http://community.brighton.ac.uk/jd29/weblog/19542.html

I have been thinking some more about the issue raised in my previous post about the importance of being first in a social system where people influence each other's behaviour. On reflection, this helps to make sense of my thoughts (and those of others) on structure influencing behaviour. In many ways, this is just a symptom of a more fundamental dynamic. It has been noted more than once that the large and slow tend to have a disproportionately large influence over the small and fast in a system of otherwise independent agents, although this is a slight oversimplification. It seems to me that this might be more properly framed as a question of context. Where other issues of scale are equal, those that come first change the context for those that come later. Similarly, the large and slow moving provide a context for the small and fast, whether as defining features of the landscape to which the small and fast adapt, or as constraints on their activities. Big events such as forest fires can change the context rapidly, as can large-scale aggregations of small-scale behaviours, such as locust plagues or riots.  And, of course, intentional human intervention can make a big difference, allowing us to sometimes overcome the large and slow and to change the context by ourselves.

Relating this to one of my favourite topics, why learning management systems are pernicious, it is easy to see how several features of context come into play. In the first place, the LMS was developed within a context. Most arose from large and slow-moving institutional environments which, for reasons that have as much to do with history as with intentional design, have a particular shape and form. They have courses, classes, lectures, lecturers and a host of other structural features that influenced the developers of such systems. Had it been otherwise, they would not have been such a good fit with institutions and would not have been taken up as enthusiastically. Once they become an institutional feature, the power of priority takes hold. When an institution has wholeheartedly embraced a given system it is incredibly hard to get out of it - the effort of transferring and rewriting alone is bad enough, but the changes in culture, ethos, not to mention the training and  marketing needed make institutional LMSs almost unassailable except through the equivalent of a forest fire, changes in the intentional policies of the large and slow moving, or massive small-scale rejection. Small scale perturbations cannot shake them. And so, with their emphasis on the traditional values and structures that first shaped them, they actually enhance and strengthen the status quo. This is depressing to those of us who once saw them as an agent for change and liberation in learning and teaching methods.

What could change this depressing pattern?

• top-down policies, outlawing (say) courses (OK, maybe asking too much too early) or at least courses that follow the same patterns as other courses
• top-down policies mandating flexible, mashable, agile computer systems that would allow the locusts to swarm and take over
• a fire - perhaps started by something like the sites that offer hired tutors on demand, or the many sites that offer assignment writing, or even the widespread avoidance of institutional learning through use of Wikipedia, Google Search, social networking sites and so on.
• large-scale disillusionment - this has both a good and a bad side. The good side is that people may recognise that the problems come from trying to replicate a system that only ever worked because of all the informal and tacit elements that come with it: things like the motivational benefits of a timetable and the presence of others (consider the benefits of watching a movie with others), the ability to talk with fellow students in corridors and coffee shops, the largely hidden conversational aspects of lectures that cause changes in presentation and content according to the needs of the audience, the simple physical existence of an educational institution and much more. The bad side might be that people become disillusioned with technology per se, rather than recognising the flaws in its design. I fear that the vicious circle that starts with students liking these systems because they let them do the institutional dance better, thus reinforcing beliefs in their value when the problem is the institutional dance in the first place, may not help this process.

Now that I teach at Athabasca University I am becoming aware of another related issue with the LMS. When it was formed in the 1970s, AU modelled itself to a large extent on the UK Open University and developed a style of teaching that was aimed at enabling independent study, very largely through correspondance. Unlike the UK OU, AU's teaching was almost entirely at a distance and it developed a model of course design and support that worked well, albeit one that only suited a relatively small group of fairly self-directed learners. This was nonetheless good, because those learners would not have been able to complete a traditional university course, so access to education was increased. Now that courses are becoming increasingly online, AU has turned to a somewhat customised version of Moodle for its learning environment. This is bad in two ways.

1. Moodle, as implemented at AU, encourages forms of interaction through discussion forums (it also has wikis, blogs and so on, but these are not yet institutionally supported to any great extent). This does not fit with the context of the correspondence model, either pedagogically or practically. It means more work for the tutors that engage (who are already quite heavily weighed down with things such as the large marking burden that the correspondence model entails) and, especially where tutors are less active, disillusionment for many students who discover that what should be a wonderful learning resource is actually often quite restricted and unengaging as, no matter how active the tutor, students enrol throughout the year and thus do not form a meaningful cohort. What is incredible is that, sometimes and not often for long, active communities do emerge.
2. Despite providing a small range of templates relating to different learning designs, Moodle is still a product of the same mentality that led to WebCT. It is suited to a different context of teaching than that practiced at AU (at least in most undergraduate courses).  

AU teaching teams are faced with two choices:

1. bow to the Moodle model and redesign courses to match its implied pedagogies. I think this would be silly as AU is free to cast aside much of the frippery of conventional university approaches to teaching and this would just bring the institution in line with others, who would therefore be competing head-on with it in the online space. AU's mission is to open education to all, not to compete for students with other universities and colleges.
2. take the perpendicular path. Build on its strengths in allowing open, unpaced courses, but surround that with a social infrastructure that fits the overall ecosystem better. This is a bit scarier, but seems the only logical path to take. It is scary because it means rethinking the whole process of course design and delivery, not replicating structures that made sense in the days when courses went out through little more than post, phone and TV. We do at least have an institutionally supported instance of Elgg, which is a step in the right direction albeit still lacking some important features (notably in terms of the ability to  mash it up through more than RSS and its rather flat and undifferentiated model of communities)
   

I am very privileged to be a part of two institutions, both great in their own right, but both with related problems. One of the joys of working at the University of Brighton is that it provides a context for the individual tutor to adapt at a very small-scale level, changing the delivery of a course in real time, as well as enabling experimentation and adaptation throughout. However, the constraints set in at a higher level and the institution as a whole, whatever its good intentions, is hard to steer in an agile fashion, moving like a mega-tanker in a small strait. Innovations spread slowly, despite good communication, because the overall structure is deeply embedded in a physical and temporal context. Athabasca University has some of those constaints, but is inherently more agile at an institutional level. Its campus is primarily virtual and thus malleable. Its mission requires it to seek those who would otherwise be denied an education, so it is not constrained by the struggle to compete for school leavers. However, the context created by its history remains a powerful barrier, and the pedagogical approach that its history entails reduces the flexibility of teachers at the smaller scale, at least in undergraduate teaching. What is needed, and what would benefit both insitutions, is the best of both worlds. Flexibility at the top and at the bottom. I think that there are ways to achieve this - my book represents an attempt to address the problem at the level of the learner, but there is a need for a similar set of principles that would enable the valorisation of diversity at the level of the institution. Technology can support this or prevent it. The kinds of things that would give support would be the provision of small, reliable, interoperable, aggregable components, and diverse templates to help people to use them effectively. The kinds of things that would prevent it are...well...the monoliths. The big, engineered hunks of code that embody cultures and patterns that, whether in keeping with or opposed to the institutional ethos, are dangerous and evil. I am encouraged by moves in industry into social spaces such as Facebook, with their increasing diversity of available applications and their agility and scalability, although I am still wary of putting eggs in a single basket, no matter how big the basket - until Facebook becomes really open (and this may happen) OpenSocial is a far better option.