Wednesday, September 12, 2012

Foundational SOOCs (Specialized Open Online Courses)

I've been wondering when The Thing will happen in higher education for at least two decades, and that's what my ChaosBook.org is about. Now that Sebastian Thrun has happened, we are in the lemmings mode, every University trying to the same thing. But when the dust settles, there will be only one 'iPhone' left standing, the rest will be crushed in competition.

Bill & Melinda Gates Foundation has come out with solicitation for Massive Open Online Courses (MOOCs) grant proposals. Theirs is a laudable goal:
...Must focus on a high-enrollment, low-success introductory level course that is a barrier to success for many students, particularly low-income/first-generation students...
If my institution decides to enter the fry on the level of introductory undergraduate courses, it should pick only very few - like one 2-semester sequence, taught by lecturers universally known to be exceptional, and these few should be given full technical and organizational support, because either you get it perfect the first time around, or do not attempt it at all. This is no arena for amateurs. And very few courses a suited to MOOCsisation - it is no accident that the first success was a rather straightforward computer course.

For research universities most research training will continue as is, by one-on-one apprenticeships. Though I think there is a market for SOOCs (Specialized Open Online Courses). There have been no resources to develop ChaosBook.org - since I moved to US everything has been done by myself, alone, in free time, so it is woefully technologically primitive - but do I keep meeting graduate students who use it.

I think a nonlemming strategy would be a boutique education approach: create for Coursera.org several high level SOOCs in areas where the institution has particular strengths. They will never be money makers, enrollment will be in hundreds at the best of times, but their advertising value would be high: they would enable us to recruit into our research programs top notch students whose quality we already know through our courses that they have taken.

[for the next installment, click here]

2 comments:

Mason said...

So I guess I still have the old bound copy of the ChaosBook in existence?

Jonathan said...

A ChaosBook class on Coursera would be fantastic. I'm doing their Probabilistic Graphical Models course right now and seems pretty good despite being on a fairly high level (maybe 4000 level). I would expect that assignments would have to focus on computation so that grading can be automated. But I think people could still get a lot out of such a course.