I got both good and bad news today on the circuit course.
The good news: it seems that we can pay the lecturer for at least part of a course, because he is only at 95% for the year, not 100%.
The bad news: the School of Engineering undergraduate curriculum committee met this morning and did not approve the course.
Somewhat predictably, EE thought that this course was too similar to their circuits course, and so did not want it offered. Somehow the undergrad director for EE (whom I’ve discussed the course with before several times over the past few years, and who had been pointed to this blog for more details) had thought that I was proposing a much lower level course originally—one at the level I think of as high-school electronics. He thought we could pick up students from the Digital Arts and New Media program (rather their own DANM 219 Introduction to Electronics for Artmaking). I had explored that possibility, but few of the arts students take calculus or physics, and even simple concepts like RMS voltage need integration. We definitely need some math sophistication in order to do phasors.
Once the EE undergrad director realized that we were proposing a more applied version of the circuits class, covering 50–70% of the same material, but with slightly reduced prereqs and less of the formal math, he was opposed.
The computer engineering undergrad rep would have liked us to propose a lower-level course, that combined some practical electronics with the physics E&M material, because they were not happy with the difficulty engineering students were having getting into the E&M course, nor with the way Physics was teaching the course. This would be really difficult, since the physics course is already a full quarter and would leave us essentially no time to teach any of the applied circuits material.
Computer Engineering would have been content with our offering the course as an alternative to the EE circuits course, as long as it was accepted as an adequate prereq for two courses their students are required to take: microprocessor system design and signals and systems. I’ve never taught the microprocessor system design class, but most of what it needs from a circuits class we would definitely be teaching, probably more than the existing circuits course (lab skills, interfacing to microprocessors).
The Signals and Systems course, though, is a can of worms, because of poor factoring of material into courses by the EE department. Rather than putting transform methods in the signals and systems course (where it would get used), they pushed it into their gateway course: circuits. As a result, there is a rather indigestible lump of transform methods in the circuits course. And that lump is almost the only reason the circuits course is a prerequisite to signals and systems. Part of what is wrong with the current circuits class for the bioengineers is all the matrix methods and transform methods that are included but only motivated by “you’ll need this in later EE classes”, which bioengineers mostly won’t take. I see no chance that EE will ever refactor their courses.
For our course to do the transform methods, we would have to put back all the math prereqs (we’d cut out the ODE course and linear algebra, leaving just single-variable calculus). We’d have to chop out two weeks of material (20% of the course) to make room for transform methods, which most of our students will then never end up using again. It would also mean that we couldn’t attract any students from biology to the course,whose math stops at single-variable calculus.
One idea that was floated was to change the course numbering so that our applied circuits course was a lower-division course (freshman or sophomore year). I’m perfectly willing to change the course number to a lower-division number, since I would prefer students to take it in the sophomore or junior year, not the senior year. I’d even be willing to have students who wanted to continue into bioelectronics have to take the EE version of the circuits course after our course in order to get the prereqs for upper-division electronics classes (though much of the material overlaps). That is, the applied circuits course could be a terminal course, with no subsequent course using it as a prereq, since most of the bioengineers will be content with that level of electronics. Somehow this idea got rejected without any arguments against it being made, so I don’t know what was the problem with it.
Another idea that was floated was to have the regular circuits course taught without significant modification, but to have the applied circuits lab offered to the bioengineering majors as an alternative to the usual labs. I think that there was recognition even from the EEs that the labs I’ve been designing were better than the ones they’ve been using. But I’ve been designing the course for the labs and the lectures to be integrated with each other, with a fair amount of book-learning needed to support the labs. Tacking the labs onto a mis-fitting lecture course is not going to work. The students will need information about electrodes and electrolytes for the electrode lab, about thermistors, strain gauges, and microphones, about biopotentials and volume conduction for the EKG, and so forth. A lot of thought went into the order of presentation so that students would just have learned concepts that were then reinforced in the lab. Without that tight integration the labs would likely fail. To make it work with an unmodified (or only slightly tweaked) circuits course that spends weeks on matrix methods and transform methods that aren’t particularly needed in the labs, the lab would have to be a separate 5-unit course, and then the students could just take the lab without the EE circuits course, getting us almost back to the current rejected design.
We can teach the course without approval this year as a “group tutorial” (which is a common workaround for the slow approval process), but I don’t see how we can ever get approval for the course if it requires EE blessing. They don’t see anything wrong with the design of their current circuits course and see no reason for there to be any “competition” for students. Not that they want to teach circuits either—they’re talking about reducing the offering from twice a year to once a year, even though there isn’t really lab space for everyone currently required to take the course to all take it the same quarter. (I think that is where the idea of our modifying our course to be an exact clone of theirs comes from—some other department paying for their course.)
EE does worry about their low enrollments and few majors, which I believe are largely caused by their gateway course being more of a barrier than a welcome (and perhaps also by how few of their faculty care about undergrad teaching). Their preferred solution, though, is to merge with the computer engineering department, which has more majors and a history of good teaching, then expect the computer engineering faculty to teach the undergrad courses while they teach only boutique grad courses.
We seem to be at an impasse, and I’m not enough of a politician to see a way forward. I’ll be meeting with the EE undergrad director (probably next week) to see if we can come up with a compromise, but I’ve no idea what that could be right now.