On Redefining Education

I've been thinking lately about the problem with schools (actually, I've been thinking about that since at least the 6th grade). I'm still trying to refine the argument, but I think it comes down to the fact that there is no so much information in the world that is ever-rapidly being produced that old ideas of learning "the basics" just don't make sense anymore. It used to be that you could learn how to design bridges, heal the sick, or, maybe build cars (probably using very few skills that you learned in school) and be set. Now, however, the job that you'll be doing in five or ten years, especially if you are a high school senior, probably doesn't exist.

As usual I see working with computers as making this case more clear, but I think that these principles are fairly universally applicable. In the late 1980s when I came across a computer problem, I would work (1) to understand the cause of the problem and (2) to remember or make notes of what I had done so that I could more easily solve similar such problems in the future. That's pretty different from what I do now.

Trying to truly understand the source of a computer-based problem is almost always just a bad idea. The systems and interactions among them are so complex and unpredictable that it is often not possible to truly understand the true source of the problem. More significant, though, is that it's usually not worth trying to remember whatever the solution was, since the set of one-time problems is increasingly large. (And arguably more-so with Open Source Software that has a much shorter life-span between revisions.)

Here's an example, that may not be a good one. I upgraded Moodle last week. Shortly thereafter I duplicated my Spring 521 course to make it be the foundation for me Summer 521 course. When I did, all of the assignment names were missing and trying to edit one of them resulted in an error. My attempts to search for an explanation came up short and I ended up posting to one of the support forums, which I almost never have need to do. Mostly Moodle works, and when I do have problems with Moodle or anything else, usually someone else has had a similar problem that they have documented (which, the astute reader will point out is perhaps antithetical to my assertion that we don't need to remember or document our problem solving experiences). This time the problem seemed to be unique, or those who'd had it hadn't asked about it somewhere that I could find. I finally found a note that someone had made about removing a certain module and that solving this problem (or maybe it was just similar). I looked for that module and didn't have it installed, but the module page said "The X module is incompatible with this version of Moodle." I removed the errant module (not a part of the standard Moodle distribution, which is why no one else had documented the problem). I did go back to reply to my posting about the problem, but I know that this particular problem isn't going to happen to me again.

So what we need to learn is not a rudimentary set of skills and knowledge about computers, but a rudimentary set of skills about how to find solutions. Solving problems with all of the tools that we have available in the real world is what we need to teach people to do, not regurgitate some facts on a test. We don't take tests in the real world. Designing tests that are true predictors of someone's success in another situation is very, very difficult. John Bransford and Dan Schwartz call this "Sequestered problem solving" (and I need to re-read Rethining Transfer).

As I love to say, Socrates distrusted those who had to rely on text to "know" things, arguing that if you had to use text, then you didn't know it and using text as a crutch in this way was misguided. This sounds rather silly today. And if you think that kids looking things up with Google and/or wikipedia is somehow inferior to walking to the library and looking it up in a book, I suggest that you're going to look just as silly.