I really loved this Politico piece by Marc Andreessen from 2014:

But policymakers shouldn’t be trying to copy Silicon Valley. Instead, they should be figuring out what domain is (or could be) specific to their region—and then removing the regulatory hurdles for that particular domain. Because we don’t want 50 Silicon Valleys; we want 50 different variations of Silicon Valley, all unique from each other and all focusing on different domains. Imagine a Bitcoin Valley, for instance, where some country fully legalizes cryptocurrencies for all financial functions. Or a Drone Valley, where a particular region removes all legal barriers to flying unmanned aerial vehicles locally. A Driverless Car Valley in a city that allows experimentation with different autonomous car designs, redesigned roadways and safety laws. A Stem Cell Valley. And so on.

I immediately thought of it when reading about “A Silicon Valley for Drones, in North Dakota” from New York Times reporter Quentin Hardy. A fortuitous combination of things is going on there, including: a) the state has a low population density (47th out of 50 states), so if a drone falls from the sky it will probably just hit dirt; b) Grand Forks Air Force Base “flies nothing but robot aircraft for the United States military and Customs and Border Protection”; c) the state spent $34 million on a civilian industrial park for drones near the air base; d) the University of North Dakota, which already trains commercial pilots and air traffic controllers, has a drone controller program; e) there’s a surprising amount of tech talent thanks to business investments from Amazon and Microsoft; f) it’s a rural state. and “rural states with farming, oil and rail lines see many practical reasons to put robots in the sky.”