Positive frustration

September 25, 2012 11:26:09 PM CDT

Thomas Edison has been quoted as saying, "If I find 10,000 ways something won't work, I haven't failed. I am not discouraged, because every wrong attempt discarded is another step forward."

I did a lot of stuff over the past few days which was progress in that sense of the term.. attempts to pound wood and wire into shape around an idea that, once complete, successfully showed the weak spots in the original idea. They aren't failures, and I know that. Sometimes the most effective way to find the holes in your thinking is to fall in them.

Doesn't mean I have a bottomless appetite for it though..

If I made anything I'd call a mistake, it was in switching between about five different projects that all involved something new. "New" is the stuff you can't predict because you haven't spent enough time poking it to see what it does. You also can't put your observations in context because "context" is what you acquire with experience. All you can do in the early stages is try something and say, "well, okay, that's what it did this time."

Keeping notes as you go helps. Dumping everything onto paper keeps you from having to carry all that unconnected experience in your head as you distill it into knowledge. The writing also slows you down. It's far too easy to start chasing one half-formed idea, then go off after another half-formed idea. Do that about twenty times in a row and it will be a miracle if you haven't gone in circles, or can recall even half of what you've done along the way.

The simple act of writing down:

  • Here's where I am right now.
  • Here's the result I want.
  • Here's what I'm going to try.

and eventually,

  • Here's what happened.

Keeps you from charging into failure like a puppy chasing a butterfly.

That doesn't mean you ignore the new ideas. You just write them down instead of acting on them the moment you think of them.

It also helps to move from one project to another. You can't assimilate the information you have while you're still collecting new info. In this case though, all the other projects also had chunks of the unknown to explore. Even though I moved from one set of tasks to another, it didn't give me a break from saying, "oh look: something else that didn't work and/or needs improvement. What do I want to do about that?"

Eventually I had to stop, clear my head, and get some perspective on the whole situation.. this post being part of that process.

In retrospect -- and going through the thirty-odd sheets of notes that I took along the way -- I can see that I've picked up a few good habits:

First, I did a pretty good job of treating partial success as success that could get better, not as failure. That's good mental hygiene. There's definite evidence of frustration (being able to swear in four languages makes for some interesting notes), but it never swallowed its own tail and turned into a bout of depression. Given that I have some bipolar tendencies, it's good to see that I've gotten better at dodging that bullet.

Second, I seem to be picking up a knack for reducing the number of unknowns that I have to tackle. The sections of the notes that make the most colorful reading are the ones where (in retrospect) I can see that I was trying to juggle two unknown factors at once. On the electronics side, there's a place where I was trying to juggle the amount of linearity you can get from 74HC04 inverters (it's possible, but the easiest solutions play hell with the input and output impedance) while also trying to translate active filter designs from op amps to inverters (heavily dependent on input and output impedance). On the physical side, there are places where I was trying to adjust the materials, dimensions, construction, and function of some storage boxes I want to make.

At the worst of those moments, I'd stop, set "good enough for right now" values for all but one or two variables for, and explore the alternatives for only a couple of unknowns.

Third, I've gotten a lot better at doing research by making small changes to a system I already understand. I have a tendency to look at the small steps and say, "yeah yeah, I know how that should work, let's think a few more steps down the road." Then when the small step turns out to be harder than (or just different from) what I expected, I have to rip out a dozen subsequent things that don't work as expected. Move-and-verify seems slower at first, but covers ground a lot faster than plugging away at random.

Random brain cookies:

The scum also rises. -- Dr. Hunter S. Thompson