The centralisation dilemma in educational IT

I wrote an article for a new journal, the International Journal of Virtual and Personal Learning Environments (IJVPLE). My piece was entitled ‘The centralisation dilemma in educational IT’. I argued that we have a centralisation – decentralisation cycle in educational IT, so we had distributed versions of VLE, which moved to a central VLE, and we are now seeing a shift back to decentralised cloud services.

The arguments for a centralised VLE are:

  1. Uniformity of student experience
  2. Centralised support
  3. Quality assurance
  4. Efficiency
  5. Robustness
  6. Integration of different tools
  7. Staff development
  8. Platform for expanding elearning offerings

Whereas the arguments for a decentralised model can be summarised as:

  1. Quality: The individual components of an integrated system will not be as good as specialist tools performing any one of these functions
  2. Flexibility
  3. Pedagogic suitability
  4. Relevance
  5. Educator control
  6. Personalisation

I then look at some of the issues around both the centralised and decentralised models before coming to this (maybe woolly) conclusion:

“Having looked at some of the issues surrounding centralised and decentralised educational IT services, we can see that there is dissatisfaction with the current centralised model, but also problems with the implementation of a decentralised model. What this may signify is that we are in a transition point as educational IT services evolve from a tightly controlled and deployed set of systems to a broader continuum of tools…

This next succession of IT services is likely to see an attempt to retain some of the benefits of a centralised system with the diversity of a decentralised one. This will see the easy, open integration of third party applications into an existing hub of central, core services.

However, this approach is still unlikely to satisfy those in the decentralisation camp, because the implementation of educational technology can also be seen as a metaphor for how the institution itself operates. The primary benefit of a centralised system is that it facilitates control – this enables universities to perform their duty of care, both pastoral and educational, to their students by controlling the environment. In order to realise a decentralised model it would necessitate a good deal of relinquishing control. But for those who favour a decentralised model, this control is both restrictive in terms of the technologies they use, but also what they do with them.”

Because it’s the first edition the journal is freely available so you can access a PDF of the full article here.

OU chum Niall Sclater also has an article in the same journal, entitled ‘elearning in the cloud’, which you can also access.

4 Comments

  1. Thanks Martin for raising again these issues which not only continue to dog us, but dog us even more than ever in the light of changing practices and tools.
    We wrote about this recently too (see http://www.heacademy.ac.uk/assets/York/documents/ourwork/learningandtech/Transforming-08.pdf) and noted that “The central challenge is to manage what seem to be competing imperatives:
    the creation of consistency, the maintenance of standards, formal explicit processes
    and procedures on the one hand, with ad hoc, flexible on-the-ground activities
    on the other hand. Taken together, our case study and the wider literature argue
    persuasively that top-down policies, understood as coercive in corporate cultures,
    are least effective for varied responsive pedagogical change. At the same time,
    fragmented, on-the-ground activities cannot be scaled up to larger success without
    systemic support. This leads to the crucial role of the middle layer in universities; to
    what has been termed ‘middle-out’ approaches.”
    Implications are important for middle layers of universities, it seems.

  2. I agree in principle with the notion that individual tools are better; but I’m not always sure better is best. For example – the OU VLE wiki is not as ‘good’ as, say, Mediawiki (in that it doesn’t have the flexibility, sophistication, and so on), but then it doesn’t need the additional functionality of Mediawiki.
    Tutors might not know or want to know how to use it given its relative sophistication; and students might not want to, either (they might not like markup, say; in that case, perhaps something WYSIWYG like Wetpaint will do the trick?)
    There’s no reason, given the modular and open quality of something like Moodle that a wiki couldn’t – theoretically – be as ‘good’ as an external tool. Or more specifically, more ‘fit for purpose’ (hate that term, but it seems to fit). Writing all this seems a bit churlish, though, when I tend to agree with you.

  3. I agree with what your thoughts about centralisation and control. It seems to me that a highly centralised approach for a VLE is more about control than anything else these days. It is often dressed up in arguments about consistency and usability but I remain to be convinced that these are crucial issues for students who are increaslingly used to working in multiple web based UIs. My experience with a highly centralised VLE is that it doesn’t meet the requirements of a large and highly diverse dual sector university and that the controls enforced by centralisation demotivate early adopters who want more control themselves. These are exactly the people that you don’t want to put offside.
    I think that there will be a desire for the next generation of VLE to be essentially centralised with third party applications as you say but I wonder whether they might be superceded by open, cloud based VLEs. Who knows.

  4. Estimado Profesor
    Es muy interesante su artículo. Abre para mí un nuevo frente de discusión al pensar en la incorporación de las ntics.
    No solo en términos macro sino también al interior del control de cada institución donde se realicen las prácticas.
    Saludos
    Jorge Apel
    p.d. Disculpas por el castellano, pero mi inglés no da para la escritura

Leave a Reply

css.php