My colleague Chris Edley is out in front of the push to get Berkeley in the on-line education business . Kevin Carey and Matt Yglesias discuss whether the new product should start upscale, like the Tesla sports car as a pilot product , or downmarket, like a knockoff LV bag that holds just as much as a real one, or maybe a Lite Vuitton that cuts some corners in the stitching and doesn’t have real leather trim .
Naturally this has all occasioned a fair amount of debate at Cal and the UC President’s office, debate that strikes me as missing a lot of points, many because of insufficiently respecting the great intellectual principle of “compared to what?”
Most people think an on-line course will be like the courses we teach now, except with the lectures on videotape and discussion sections either finessed or somehow recast into something like GoToMeeting . I think this approach is a big mistake for several reasons. First, and fundamentally, why are we still giving live lectures with everyone in a room together at all, never mind trying to fossilize them into mpegs? This used to be the only way to efficiently use instructor time, but as the students almost never interact with each other in a class larger than a hundred (often in much smaller courses as well), why should they have to look at the prof as a tiny fraction of a steradian, and a blackboard limited to chalk or marker handwriting, now that we have computers, much less do it when some of them could be playing soccer and others could be doing fieldwork in the rain forest? No-one outside of our business uses this method of enlightenment any more. Hold your thumb up at arms length, in front of your computer screen: that’s bigger than my head in a lecture class; what’s the rest of your visual field doing? Why should notes and equations be handwritten, forty feet away, instead of right in front of you on your laptop in nice fonts and thoughtful layout?
Didactic transmission of declarative knowledge is the function of a textbook, not a live lecture, and indeed, the textbook for one of my courses has a cd of software inside the back cover, and many textbooks include access to all sorts of online material. Now if we can start to develop proper 21st century textbooks in the form of multimedia interactive DVDs or websites, with text, exercises and labs, pictures, videos, and all that good stuff, with lots of branching and options, we would be going in the right direction. Customizing learning sites like this is the place for individual professors to be creating value. Don’t put lectures on video; forget them (OK, maybe a prof could pop up now and then in the lower left-hand corner of the screen and say a few words of guidance…but why not a star from the drama department?). All the research I’ve seen indicates lectures are really lame devices for retention or any real learning. If the ego boost for the prof of having a roomful of students listening to him for ninety minutes twice a week is important, let’s find a cheaper way to deliver it, maybe with medals or parking spaces.
I think we’re naive to think on-line education should be benchmarked only against what we deliver now, or to think that’s the way the world will see it. In the first place, an enormous fraction of our current education is already on-line; I communicate with students by email much more now than by phone or in office hrs, and almost every reading for all my courses is a link in a syllabus distributed as a Word file. More important, the Discovery Channel, Simon Schama’s art history programs, yes, and Dogs 101, are on-line education, with very high production values that match what we’ve come to expect on a screen. The only things they don’t have are interactivity and a monopoly charter to award degrees. Maybe we can stay in business behind our market power, but as education blurs out across a working lifetime (as it should) rather than being delivered in a big dollop before most people really know why they’re having it, the credential will probably lose value, or be diffused across a variety of certificates and other evidence of learning. Meanwhile, ascertaining evidence of traits like being smart or glib or a good memorizer has already been pried away from my outfit by testing companies.
We’re also contemplating starting up the on-line learning curve in the worst place possible, namely introductory large-enrollment courses where failure is most damaging to students and where top-level, effective pedagogy is most important (not to mention wisdom, experience, and the confidence that allows a prof to listen to the students when they’re hitting the wall instead of desperately saying everything again faster). A new product like this needs to be developed in skunk works, at a small scale, with lots of ready-fire-aim experimentation at low stakes, not in public and with the most difficult item to deliver in our product line. Graduate seminars and specialized upper-division courses with students who are already committed to the material and have learned how to learn is where we should be learning this new skill, and before we rush off offering online degrees, we need to get really good at the large on-line fraction of our current residential programs.
Down the line, there is real cause for concern about losing the residential on-site experience of college, most of which is a network externality among the students. Edley correctly notes that we won’t be delivering that to any fewer students if we are also offering online degrees to lots of others, but society needs to think about whether a student far from Cal is better off at a local college, where he can have a beer with people who are taking the same courses he is, than sitting in his basement in pajamas getting a Berkeley degree. I don’t know what I think about that because I haven’t heard enough serious discussion of it either way. In any case, let’s modestly get good at delivering online advanced courses to seniors who are off-campus in the jungle or in Florence for a semester before we try to deliver a whole degree through a monitor.
Finally, I definitely come down on Carey’s side of the quality discussion. There is no reason to believe we face a cost/quality trade doing education in a way that really embraces the technology at our disposal. We should assume we are nowhere near the production possibility frontier on those dimensions, and strike out to the northeast, expecting to deliver something much better and much cheaper that we make now. If we surrender to imaginary constraints with unwatchable amateurish videos of professors lecturing, we are doomed, some outfit like this [check those prices for a DVD!] or this will eat our lunch, and we will deserve it.
[Update 9/VIII: Interesting reflections on a real on-line course here]