The Starfish and the Spider
A business book examining how decentralized, leaderless organizations (starfish) can be more adaptable and resilient than hierarchical ones (spiders).
Get this: for the starfish to move, one of the arms must convince the other arms that it’s a good idea to do so. The arm starts moving, and then—in a process that no one fully understands—the other arms cooperate and move as well. The brain doesn’t “yea” or “nay” the decision. In truth, there isn’t even a brain to declare a “yea” or “nay.” The starfish doesn’t have a brain. There is no central command.
At Alcoholics Anonymous, no one’s in charge. And yet, at the same time, everyone’s in charge. It’s Nevins’s open system in action. The organization functions just like a starfish. You automatically become part of the leadership—an arm of the starfish, if you will—the moment you join. Thus, AA is constantly changing form as new members come in and others leave.
It’s not that open systems necessarily make better decisions. It’s just that they’re able to respond more quickly because each member has access to knowledge and the ability to make direct use of it. This brings us to the third principle of decentralization: an open system doesn’t have central intelligence; the intelligence is spread throughout the system. Information and knowledge naturally filter in at the edges, closer to where the action is.
The fourth principle of decentralization is that open systems can easily mutate.
the fifth principle of decentralization: the decentralized organization sneaks up on you. Because the decentralized organization mutates so quickly, it can also grow incredibly quickly.
put people into an open system and they’ll automatically want to contribute.
In open organizations, a catalyst is the person who initiates a circle and then fades away into the background.
Decentralization brings out creativity, but it also creates variance.
when measuring a decentralized network, it’s better, as the saying goes, to be vaguely right than precisely wrong. Even if we could, it wouldn’t really matter if we were able to get a precise count of how many members are in a network. What matters more is looking at circles. How active are they? How distributed is the network? Are circles independent? What kind of connections do they have between them?
How’s the circle’s health? Do members continue participating? Is the network growing? Is it spreading? Is it mutating? Is it becoming more or less decentralized?