Tuesday, 31 January 2006

Planning on two fronts

Modern computer processors have this feature called “speculative execution,” in which they try to guess which code path is more likely to be executed, and actually go ahead and execute that code before seeing which path is taken. This is a perfectly sensible strategy; however, if the prediction is wrong, the computer has to “undo” all that work and take the other branch.

Today I engaged in a little speculative execution of my own by putting in the paperwork to teach my (hypothetical) research methods class in the fall in an Interactive Computer Classroom. It’s not a huge investment by any stretch of the imagination, but every little commitment edges me down the path of spending another year in Durham. And given the state of the job market these days (and perhaps my generally picky nature), the chance of a branch misprediction seems rather low at this point.

Wednesday, 1 February 2006

Branch misprediction

It seems I spoke too soon; the financial powers that be apparently are unconvinced that keeping around a visitor to do something (teach sections of undergraduate methods) that perhaps a dozen or more of the department’s tenured or tenure-track faculty members are nominally qualified to do needs a bit more justification, particularly in the midst of a “budget crunch,” at least by “university that has more money than the Queen of England” standards. Nothing definite, but playing the odds in the presence of asymmetrical payoffs for misprediction seems like a bad idea at this point—potentially wasting hours of my life on applications beats potentially having to beg for a job at Best Buy or Red Hat, any day.

On the upside, a whole new vista of postdocs and one-years have been opened to me. Happy happy, joy joy.

Incidentally, I’m reminded of one of my favorite NewsRadio quotes: “You can’t take something off the Internet. It’s like taking pee out of a swimming pool.”