What book do you think every engineer should read? I'd say Atomic Accidents -- a fun read and fundamentally about how/why things fail. If you want to build things which don't fail, you need to recognize how things do (and how often it's about people). www.simonandschuster.com/books/Atomic... - ThreadSky

leak.bsky.social • 47 days ago

What book do you think every engineer should read?

I'd say Atomic Accidents -- a fun read and fundamentally about how/why things fail. If you want to build things which don't fail, you need to recognize how things do (and how often it's about people).

https://www.simonandschuster.com/books/Atomic-Accidents/James-Mahaffey/9781605986807

Comments

likescalendars.bsky.social•46 days ago

This looks very cool.

I would like to add "The Checklist Manifesto", I think it's pretty good:

https://www.goodreads.com/book/show/6667514-the-checklist-manifesto

leak.bsky.social•46 days ago

The amount of time I spend saying "people gonna people" because humans really are not good at things like remembering every single step of a complex task. If you can't remove the need for the step, plan for it.

likescalendars.bsky.social•46 days ago

I agree in full with what you just said.

Also, it increases velocity: you can do the steps faster, because you don't have to double check every single one.

carste1n.bsky.social•47 days ago

Jurassic Park - as an illustration of what happens when your staff is overworked and underpaid.

on more serious note: "The Art of Doing Science and Engineering"

adamarellano.bsky.social•46 days ago

Oooooooh great suggestions

aleatha.bsky.social•46 days ago

Thinking in Bets, especially when working on high-ambiguity projects. It's a nice soft intro to Bayesian decision theory. I find I'm coaching engineers on "efficiently burning down uncertainty" a lot these days.

snarfed.org•47 days ago

Funny, my pick is Normal Accidents, same topic and nearly the same title 😆 https://snarfed.org/2011-12-09_can_software_learn_from_nuclear_power

thebadcode.com•47 days ago

Also, Atomic Accidents is really good too!

thebadcode.com•47 days ago

"Field Guide to Understanding Human Error". The takeaway is you should never be satisfied with an explanation that blames human error: if you want to improve the resilience of the system, you have to care about how "not making errors" became the only line of defense.

https://www.amazon.com/Field-Guide-Understanding-Human-Error/dp/1472439058

adamarellano.bsky.social•46 days ago

Noise by Danny K et al. Last book before he died. Why humans make bad decisions because of noise, so maybe plan for it.

leak.bsky.social•46 days ago

Like literal noise? Because that definitely doesn't help me think.

adamarellano.bsky.social•46 days ago

Ha. Kind of. If humans made decisions logically and reliably the "signal" of their decision pattern would be smooth and predictable, but humans are really bad at it so the "signal" is actually all over the place and unpredictable. Think really bad long distance connection in the 90s

Comments

Posting Rules

Reply