In almost any undertaking, there is the ‘right’ way to do it … and a way that works best. The trick to being effective is to realize that the two aren’t always the same, and when that
Often, knowing when ‘right’ is not best comes from experience. In almost any profession, it seems like there is the old-timer for whom the process is more of an art than a science. Think of the machinist who knows exactly when the machine should be pushed beyond its nominal limits, the teacher who knows which students to motivate through praise and which are driven by criticism, or the plumber who knows that sometimes it just needs a good whack.
One of the (many) issues I have with bureaucracy is that it dictates a ‘right’ way it demands be followed, regardless of whether it’s the best way.
I experience this dichotomy all the time (it seems especially prevalent when building things) but two recent run-ins with it inspired this post:
- Last night I visited Evil Mad Scientist Labs, which is slightly more mundane than the name would suggest, but still freaking cool. All of the clever tricks and tweaks they used to get their products to work made the traditionally educated engineer in me cringe, but that feeling was quickly overridden by the excitement from the by-whatever-means-necessary builder.
- In order to get the micro-gravity test apparatus working at Ames, I need to find the correct PID gains. The ‘right’ way to find them [warning: irrelevant technical jargon] is to model the system, transform it into the frequency domain, plot the poles and zeros, and create a Bode plot to make sure it will perform as desired. Unfortunately, the system is so far from ideal that such an approach would never work. I’ve learned that in reality, the best way is to essentially try a bunch of different combinations in an only vaguely scientific way.