Write Yourself a Manual

Inspiration comes from unexpected places.

I went to BYU's Acoustics Research Group (ARG) meeting this morning. For those of you who aren't familiar with all the weird things I'm into, I spent a large portion of last year conducting acoustics research on the Nigerian Udu clay pot drum, culminating in a presentation at the meeting of the Acoustical Society of America in San Diego in the fall. We are still working on finishing the paper and publishing it (look in all your upcoming acoustics journals for "Equivalent circuit modeling and vibrometry analysis of the Udu Utar Nigerian drum!"), though progress has slowed a bit due to my professor taking a post at Los Alamos National Laboratory and me having, as always, a rather busy schedule. Anywho, I like to go to the meetings for the free food and the friends I have made, as well as to look for more opportunities for research on weird and obscure instruments. Hopefully I will get to do an analysis of Gamelan instrument tuning, but that probably won't happen until after India comes and goes.

The meeting focused on proper and thorough documentation of one's work. A big problem, especially at university labs, is the high turnover rate of researchers. People graduate, switch majors, get married, etc., etc. Large and small projects often have to be set aside for months or years. When new folks come along and see a half-completed project, they jump on board only to find that they basically have to start from ground zero because all the last researcher left behind were results of experiments without any notation as to how the experiment was set up, what the 239th line of code does, what the names of their files mean (ttl2Version1.8? What is that?! The person who wrote the file name probably forgot too.), ad nauseum. This problem also occurs when time comes for thesis defenses, dissertations, or graduation boards.

"What protocol did you use to obtain these results?"
Trying to remember, thinking back four years...
"Umm... Seven?"

To avoid problems like this in the future, the presenter said, it's a good idea - a great idea - to write yourself a manual as you go along. Include things like:

Research methods
Motivations
Findings
Purchase orders
Equipment and settings
Plans

I can't tell you how many times I've worked through half of a math or physics problem, set it down to get a drink of water, and come back wondering, "now why in the heck did I do that? I'm sure I was going somewhere with that derivation..." and had to start over. There's a mindset that occurs when in the middle of hard work that has to be recaptured when work is resumed, and methinks something like a manual would help with speedifying the recapture process. The idea is to not trust your brain, and to leave notes detailed enough so that someone else could easily pick up where you left off.

I don't know about you, but I don't trust my brain. My plan is to keep more detailed notes, including the items above and probably adding in more things as trial and error works its magic. This will help with the field study. It will probably also help with whatever else I end up doing with my life.

On a closing note, there is also a bit of sick pleasure I get from knowing I will leave behind notebooks upon notebooks of mostly useless information for my great-grandkids to dig through when they're doing family history. Buahaha. Beau Hilton VII, you just try to find the touching anecdotes. I dare you.