s20451's Journal: When the professor gets schooled

I'm an assistant professor of computer engineering at a large public university. I'm also the department newbie, having been hired over the summer, so my first class as a real, live, full-time, tenure-track, daddypants-wearing prof wrapped up on Friday. I was teaching a digital logic design course, which was assigned to me randomly, and outside my area of specialty. Still, I was very pleased when I found out I would be teaching it - I liked it as an undergrad, and was a lab assistant for a similar course as a graduate student.

(As an aside, I should mention something for those who have never been on the salaried side of the desk at an institution of higher learning. At my school, the sum total of the direction I received was to be told what course I would be teaching, and to be given the calendar description of that course. Nobody ever asked to see my course notes, my handouts, my exams, or my assignments; nobody in the administration ever sat in on a lecture. I could have been teaching Faust or talking about my dog and nobody would have been the wiser. That's not to say that support was unavailable: the department bends over backwards to give help to faculty who want it, but they stay completely out of the way unless you ask. It's an exhilarating amount of freedom for someone fresh off of the PhD-postdoc treadmill.)

I thought I would do something a little different from those professors who had taught the class before, and borrow an idea from the prof who taught it when I was a lab assistant. I decided it would be a great experience for the students to take the last three weeks of lab periods and do a design project, where they would build something interesting such as a digital clock, a game, or something else similar. Using the in-house FPGA board (we use the Altera UP2, which is standard for a lot of computer engineering programs), it should have been easy for the students to come up with some interesting, non-trivial design problem, giving them design experience and reinforcing the lessons from the class. And everyone would be in awe of my pedagogical greatness.

Should have been. I forgot the most important rule of engineering design, which is that designs never fail in theory, they fail in practice.

The students had no end of problems, most of which were the university's fault for having poor equipment (and mine for not anticipating their needs). For example, the UP2 board has two chips: an EPM7128 for small projects (perfectly suitable for simple lab-bench stuff), and a Flex10K for larger projects (which most of the students ended up having to use). I guess nobody had ever used the Flex10Ks before, because nobody had soldered the appropriate headers onto the boards. After some scurrying around to get that problem fixed, there were no ribbon cable connectors to attach to the headers, meaning that an ad-hoc and very inefficient solution had to be found.

Then the Flex10K was found not to provide enough current to power external seven-segment displays. There were voltage problems with externally-attached chips - some students wasted their money buying external components that did not end up working, which surprised me in a lot of cases. There was a shortage of boards: three boards were left for the students to sign out after hours; because of the lack of ribbon-cable connectors, it was basically impossible for them to share, because they couldn't just unplug their circuit and hand the board to their friend.

And then - because I underestimated the difficulty of the project - I only assigned it to be worth 10% of the students' grade, in spite of the fact that everyone was putting in long hours in the lab for a solid week or more before the due date.

In the end I was very impressed with what the students produced: a couple of clocks, a "slot machine", an LCD text display, a tic-tac-toe game, and so on. A lot of the projects achieved their main objectives, in spite of the problems, and I marked leniently to acknowledge the long hours that everyone put in.

In the wrap-up class on Friday, I asked for feedback on the project, and got a host of these complaints. I am dreading my course evaluation results (but hey: get a low mark in the beginning and then move up; it's all about the improvement).

I'm having the students write up a list of hardware problems they encountered, and with the benefit of this experience, I think I will do the project again next time. But we professors learn things every day, too.