree with his underlying point that academics should not be forced to use the institutional approach. Given any option I would not suggest such an approach.

Incompetent paternalism

However, at least within some Australian institutions academics are being forced to accept an institutional approach. That approach is typically expressed as “minimum service standards” which are specified by management. The academics are than expected to manually fulfill those standards. I have a problem with this approach, but if it is being adopted, then at the very least implement it in an efficient and effective way.

What is happening in these situations could be described as “incompetent” paternalism. Academics are being treated as children (a theory X perspective of academics underpins this approach). Management as the parents have to specify codes of behaviour. But when it comes to implementing this code of behaviour, management are actually making using an inefficient and ineffective approach.

I disagree strongly with this approach, but if management believe it, is it too much to ask them to do it efficiently and effectively? That’s one perspective, my real interest is in a third way that tries to effectively merge features of both incompetent paternalism and the academic free for all.

The “libertarian” paternalist alternative

The model that evolved in the early part of this decade could be described as a “libertarian” paternalist approach. It’s a bit of a stretch but I think the metaphor works.

The theory was/is that an appropriately skilled group, taking an adopter-focused and emergent development approach could develop a default course site that could effectively be used by a group of courses. That default course site could be automatically created. But since the group was using an emergent development approach, the default course site would continue evolving.

The default course site did not remove the academics freedom of choice. As implemented, academics could modify the default course sites in two ways:

  • Use the “LMS” to modify or add to the default course site; or
    Here’s an example default course site from 2006.
  • Create a real course site.
    Here’s a default course site where the academic has created a real course site.

This wasn’t a perfect solution, it still wasn’t flexible enough. We had plans to enable better merging of the default and real course sites. i.e. if a real course site was created, it could replace the default course site and make better use of the services offered by the “LMS”. But they never got off the drawing board.

In the almost 8 years that this approach was used, the “LMS” averaged around 300 course sites a year. The number of real course sites (i.e. academics doing their own thing) was never reach 10% in any given year. It averaged around 4% of course sites a year.

Lack of appropriation

Importantly, where possible, the aim was to observe what everyone was doing, especially the <10% creating the real courses sites, and use those insights to modify the default course sites.

The current management approach of specifying minimum standards is being driven by external desires, not by the experience of academics and students using the minimum standards.


Jones, D. and T. Lynch (1999). A Model for the Design of Web-based Systems that supports Adoption, Appropriation and Evolution. First ICSE Workshop on Web Engineering, Los Angeles.