Seasteads are man-made islands that float permanently on the ocean with any measure of a political autonomy. They would essentially be startup societies where people could form whatever kind of community they wanted.

Nearly half the earth's surface is unclaimed by any country, so seasteads would be startup countries on the blue frontier. Patri Friedman is a Google engineer and theorist of political economy who realized that if society floated, it would completely change the nature of governance itself. If seasteads are modular and can be moved about, allowing people to choose new societies, we'd create a market of governance providers, competing to attract residents.

Seasteads cost money, and if you want to succeed as a Seastead you have to find ways to attract people to move there. If I was a billionaire I wouldn't want to move to a seastead, but if I was a member of the bottom billion, most of whom want to leave their dysfunctional governments, I might want to move to a seastead.

A lot of Pacific island nations are sinking below sea level; they could easily transition slowly into becoming floating nations.

Thinking about seasteading requires us to free ourselves of these broad political categories we're stuck with on land. People can make whatever community system they want on a seastead. What emerges will totally defy the broad categories we debate about now.

Seasteads are a technology for anybody to form an alternative community based on their unique values - for communities to organize themselves however they want. Seasteads are their chance to demonstrate their vision can work. All that matters is that people can create, join, and leave seasteads voluntarily. As long as people can choose among seasteads, the best ways of living together will prosper, and the ones that people don't like will fail.

Evolution is variation and selection. If you can vary alternatives, and select among them, improvement emerges. It works in technology, in apps, and in life itself. What stunned me about seasteading is that it's a technology for variation and selection in governance itself. The reason some two hundred nation-states do a poor job of governing seven billion people is that they don't vary, and people don't select.