Thread by @ShriramKMurthi

A short thread about my "other 90%" rule, frustration with courses, and a nice thing that happened.

My "other 90%" rule is that ugrad courses should be designed for the 90% of students who are NOT like us: not crazy about our subject, bound to grad school, etc. 1/n

What can THEY take away from our subject? What can they learn that might most impact them? The impact can and maybe should be both intellectual (expand their mind) and practical (give them new skills). I've always tried to apply this to my PL course. 2/n

When I talk about this to colleagues, they always trot out a "but who will think of the children" argument for not wanting to do the same: they're thinking of the student exactly like themselves. I get the feeling. I just think we should resist it. 3/n

My view is that that student will thrive in almost any course. Sure, they could have had their development bumped along a bit more. But if they're going to grad school anyway, there's plenty of time. I'm more worried about the "collateral damage". 4/n

This has led to my long, gradual, frustrating PL course evolution from an arch Dan Friedman disciple into something…else. I'm still trying to figure it out. And the more I've studied CS ed, etc., the more frustrated I've grown. 5/n

The running joke in my students' office is every Oct I walk in and say "this course is crap, I really can't stand it, we need to do something about it". They took me seriously for some time, then started to laugh after 2-3 years and now they all do. The struggle is real. 6/n

The mid-2010s where when my frustration was really growing. The little tweaks I was making were not really helping. Frustratingly, you'd never sense the problem from the course evals, which were great. It was working for the other 90%. 7/n

But I needed to take the course to another plane. Thanks to some insights this year, I think I've finally figured out the next step, and I'm working it into PLAI 3/e. @jbclements has been invaluable. So after 20 years I *think* I have *some* idea of how to teach PL. 8/n

But something awesome happened today. Someone emailed me about a former student on the fac market. I checked: *20%* of that PL class is now doing great work in top PL PhD programs. Some of y'all are their advisors. 9/n

Some of it was just the joy of "ooh, yet another @BrownCSDept grad who's gone off to do formal methods"—the fruit of the program @logicpreacher and I have been putting in place. But also, concrete evidence that "the 10%" are DOING JUST FINE. Weren't hurt by my "populism". 10/n

(This, incidentally, is a student who worked mainly with @logicpreacher. His "Logic for Systems" is also built entirely around the "other 90%" principle, from the title onwards. So this really is a consistent theme.) 11/n

So if you feel "I could teach a course that attracts a lot of students, or I could teach one just for students like me"—if you think your area can change the world, go the former route. Be brave. It's probably not the course your advisor teaches. It's okay. 12/n

But if you feel "but then they won't be shovel-ready to work on my research project", well, I'm going to be judgmental here: that's wrong, and you're being bad. They didn't come to college just to work on your research. Let it go. Teach them what they need later. •