1, #194 - Context doesn't scale.
Estimated Read Time: 59 seconds.
Last week I found a new word I love: metis. It’s a Greek term for wisdom or cunning. More specifically: It’s intelligence learned from experience.
I found it in a book called Seeing Like a State, which explains why top-down management so often fails. Countries (and companies, too) need a standardized system to manage things efficiently, but that centralized view doesn’t (and can’t) completely map to the messy reality of the details on the ground. Local citizens and front-line employees develop an effective way of doing things, which management doesn’t understand … and then proceeds to ruin, in the name of “making it better.”
All of this comes back to context.
Metis is inherently local: built from one context, for that context.
The logic of the managers is the opposite: it’s abstract knowledge intended to be useful in all circumstances, with no particular context. In theory, it works everywhere, but in reality, it doesn’t really work anywhere.
The natural instinct for me and probably a lot of other people in the tech world is: “Metis is more effective? Great, let’s scale it up, so everyone can have it.” But that’s exactly the point: Metis is hyper-contextual. Flattening it out to make it more broadly available is removing the specific context that makes it work.
If you’re drawn to expanding or scaling what you do, consider: Will that undermine the best parts? Do you want what you’re doing to be as big as possible, or as good as possible?
Insight inspired by Man as a Rationalist Animal, an essay exploring James C. Scott and his book. Thanks to Dawson for the tip!
Bonus! It turns out metis connects a lot of dots I’ve been exploring, so here are some (unexpectedly) related posts, if you’re interested:
1, #181 - The Garden & The Fence: The danger of building without context.
1, #85 - Is quality possible at scale?: Can big be good?
1, #24 - The truth in cliches: “Through the doing comes the knowing.”
1, #101 - 2 ways of knowing: The difference between knowing facts vs. familiarity.
1, #132 - It feels obvious: Great design requires deep understanding of context.
1, #42 - Can information solve our problems?: Abstract info doesn’t solve things.
1, #182 - Too clear?: Organizational clarity rarely reflects reality.