For as long as I’ve been doing research, I’ve tried to come up with a “system” that works for keeping track of the loose ideas that filter through during talks or in idle moments. Earlier on, the focus was on recording technologies — scraps of paper, an audio tape recorder, the Palm Pilot and its inscrutable Graffiti. The Moleskine reared its sightless head, promising an end to the torment of lost epiphanies. The sad truth of course is that progress doesn’t come in lightning flashes of brilliance while strolling down a boulevard, but instead in obsessive fits and starts. So instead I’m trying to settle on a simpler method of keeping track of research questions — a text file.
Once the medium is fixed (and mutable), the central question is taxonomy. For a while I was using subject matter — Shannon theory, networking, etc. More recently I decided that it’s better to classify things by “how much I have thought about them.” I have 5 categories :
- Pie-in-the-sky : wow, that seems interesting… I should think about it for more than 5 minutes…
- Nebulous hand-waving : there’s a problem there, but what is the right framework?
- Percolating : ok, but what is the actual formal problem?
- In progress : finding the proof in the pudding.
- Writing : oh no, a deadline! Gotta figure out how to fit the page limit!
This seems to be working ok for me at the moment, but it’s chasing a moving target, it seems. Do any of you readers have systems that have worked out for you? Yes, this is a blatant plea to those lurkers out there…
An Ergodic Walk 
I have an “idea box” – a little index card box on my desk with little index card dividers for different genres of ideas (since some are more design than research). Some of them I just have a title scribbled at the top to remind me what it was, some I have more concrete steps written out below.
Maybe someday I will be a grownup researcher and need those 5 categories, too. For now, very little is beyond 1 or 2.
I go through organizational systems like once or twice a year, though, as far as actually organizing the projects. I just can’t stay on top of one, and then the next one looks oh so shiny.
The “shiny” problem is killer — i guess the filing system is always greener on the other side… of the cubicle. Ok I’ll stop analogizing now.
My text file generally has sections for various classifications of ideas. It’s become three text files for three major categories with subcategories semi-hierarchically arranged. The most important ideas are on the top of each subsection. Eventually some of these ideas graduate to actual formal documents, or at least comments on these documents. However, this is one of several systems I have, including notepaper (for non-text), LaTeX (for embryonic papers), C and MATLAB with comments (for algorithms), emails (for collaborative ideas) and small special-purpose text files (for whatever doesn’t fit elsewhere). I keep all these files synchronized to a mobile device so that if two of the three computing devices containing them are destroyed, I’ll still have the electronic documents. Redundancy, as always, is key.
I keep what I call ‘thinking books’ – large, hard-back bound notebooks in which I mull over ideas in old-fashioned handwriting. Because they are bound rather than loose-leaf, the organisation is inevitably by date – they are in effect journals of how my thinking develops on a project. The ideas usually need more formal collating and processing before they get into full written form, but the chronolgical organisation makes the act of revisiting and re-working ideas very clear, and reminds me that theorising is an interative rather than linear process. Very low-tech I know, but it’s a system that’s worked for a PhD and two books, so at least it’s tried and tested!
liz