I go up for tenure this year (as in, right NOW), and as part of the hazing process, I have to write a prospectus. A prospectus is kind of like a mongo personal statement/teaching philosophy/research statement/where do I see myself in ten years sort of thing, all in 10 pages or less. Good times. Anyway, as part of the writing of the prospectus, I’m reflecting on my past six years at Carleton, and particularly the last three or so since the third-year review. Since I had a rather lengthy pre-tenure sabbatical, one of the things I have to discuss in my prospectus is how I spent this sabbatical time and how it helped my research.
The thing is, on paper, my sabbatical looks kind of like a failure. I had a long list of things I meant to accomplish, that I really wanted and thought I needed to get done….and I did approximately none of those. In fact, the one thing I spent the most time on during my sabbatical was working on a journal article that ultimately I never submitted. F-A-I-L-U-R-E.
This got me thinking about “sunk costs”—time we spend on things that never see the light of day. On paper, this journal article is a sunk cost, time and research energy I will never get back that didn’t ultimately pan out. My research notebook is full of sunk costs: things we tried that didn’t work, analyses we performed that didn’t yield conclusive results, sketches for systems never realized or implemented. Wastes of valuable research time which is all too precious of a commodity at a liberal arts college.
Except that my sunk costs, my failures, have a funny way of reinventing themselves as successes. While I was writing that doomed journal article, I kept having to rerun analyses and rework some results, which ended up yielding some insights that got me unstuck from a totally different aspect of the project that I had given up as a dead end….and is now a conference paper. The act of redoing the analyses also helped me refine some of my analysis methods and some of the ways I was looking at the data….which led to yet another conference paper. And part of that doomed conference paper helped me with an invited paper I had to submit last week on very short notice….which, most likely, will be published this fall.
So now, I’m less likely to think of research failures as sunk costs, but as opportunities. The failures, after all, are often way more interesting and fruitful than the successes.
This is what a computer scientist looks like 