Worked on the PyCon paper some more this evening (when I wasn't bumping into bugs here and there in OOo). Mostly just getting the section introducing data descriptors roughed out and doing some more work on the functions-to-non-data-descriptors generalisation.

Problem is, I still have a lot of material to cover and I'm spending entirely too much time on these "transitional" sections, which probably means that I'll need to come back and rework them later anyway. If the design isn't flowing in a particular area, it's probably better to just charge through and come back when you have a better idea of the sketch for the whole project.

Think I'd better set a deadline of Friday evening for getting the paper finished so that I can get the first draft of the presentation done over the weekend... which means not likely much work on OpenGL.

Still, the paper and presentation need to get done, I guess they're now officially my top non-work-project priority.