The bitcoin meta-problem

Today is the 14th anniversary of the creation of bitcoin’s genesis block. As I reflect on bitcoin’s progress over the past 14 years — which is by all measures amazing and something the contributor community can be deeply proud of — my mind soon drifts to the topic of what we have left to do. Time is limited, material resources are limited, and there’s a real possibility that bitcoin ossifies in the not-so-distant future and no more consensus changes can be made. I wonder: what is the most important unsolved problem in bitcoin?

My gut answer is “privacy”. Whenever I’m teaching newcoiners, I’m always bothered by how many caveats and instructions are required to teach them how to gain some semblance of privacy when using bitcoin. And even after that, I come away from the conversation both uncertain that the tools recommended will actually protect the newcoiner’s privacy in the long-run, and also quite certain that even if the tools did work, the newcoiner will make a mistake at some point that will un-do the work they had done to protect their privacy up until that point. This is unacceptable and untenable. We need a solution that people can use to store and transfer coins in a self-custodial, p2p manner that offers privacy and ease-of-use as close to physical cash as possible. Bitcoin is too risky and inconvenient for most people to use otherwise.

Then I think, ok, if we somehow solved that, then what’s the next most important problem to solve? And the one after that? And after that? And so on. There will always be some new problem to solve, because human desire and creativity is neverending. And I realize that actually, perhaps the most important problem to solve is giving bitcoin developers the ability to easily solve ~any problem within the domain of electronic cash. I’m calling this “the bitcoin meta-problem”.

Currently, to solve some problem that users have, a bitcoin developer must either work within the limited confines of bitcoin’s existing scripting capabilities, or else propose consensus changes that require buy-in from the majority of miners and economic full nodes. If the developer can’t or doesn’t want to do either of those, then they have to convince the user to move their BTC into some (centralized/federated/collateralized) custodial system, sacrificing to some degree the core value propositions of bitcoin for some new functionality.

What is needed is a solution (a “meta-solution” to the “meta-problem”) that enables a bitcoin developer to build new functionality that solves their users’ problem without requiring the developer to either make consensus changes to bitcoin or convince their users to give up self-custody.

One meta-solution would be to directly add a fully expressive smart contract language to bitcoin, capable of supporting any kind of logic that developers would need to solve any kind of user problem. This is the approach pioneered by Ethereum. Bitcoin developers have so far rejected this approach because of the relatively heavy computational burden that it would place on bitcoin full nodes, and this consensus is unlikely to change. So we need to add one more requirement: the meta-solution can’t impose heavy computational costs on bitcoin full nodes.

Another meta-solution would be to add support for validity rollups to bitcoin. Validity rollups can enable developers to deploy new functionality without having to make consensus changes to bitcoin, convince users to give up self-custody, or put a heavy computational load on bitcoin full nodes. To support any kind of new functionality (new privacy protocols, new smart contracts, etc), bitcoin full nodes only have to verify a single cryptographic proof per rollup block.

Based on my research, I consider validity rollups to be the best meta-solution that has been invented so far. Validity rollups might even be the best meta-solution possible as a protocol that can check all of the relevant boxes: self-custodial (trustless), infinitely extensible, and easy for bitcoin full nodes to verify and stay in sync.

I look forward to seeing what progress we make toward solving the bitcoin meta-problem this year, 14 Anno Satoshi