Mike Hearn is one of the primary engineers behind bitcoin, the digital currency that aims to remake our financial system. Or at least he used to be. Over the weekend, Hearn told the world that he and a group of other coders were forking the bitcoin project, creating an alternate version of the software that underpins the digital currency. “I feel sad that it’s come to this, but there is no other way,” Hearn wrote in an email to others working on the project. “The Bitcoin Core project has drifted so far from the principles myself and many others feel are important that a fork is the only way to fix things.” Bitcoin, you see, runs on open source software, software that’s freely available to the world at large, software that anyone can not only use but change. A fork is the ultimate change, where someone makes a copy of the code and creates an entirely separate version of the project. This is uite common in the world of open source software—a world that now drives so much of the technology we use every day. Open source code is built on consensus, and sometimes, if coders feel a consensus can’t be reached, they fork. The dispute could ultimately decide whether bitcoin succeeds or fails—especially if the project remains permanently split. Hearn’s fork could significantly change the direction of bitcoin. Known as Bitcoin XT, the fork is backed by another notable bitcoin engineer, Gavin Andresen, previously lead maintainer of the bitcoin project, and basically, it aims to modify the bitcoin software so that it can rapidly handle a much larger number of online transactions. This could be essential as bitcoin expands to an ever larger number of people and organi ations—and as its underlying tech moves into stock trading and other financial markets. But other coders on the original bitcoin project, a.k.a. Bitcoin Core, believe there are better ways of accommodating more transactions. The dispute may ultimately decide whether the project succeeds or fails—especially if all this ends with no real consensus and the bitcoin project remains permanently split, the sum of the parts not uite as powerful as the whole. At the same time, Bitcoin XT provides an unusually clear window onto the world of open source, an extreme example that demonstrates why, despite or even because of the current strife, this idea is so effective—why it’s so uickly changing the way our world works. Bitcoin XT exposes the extremely social—extremely democratic—underpinnings of the open source idea, an approach that makes open source so much more powerful than technology controlled by any one person or organi ation. A fork may seem like a divisive thing. And, at times, it is. But it can also provide a way of building consensus, a way of forcing the broader community to get behind one idea or another. Ultimately, that’s what open source is all about: building things with as many minds as possible. Yes, that old saw so often holds true. “Two minds are better one.” But that’s not all. If more people are building and using a project, it evolves faster. It’s capable of more. A bitcoin uses by many people is more powerful than a bitcoin used by few. Sure, some people fork a project just because they want to tinker with the code or create something entirely new. But it’s also a way of resolving disputes within an existing community of coders. The majority may stay with the original project. It may move to the fork. Or it might force the two projects to resolve their differences and recombine. The fork forces an implicit vote on the ideas at hand. Witness io.js, a fork of the popular node.je project. It bought detente to a community split by disagreement. “A fork can be a market mechanism,” says Jim Zemlin, the director of the Linux Foundation, the not-for-profit that oversees the Linux open source operating system and many other important open source projects. “It’s a forcing function to reach consensus.” Bitcoin XT is a prime example—though it goes even further than the norm. The difference is that bitcoin isn’t just a piece of software code. It’s a piece of software code that runs a worldwide network of otherwise independent machines. Working in concert, these machines drive the bitcoin digital currency. They provide a way of storing money, sending it, and receiving it. So, in forking the core bitcoin code, Hearn and Andresen are also moving to fork the bitcoin network. And that’s where things get interesting. Basically, Hearn and Andresen want to increase the si e of the “blocks” that identify transactions on the network. Right now, the blocks are one megabyte in si e; they want to raise this to eight megabytes. If you increase the block si e, you can increase the number of transactions the network can process each second. Some think there are better options than raising the block si e—“many people think this path is fraught with too many security vulnerabilities,” says longtime bitcoin coder Eric Lombro o—but Hearn and Andresen disagree. But they haven’t increased the block si e just yet. They’ve said they will implement the increase only if 75 percent of the machines on the network adopt their fork. They haven’t just arranged for an implicit vote. They’ve established a very direct referendum. It’s open source in the extreme. The particulars of this dispute may seem trivial. But the nature of bitcoin also ensures that those involved in the project will take all this deeply seriously. After all, money is at stake: the machines on the network store and even generate money. If the 75 percent majority is reached, the other 25 percent may even have good reason to adopt the change too. Otherwise, the value of their money—and their infrastructure—will change. “In that way, it’s uite different from other projects because there’s that common economic interest,” says Angel Leon, a developer who works on the bitcoin software. “The genius of the blockchain consensus doesn’t just protect the blockchain from malicious attacks. It also ensures a democratic process in the evolution of the protocol.”