Too often, the defenders of free markets forget that what we really want is free men.
— Michael Crawford
Someone once said to my dad, “Nothing great was ever created by a committee,” to which my dad replied, “Except the King James Bible and the US Constitution.” I don’t know as much about the design goals for the former, but I do know the latter was intended to achieve good government using flawed people. Pre-Enlightenment philosophers talked about just kings with wise advisors; the genius of those who gathered in Philadelphia in 1787 was to realize that they couldn’t rely on either being available, and to create a system that would produce acceptable results with real people. (This speech is fiction, but it’s still the clearest and most inspiring statement of this principle in recent times.)
Which brings me to an exchange on Twitter today. JD Long asked, “OK, we’ve been burned by DataCamp. What do we need from an ed site that would prevent this in the future?” One of the tweets in the stream that followed said:
+1 to the team teaching model. Anything that runs like this requires a well-trained team of educators (instructors and assistants) supported by instructional designers, education developers and media specialists.
With respect, anything that requires that isn’t going to work. Teaching is always a day late and a dollar short: it’s the largest part of what universities do, but most faculty aren’t taught how to teach and aren’t given the resources to teach well, and while many companies pay lip service to “lifetime learning”, my experience is that training and re-training usually wind up being something staff get to do in their “other” 40 hours a week.
I didn’t design Software Carpentry with this in mind (mostly because I never actually designed it), but one of the things we got right was a structure that could achieve good results with the people we had and the time they could give us. Instead of a well-trained team of educators, we had volunteers with a one- or two-day introduction to a few key ideas. Instead of instructional designers and media specialists, we had other volunteers with commit privileges to GitHub repos and the authority to use their own best judgment. And while many of us may have read How Learning Works, we drew more on Lang’s Small Teaching and Huston’s Teaching What You Don’t Know.
The thing I want most in an ethical replacement for DataCamp is checks and balances. Make it harder for those with power to cover up misbehavior. Ensure that information about users won’t be sold to third parties, even if the platform is bought or the idealists in the C-suite are sidelined. But the thing I want second is to create a commons or a co-operative by erasing the distinction between teachers and learners. I don’t just want a platform for teaching data science; I want one that teaches participation in the way that the public education system my parents devoted their lives to taught people how to be citizens.
Stack Overflow is proof by example that people will create a “chorus of explanations” if given the means to do so, thereby erasing the distinction between “teacher” and “learner” in the way that the men who gathered in Philadelphia erased the distinction between government and (some of) the governed. If we teach people a little bit about teaching and a little bit about how to organize, I believe we can create a republic of learning: a place where imperfect people with too little time and too few resources to do the job right can nevertheless do it well.
I organized an online study group two years ago to work through Brown’s Building Powerful Community Organizations, which is the most practical guide I know to grassroots organization. If you would like to take part in another such group this summer, please let me know.