Do financial incentives increase the threat of forks?
July 2, 2018
I’ve been trying to understand why the threat of forks is so frequently discussed by cryptocurrency projects, compared to open source at large.
The most common explanation I’ve heard is that there’s a financial incentive to fork in crypto that doesn’t exist for open source at large. This is true for both the developers, who could fork the source code and raise money on someone else’s idea, and for the miners, who stand to receive sizeable financial rewards for mining new blocks. To make things even more complicated, forking a cryptocurrency project implies not just forking the code, but the ledger, coin, community (including critical functions like miners), and third-party partner support (such as wallet or exchange listings).
I’ve been mulling on whether this is true, and if it is, what it might mean about how incentives change in open source when people suddenly stand to make a ton of money off of it.
Andrey Petrov and I discussed it in an email exchange. I liked his response so much that I’m sharing it with his permission below.
Forks just aren’t necessary to raise cheap money, they just put up a fancy landing page with a dozen impossible promises and off they go.
The forks that do exist are mostly based off things that have been around for a long while, like Bitcoin which has been around for >3 years before the first wave of forks (Namecoin/Litecoin/Peercoin/etc). Most recently projects are just ripping off Ethereum, but Ethereum has been live for several years and was developed in the open from the beginning (almost 5 years ago). Why fork off something that is just a few months old and unproven when there are mature building blocks around, right?
I feel the biggest defensibility in the crypto community turns out to be momentum (which imo encompasses community as well as other things). A fork basically starts from a stationary position and it takes herculean effort to reach any kind of relevant momentum. Being committed to a large codebase is like mass, it takes proportional effort to ramp up or shift directions. If all they have is some marketing props with no mass behind it, they can turn on a dime.
If the original fork stagnates (loses momentum) and whatever competing fork overtakes it, then that’s the natural course of things. I think that is the real thing people are afraid of, but it’s really the best case outcome for everyone. (Bitcoin Core vs Bitcoin Cash might be an interesting case study someday, even if it doesn’t overtake it.)
Realistically, from an open source perspective, anyone who has the motivation to start a genuine competing fork is signing up for a lifetime of thankless maintenance work.
So here’s my theory. I think crypto is concerned about forks, not because financial incentives actually change the game, but simply because it’s early days.
In a time of rapid speciation, it makes sense to be concerned about forks. But over time, we’ll see things settle into fewer, more stable projects with defensible brands, chains, communities, and developer-miner partnerships.
As evidence of this, one of the clear cultural divides I’ve noticed between early and contemporary open source is their interest in forks. Early open source developers talk about forks a lot. Contemporary open source developers, less so. (If you’re interested, Mikeal Rogers and I once discussed this with Karl Fogel, especially with regard to network effects and governance: episode 1, episode 2.)
I’m not sure exactly why this is the case, but my guess is it’s partly because the bigger project communities have stabilized, and partly because smaller software projects are modularized, with more substitutes than ever before, because so many more developers are building and releasing similar ideas in parallel. The right to fork is still critical as an implied threat to keep the system in check. But major political forks aren’t as common, and certainly not as top-of-mind, anymore.
I’m guessing there are other behaviors that we attribute to being unusually characteristic of crypto, which are actually just signs of an early-stage open source ecosystem.
(Note: I think this is also true for startups. We saw an explosion of new startups created in the first few decades, so we assumed that anybody being able to build The Next Big Thing was an innate characteristic of the tech ecosystem. Today, technology has consolidated into a few big players, and it is very hard for even a large, well-funded company to escape the jaws of acquisition. Will a decentralized ecosystem be different? I’m not sure. Networks benefit from monopolies.)