I’ve heard people say that research is a long slog rather than moments of sudden brilliance. I’m wondering if it can’t be both. A bit like geology itself, a combination of slow gradual changes interspersed with rapid events. Research can certainly be a long slog, hours spent in the lab, or in the office reading and processing data. But there are moments of excitement too. New ideas don’t tend to form gradually in my mind, they tend to jump out, and wave surprise.
It doesn’t happen very often and these new ideas typically result in me running through the corridors of the department between my office and the printer, grinning like a maniac, and occasionally dancing, or fist-pumping or something that makes me look like a fool.
Two weeks ago I was feeling a bit down about my research. I’d just got back from Goldschmidt, and all the excitement of the conference and the ideas that I’d had began to fade. I was left with the realisation that my research was still in the same uncomfortable place that it had been one and a half months ago, when I downed tools to work on my conference presentation and then headed off to Canada.
Machines that I wanted to use weren’t working, still. One had been down for months and there was the possibility that I wouldn’t get on the machine for another few months. Another thing that I wanted to do was languishing somewhere between impossible and never going to work, and merely just very difficult. I’m still waiting on dates for my samples that I sent off back at the start of December. I was in a bit of a research funk.
So with no prospect of getting any new data in the next month or so, I decided to sit down and really pound the data I already had into submission. I spent the last two weeks screaming at Matlab again. I produced graphs, I’ve churned through my data. I’ve poured over other peoples data, looking for connections, similarities, differences. I’ve analysed the living daylights out of any relevant data I can lay my hands on. I’ve performed all kinds of tests, discovered what a z-score is, compared means, and standard deviations. By Friday afternoon I knew I was getting close to something. That my week of data processing was leading to something special.
Now, 6pm on a Friday evening is a kind of inconvenient time to discover something new about your research. As I was frantically scribbling away at a graph, trying to work out exactly what my residual z-score meant in terms of the real world, my friend Alex came to my office door, enquired about why I hadn’t been at the department’s weekly social football match and if I was coming to the pub for beer.
I had to turn him down, turning down beer is tough, but this was an exciting moment of science. I had produced an amazing graph, detailing 1) my data, 2) the similarities and differences between my data and another person’s data, and what that difference meant in terms of physical processes and 3) a beautiful correlation with a potential mechanism as to what is causing the changes in 1 and 2. All three of a holy triumvirate of scientific discovery, sitting there prettily on my beautiful, beautiful plot.
I had done it, from low to high in a couple of weeks. And I still made it to the pub for beer.