Most CS Professors Don’t Program

I visited Brendan Murphy’s blog recently, and his latest post reminded me of Alan’s comments to my last post, but in a different direction than (I think) either of them were talking about.

Should passionate teachers who aren’t really good teachers be allowed in the classroom? We aren’t even close to being able to ask that question. I think if we had a perfect world where we had the best and most passionate teachers in all of our schools we would be right in asking that question. This isn’t a perfect world. We don’t have passionate teachers in all of our classrooms. I think a bad, but passionate teachers are more likely to, at least be willing to learn. While just plain bad teachers likely just don’t care enough to try to improve.

via Philosophy Without A Home.

What does Brendan’s point, and Alan’s about teachers who don’t care to learn the topic themselves anymore, say about higher education faculty?  There are certainly passionate computing teachers in higher education.  Over 1000 of them gather each March at the SIGCSE Symposium.  I have learned a lot from these instructors, and have been completely amazed at the passion and insight of these teaching superstars like Susan Rodger, Owen Astrachan, Christine Alvarado, Eric Roberts, Eugene Wallingford, Mike Clancy, Dan Garcia, and Nick Parlante.  (You know that once I named a half-dozen, I could easily name another two dozen, but I stopped at those that I could easily get from my at-hand lists of Web URL’s because I visit their pages often enough.)  Our instructors at Georgia Tech are (literally) award-winning.  These are teachers who inspire students to go on, but more importantly (as Alan points  out), are constantly learning themselves, who engage in the “hard fun” of their subjects constantly.

Bjarne Stroustrup pointed out in his CACM piece on Jan 2010: Most CS faculty don’t program.  He was particularly focusing on developing systems, and the importance of knowing how to build systems in order to teach students about building systems.  I’m concerned about a less complex goal.  Introductory courses in computer science use programming — whether it’s all about programming (it shouldn’t be), or programming is just a notation, or if programming is used in order to rediscover the beauty, joy, and awe of computer science.  Higher-education teachers that I know, who succeed in teaching those introductory courses, still program, to produce examples and often to live-code in their classes.

There are probably very few computer science professors who have never programmed.  There are relatively few who still do.  This is where Alan’s point, about “this stuff is a kind of yukky medicine that you have to take to get well, but I’m healthy so I don’t have to do this anymore” is relevant.  To get through their graduate degrees, these professors all had that passion about their subject, at least at some point.  Brendan asks if that’s enough.  Is it enough to have had passion once for the subject?  Or do we convey to students that programming (and maybe all of computer science) is “yukky medicine”?

Stroustrup’s piece touches on this point. He says that we have to change faculty reward structures so that faculty are rewarded for building systems.  Faculty should also be rewarded for teaching well, so that those behaviors are incentivized.  Maybe these goals are related.  Faculty should be incentivized to engage in the activities that they’re teaching, to convey positive messages about them.  If we can, we should reward for passion and engagement in teaching.