Matthew's Journal

Some days at work I really wish I had cow-orkers who also did a lot of coding. Today has been one of those days. I've spent the day thus far debugging a piece of code I wrote the other day. After many runs, and adding a lot of fprintf calls for debugging, I eventually tracked the problem down. It wasn't in the new and quite complex algorithm I'd written recently, but rather in a tiny utility function I wrote and tested ages ago. Under one rare edge case it would return the wrong length answer, leading to a bit of code calculating the dyad census wrongly.

All a bit frustrating, really. Still, the machine is now crunching data and running simulations, and by tomorrow I'll have some more stuff to put in the talk I'm giving next week...