11 Aug 2008--starting out
Aaron, Soojin, Joshua, King, Jung, and Mark met in 3100 Klaus, and Charles joined us by conference call.
The meeting focused on the computational needs of students in Biology.
- There are issues to be addressed at the undergraduate level, but even greater issues at the MS in Bioinformatics and PhD levels.
- There was interest in more "software creation skills" for students.
- The Bio faculty felt that CS4710 could use more object-orientation.
- Bio 2400, the course on mathematical and computational models of biology, will probably move away from Excel. Joshua's work is in MATLAB, because he does more work with dynamics, so he may shift it that way.
We came with a set of action items:
- Biology would like CS to offer a graduate version of CS1331 – one that graduate students could get credit for. It might be the exact same course, just numbered CS6331 (or such), and if there's enough interest, offered in a section apart from all the undergrads. Then, MS BioInf and PhD students could take that, if they need more focus on software deve
- Biology undergraduates need to have their computing requirement completed before they take Bio2400, so that that class can assume some software construction skills – that the students are ready to talk about modeling.
- CS1315 (MediaComp, or Media Computation) is not an appropriate computing requirement for Biology undergraduates, not if we expect them to be able to build programs in Bio2400. Biology undergraduates should probably take CS1371 (MATLAB) or CS1316 (Java, with CS1315 left as an implied requirement for those without the computing/mathematics high school background to get through CS1371).
There was a good bit of interest in developing a NEW CS1 (in parallel with CS1315, CS1301, and CS1371) which would use Biology as a context. The idea would be to create a class in computational science, using Biology as the particular science on which the class would be designed. Charles told us about their experience at Spelman, and the need for a textbook to make a course for freshmen work.
Mark described the process used in the MediaComp and Robotics CS1 efforts.
- Basically, we came up with some particularly cool assignments – things that students would want to be able to do, that they would find relevant and interesting.
- For Media Computation, our initial examples were background subtraction, chromakey, negating an image, reversing a sound, and splicing sounds. (See the old page where we laid these out)
- For Robotics, it was for the robot to find its way out of a maze, and later, became storytelling with the robot (with robots as characters, and robots as cameras).
- We then figured out what computing the students would need in order to be able to do these assignments. Negation didn't require more than iteration, but background subtraction and chromakey also needed conditionals.
- We then continued to backfill, finding other examples from the domain to serve as intermediate projects working toward the bigger goals. So we do increasing/decreasing volume, and changing individual color values in MediaComp.
- We then review the CS concepts we're covering and see if there's anything left to be a real CS1. We then add CS concepts and projects from the domain to complete our desired scope.
Mark will call (in early Sept.) for a monthly meeting. In the meantime, everyone is charged with coming up with the canonical examples that we can design a BioCS1 around.
Here's the Whiteboard as we left: (Click on it to see a full-size version.)