Since the likelihood is that venture-capital funded MOOCs are going to fade out, and (in one way or another) the format is going to come more closely under the control of universities, it’s worth thinking more about where exactly MOOCs can be of greatest use within higher education systems.

The basic challenge is that MOOCs are individual courses, but what matters for most students is a degree.  The only way MOOCs genuinely make sense as part of a higher education system (as opposed to being a stand-alone edutainment product) is if they can be arranged in such a way as to be part of a larger curriculum.  The key issue, then, is how to integrate MOOCs into such a framework.

I think at the moment there are two settings in which this is both possible and desirable.  The first, which was recently profiled in the Chronicle, is exemplified by a project called Kepler in Rwanda.  The project advises students on how to put together a set of MOOCs in such a way that they can pass challenge exams for competency-based degrees, such as those offered by places like Southern New Hampshire State.  The MOOCs can be from anywhere; the value-added role of the institutions is to create a curriculum out of the many hundreds of MOOCs on offer, and to certify that learning has occurred.  This isn’t a model that would be especially appealing to North American students, but it could work in much of Africa and Asia where the local alternatives are weak.

The second setting is North American Liberal Arts colleges.  Yes, really.  Liberal Arts’ schools’ Achilles heel is scale; when you’ve only got 100-200 faculty, it’s tough to offer anything like a full range of courses.  Now, imagine you really want to offer a course on Renaissance Art, but you simply can’t afford permanent faculty with that degree of specialization.  Well, why not allow a MOOC for credit (assuming there’s already a decent one out there)?  That is, use the MOOC as a base, but have your own instructors (possibly sessionals) run discussion sessions, and also set and grade a test at the end?  In this scenario, the institution’s role would be to guide discussions and, as in Kepler, to certify the learning – but the main content provision would be unbundled and outsourced.  It’s not ideal, perhaps, but this kind of approach would allow small, cash-strapped institutions to maintain or expand their breadth at reasonable cost – and it could preserve the idea that intense discussions are the core of Liberal Arts.

(Why not do this at larger schools, too?  No reason, but by definition if they are larger they probably have more staff to start with, and hence fewer gaps in the curriculum).

The key in both cases, as you can see, is that MOOCs themselves need to be unbundled.  They have (the better ones do anyway) plenty of good qualities as learning tools,  but outside math-based disciplines, where right and wrong answers can be identified by machines, no one thinks MOOC assessment mechanisms are worth a damn.  Putting that part of the learning process back in institutions’ hands is key to making MOOCs fit for purpose.