Following my habit of posting Before/After notes on my semesters, some impressions for Fall 2013 now that every class has met. I’m getting to the point where the bulk of coursework I can and would sign up for tends to be special topics courses, which is a very interesting, if sometimes strange, phenomenon.
Taking: CS515 ALGORITHM DESIGN
This one stresses me out, even with a very laid back instructor. I’m not terribly mathematically inclined, and my model for computing is based on computers, while classical algorithms are very explicitly not. I consider classical algorithms to be one of those things that everyone in the field should know, and everyone should also know is a model that doesn’t quite match reality. I read/listen along and am constantly suppressing a stream of phrases like “NOOOO you can’t do that, your cost model doesn’t correspond to any computer I know of.” “No, there are additions, you need a counter unless you do it recursively, in which case the function call overhead becomes the dominating factor.” “That nonlinear memory access pattern will be the limiting factor.” etc. I don’t expect it will be too awful, but it will be uncomfortable, and probably in a healthy and enriching way.
Taking: GS600 SPEC TOPICAL GRAD COURSE/ University Curriculum
I’m taking it because it sounded fun. It is apparently also being taught because it sounded fun. This is a good combination. The class is composed of ~9 grad students in a variety of fields, meets for two hours in the evening once a week, and is a discussion setup, which is likely to actually discuss. I’m hoping to get to devote some cycles to considering and discussing the unusual practical/theoretical balance issues that show up in CS and Engineering disciplines. No textbook, selected source readings, and other than participation, there are only two graded assignments; one paper about designing an undergraduate curricula, and one talk about our own undergraduate curriculum -which since I gamed the system at the time, I already have the documentation and cross-referenced spreadsheet for. The paper parameters were defined as “I’m not specifying a length, I’m pretty sure you all can write” which is a strong sign that this course will get to follow educational ideals a bit more than most.
Taking: EE699 TOPS IN ELEC ENGR/ Cameras as Computing Devices
Also, to be fair, taking this for fun – I wanted an excuse to misuse consumer electronics for credit, this provided one. The fact that I know the instructor tends to teach interesting and low-pain classes helps. The class is split into thirds; [digital] photography principles, image capture, and processing. Camera and image processing is one of those spaces I’ve dabbled in but never done any serious study, so it will be nice to fill in the gaps, and the refresher on largish embedded systems is also a desirable feature. It’s a small class, which helps enable getting really hands-on with hardware.
TAing: CS275 DISCRETE MATHEMATICS
Third time around. I’m now pretty comfortable with the material and process at this point, and the primary instructor is one I have done it under before, so I’m not expecting any surprises. I do have something like 75 students across two sections, which will be a pretty big consumer of my time for grading and the like. The only known uglyness is that one of the rooms is very poorly configured for the course – not having at least a double-width board sucks for demonstrating involved problems. I do have some disability accommodation to work in, but so far the back and forth on that has been very successful, and we are quickly arriving at a set of tools that work.
I expect a professionally enriching semester, and hope that the fact it is scheduled such that I only have fixed obligations TWR will let me get some of the research I’ve been hoping to get off the ground going.