Eigenmumbles

It seems that every couple years, I go a little nuts around the start of the new year. Some years, this involves sticking my head under a blanket and going on a two-week-long fiction reading binge, with breaks to eat and to shower and (grudgingly) to sleep. Somehow, I don’t think that’s going to happen any more, or at least not until the little one becomes sufficiently big that she also thinks that two weeks of nothing but reading sounds like fun. But some years, a little voice in the back of my head says “hey, you’ve finally got a little time to fiddle with all those things you thought you were going to fiddle with earlier in the year!” This little voice is, of course, somewhat deluded. The lack of immediate deadlines is not the same things as an excess of time, and what I really should do is mark things of my list of deferred chores. But I listen to the little voice anyhow. As I say, I go a little nuts.

Some years, my fiddling involves rewriting chunks of software, or launching headlong into a new bit of code. This year, it looks like the part of my crocodile brain that evaluates shiny things has decided that infrastructure projects are particularly attractive. So I now have a Windows development environment again (hosted under a virtual machine using Parallels), I have my repositories up on bitbucket, and I’ve mapped out some major clean-up tasks for my research codes. I’m itching to finish the revisions to my matexpr tool, and to use this more powerful tool to factor out some redundant garbage in one of my finite element codes. I’ve figured out how to convert PDF files to WMF files on my new Lion laptop, which seems to dislike pstoedit for reasons I cannot fathom. This was susprisingly difficult. And I figured out how to record screencasts on the same laptop. This was surprisingly easy (start QuickTime Player and click “Record New Screencast” in the menu), though I’m still figuring out what precisely I want from these screencasts.

I could expend a lot of energy fighting these impulses, but that seems counterproductive. I enjoy these infrastructure projects because they are fairly limited in scope and they involve a different sort of head scratching from research, teaching, or proposal writing. But I do need to keep up with my other tasks, too. So: this winter session, my mornings are for writing. If I do that, I figure I can keep the afternoons for hacking around, particularly if much of that hacking around produces something useful.

Could this be an early New Year’s resolution? Or would calling it that just jinx my plans?