Mechanical timers & reaching for complexity instead of better solutions

My sauna (see other blog post) is too hot. I’m sitting inside, sweating it out as we get north of 140°F, and venting it to cool down seems like a waste. Also - once you get in, you need to reheat it. so it’s a real (1st world) problem!

In the current state, the control is outside; it involves unplugging one of the two steamers whilst sopping wet.

My brain immediately went full Silicon Valley: use a microcontroller with a temperature sensor, voice command a la “the clapper”, maybe some machine learning to predict optimal temperature curves based on heart rate variability while I’m at it.

Then I remembered Tachles - that perfect Yiddish word meaning “to the point, no bullshit.” Getting down to brass tacks.

I bought a mechanical timer. Set it to 10 minutes on one of the steamers before i hop in. It warms back up to the perfect temp now and then shuts off. There was No need for another device on my home network, likely an internet of rubbish tool with my wifi password. No firmware updates. No dependency on the dumpster fire that is AWS us-east-1.

As software engineers, we’re driven to solve hard problems that often require a complex solutions. We get rewarded and lauded by our peers, they’re more interesting ways to spend our time; thus every nail looks like a distributed systems problem. But many problems aren’t that interesting. Many problems we could solve at scale (with great cost) just need some Flex Seal in order to focus on the real problems that stop us from getting to said scale.

Think about it: where can you pick a metaphorical mechanical timer that works more reliably, with less effort to setup, and well cheaper?

Embrace Tachles.