10/07/2018

I.

In contrast to centered (even polycentric) systems with hierarchical modes of communication and pre-established paths, the rhizome is an acentered, nonhierarchical, nonsignifying system without a General and without an organizing memory or central automaton, defined solely by a circulation of states.

— Gilles Deleuze and Félix Guattari, A Thousand Plateaus

If you've been living anywhere other than under a rock for the last few years, you already know that there has been a resurgence of interest in decentralized systems. There's the super well-known ones like Bitcoin and Ethereum of course, and also a myriad of other technologies and ecosystems: Mastodon, ActivityPub, Perkeep, Scuttlebutt, Dweb, Solid... the list goes on.

Apparently, the answer to life, the universe, and everything is to put a blockchain on it. (Optional extra steps: publish a buzzword-soup whitepaper, ICO for a bajillion dollars, and develop a grand total of no new products.) I'm being snarky, but it isn't because I don't believe in the idea of decentralized systems.

The classic rebuttal against blockchains and other distributed technologies goes something like this: there's nothing a blockchain can do that a plain old SQL database can't do cheaper and more efficiently. That's true, but it's also missing the point; it ignores all of the questions around things like data governance and platform power that the new decentralized technologies raise.

Instead, I want to attack this from another angle. I want to talk about how the hype around decentralized technologies represents the latest in a long line of technological determinism, a kind of belief that says, as William Gibson puts it, "all cultural change is essentially technology-driven". In other words, I want to argue that technological solutions are never sufficient.

I try to approach this from two different angles. I'll first build upon a few takes around the idea of uneven development, and then talk a bit about the competitive tendencies of decentralized systems using Scott Alexander's idea of Moloch.

II.

The centralization of capital allows for a more rapid expansion in the scale of production (and potentially therefore to a more rapid increase in the productiveness of labor) than could be achieved by simple concentration of capital in existing units.

— Neil Smith, Uneven Development

I'm slightly misappropriating the term from scholars like Neil Smith, but I basically want to use the term uneven development to describe a pattern like this: in any competitive scenario involving multiple agents (individuals, tribes, corporations, whatever), all agents are not created equal. There tends to be a natural unevenness in the abilities and fitness of agents. But over time, as the fitter agents capitalize on their advantage, this unevenness compounds, and power consolidates. Over time, a power law distribution asserts itself.

This already applies itself in the technologies that we already use day-to-day. A lot of people have already observed that most of the technologies we use started off as decentralized systems — the main examples being things like HTTP and email and Git. It's just that in each of these cases, most of the users end up gravitating towards one or few centralized systems. For HTTP, it's Google and Facebook. For email, it's Gmail. For Git, it's GitHub and GitLab.

Furthermore, even newer decentralized technologies don't really negate this. Proof-of-work blockchains merely delay it, and proof-of-stake blockchains even embrace it.

I think coming to terms with this uneven development leads to two possible conclusions. If uneven development is seen as a bad thing, then a decentralized system is by itself insufficient for maintaining equality. And if uneven development is seen as a good (or benign) thing, doesn't that mean the decentralized system was never needed in the first place?

III.

Each step of the Poor Coordination Polka makes your life worse.