The number of UNIX installations has grown to 10, with more expected
- The UNIX Programmer’s Manual, 1972
Post-scarcity economics is a fun thought experiment: what would you do if everything you needed was available for free? As rich countries asymptotically approach that state (we’re all post-scarcity with respect to the material needs of a generation or two ago), one of the things we’ve discovered is: post-scarcity economies have tricky incentives. Take operating systems: if the OS is completely free, the result is a massive proliferation of options. Anyone technical enough to develop an opinion about the one right way to do Linux is probably technical enough to roll their own just-right distribution, and the result is countless hours of developer time spent re-implementing similar features across different distributions. A single main Linux would slightly frustrate lots of developers, at the cost of eliminating duplicated efforts.
This isn’t an unmitigated bad thing. It’s great for hobbyists, for example — I spent a fun six hours this weekend getting Arch working again after what was supposed to be a minor upgrade. But it means that a lot of effort gets frittered away in fairly unproductive pursuits. When the cost of exit is zero, voice and loyalty go unused.
And it’s also a useless thought experiment to talk about a single canonical Linux distribution. (The one made by Canonical Inc. appears to have a plurality of the market, though.) Nothing stops developers from building their own, so they generally will. As this excellent podcast points out, most protocols — and Unix is a bundle of protocols packaged as an operating system — were standardized in the 70s and 80s. New open protocols like RSS and OAuth mostly failed.
The protocols that didn’t fail were the ones that avoided openness and imposed artificial scarcity: Google uses pricing signals to control the supply of search clicks. Facebook makes it hard to reach someone with a post unless you’re either socially close to them or willing to pay to target them. And the net present value of Facebook’s mostly-private interactions is massively higher than the net present value of Twitter’s mostly-public ones. And there’s a big cost to building your own copy of Google or Facebook: the network effects are impossible to duplicate.
The tradeoff seems to be: a protocol owned by a monopolist can thrive, because the monopolist taxes the protocol at the Laffer maximum and invests the tax revenue in expanding its reach; open protocols are chronically under-resourced.
But crypto-currencies upend that relationship: they’re open — anyone can download the Bitcoin source code, change a few things, and launch their own coin. But Bitcoin holders can still get rich, so they have incentives to invest in the ecosystem. Early Linux and RSS adopters have bragging rights; early Bitcoin adopters have vacation homes they paid for in cash.
By introducing artificial scarcity into a post-scarcity system, token-based protocols recreate some of the incentives that closed protocols like Facebook and Twitter have, while still retaining the benefits of openness. This doesn’t mean Bitcoin is guaranteed to win: as the current block-size debate shows, Bitcoin is capitalism without constitutional governance. Projects like Tezos and Urbit add a layer of governance to the project, reducing the risk of breaking changes while still retaining scarce tokens.
One cynical theory about Bitcoin is that it’s basically a distributed Ponzi scheme: the limited supply of tokens gives people who buy into the system a reason to hype the system. That’s not completely damning, though. Modern banking is a benign descendant of John Law’s inflationary scams. As it turns out, there’s such a thing as a homeopathic dose of Ponzi, which gives people the right incentive to create enough wealth to make good on their own hype.
Disclosure: I am long Bitcoin, and long Urbit through ownership of an Urbit galaxy.