I just limped across the finish line of my first MOOC, Galaxies and Cosmology (10 out of 15 on the last quiz = pass!). I got what I wanted out of the course; I learned the time scales for events in the history of the universe, the types of galaxies and how they evolve, and most importantly, a lot of jargon that will help me better follow astronomy seminars. I also learned more about how I'd develop my own MOOC -- that displaying slides with narration was mostly OK, but that I'd still want some animation for complex math, and that there would be no need to show my face while taking! But I'm also left with a new reservation about the recent explosion of MOOC interest. If we are looking at MOOCs as an option to classroom instruction, was this a fair test of the system?
With a couple of exceptions (coding and poetry courses about which others have posted), most of the online courses I have seen advertised, and certainly the one I took, were what we would call survey courses, or general overviews of a field. And I think this is part of what is fueling the MOOC excitement -- it's fun to have the horizon suddenly pushed way back, and have whole new expanses of knowledge available to you. It's the same thing we saw with the advent of search engines; it's incredibly addictive to watch a movie and say, "What else has that person been in?", or watch Jeopardy! and ask, "Where the heck is Suriname, anyway?", and have the answer immediately available. But while survey courses are great introductions to a field, they hide much of the work required to become a practicing member of the field.
Through the Galaxies and Cosmology course, I know what the Friedmann equation is, and what the solutions look like in certain limits. But I haven't had to solve it myself, and in no way feel prepared to use it for any real astronomical calculations. I learned some new terminology and have shown (increasingly less) retention of it, but I'm not ready to apply it beyond reading comprehension. Which is fine, because that's what the course was intended to do. But add in the requirement of learning applications, and all of the warts of the MOOC process (limited interactivity with instructors, poor signal-to-noise in message boards, grading process versus result) become magnified.
I think Gabriel Ferrer said something very important in the previous blog post. The current incarnation of MOOCs cannot replace classroom instruction, but they can free it up, and allow it to concentrate on its strengths. Present your material through online instruction, where students can choose their own hours and pace of learning. Then devote classroom time to application, problem solving, and interactivity, and you get the best of both worlds.