March 4, 2011
For the last dozen years, I've listened to Bill Gates explain how to improve education. First, it was small learning communities (which he now says the Gates Foundation wasted $2 billion upon), then it was making everybody pass Algebra II to graduate from high school, then it was something else, now it's giving the best teachers bigger classes (see Gates's latest op-ed: "How Teacher Development Could Revolutionize Our Schools").
The weird thing is that the Way to Fix the Schools has basically never been, according to Gates, about the main way the rest of economy gets more productive -- and also the one thing Bill Gates definitely knows a lot about: information technology.
And yet, common sense says that information technology offers the main hope of us ever being able to afford on a mass scale the one educational tool that works more often than anything else, especially with math: individualized tutoring. It often doesn't work, but over thousands of years it's tended to work enough that that's what rich people get for their kids. And it's a lot more likely to work than the latest fad.
Unfortunately, assigning one human tutor with patience, insight, and communications skills per student is mind-bogglingly expensive.
So, the standard Ed School solution is "differentiated instruction:" i.e., the teacher should be every student's personal tutor. The teacher is supposed to walk around the classroom instantly diagnosing why each individual student is screwing up and giving the exact help he or she needs. Thus, the need for Superman.
Yet, assigning one computer per student is getting cheaper all the time. And computers have all the patience in the world. It's easy for a program like Aleks to generate math problems adapted on the fly to the exact level of the student -- if you miss a question, the next one is easier, if you get it right, the next one is harder. That's how big tests like the GRE and the ASVAB work today.
What's harder is getting the computer to figure out why the student gets wrong a problem at his appropriate level. Yet, that's not an impossible task in math, where there are a finite number of ways to screw up.
Folks, it's 2011. Way back in 1998, my Palm Pilot could humiliate me in chess. We're not talking about beating Gary Kasparov or Ken Jennings here, we're talking about reminding a kid who thinks that -3 times -3 equals -9 that a negative number times a negative number is a positive number.
So, what would have happened if instead of investing billions naively chasing social theory fads, Bill Gates had invested billions over the last 12 years in something he knows about: software.
Would that have done more good than investing billions in the Ayres Brothers' small learning communities idea? Maybe not, even probably not. But would it have done less good?
So, why not?
Sometimes I wonder if all of Gates's interventions in American education aren't a sideshow to distract from Microsoft Windows' unsuitability for integrating into classrooms. Lots of schools have Computer Classrooms where the students file in and do stuff on computers under the eye of the Computer Teacher who comes around and fixes stuff when it breaks.
What hasn't worked well, at least up through Windows Vista (I have no experience with Windows 7) is integrating PCs into regular classroom work. When the solution to hangups it to reboot and wait ten minutes for the PC reload Windows, well, as Glaivester pointed out a few years ago, when he was teaching and the whole school was given laptops, learning dropped precipitously because he spent half his time fixing kids' computer problems.
That's why Steve Jobs' Winston-Smith-loved-Big-Brother approach to limiting what you can do on Apple products (e.g., no Flash) to what Steve thinks you should do may be more promising for classrooms.
(By the way, here's Ridley Scott's "1984" Apple commercial announcing the Mac during the 1984 Super Bowl broadcast -- talk about irony.)