The Scroll: UTXO Sharing
In this new recurring newsletter, The Scroll, we’ll be highlighting the highly technical work being done by our Bitcoin Wizards team, who, as it happens, are in the market for a third wizard.
If you’ve got your finger on bitcoin’s pulse, then you know that people have been talking about doing upgrades to the Bitcoin Scripting system to enable scaling beyond the three things covered here. This is an introduction to those that we’ll be elaborating on in the future, such as the OP_NEXT conference.
During the Blocksize Wars of years past, proposals to “scale bitcoin” typically fell into one of a few distinct, actionable categories:
Increase the block size: Heavier resource requirements allow more UTXOs to be consumed and created over a given period.
Use Payment Channels: If there are repeated economic engagements between actors, payment channel networks such as the Lightning Network seem to make incredible efficiency gains, where a UTXO corresponds to a person rather than a transaction.
Shrink transactions: Increases support for higher average transactions per second.
Throughout most of the bitcoin community, solution one has technically been deemed a dead-end. Solution 2’s core limitation is that even if we assume that LN participants never go to a chain, only a small fraction of the world’s population can reasonably create channels with the block space we have. Solution 3, by definition, cannot effectively increase by more than a modest multiple while remaining trustless. That leaves all three being at least worthy of consideration.
That said, how else can we scale UTXO ownership?
One idea is that in order to scale UTXO ownership, we can enable what is called “UTXO sharing.” Imagine this strawman example:
The entire world is onboarded onto bitcoin. There is a single UTXO on the network, where the true distribution of bitcoin lies under the hood through a series of pre-signed transactions. These pre-signed transactions construct a binary tree, ending in “virtual” UTXOs, each preserving the global population’s balance of wealth.
This idea is not impossible. The tree would only be 32 layers deep! All it would take is getting 8 billion people together online at the same time to send a few hundred billion messages and, uh, yeah.
As this strawman makes clear, *interactivity* is one key part where these naive schemes fall apart and where most of the work is on making massive UTXO sharing schemes practical. What UTXO sharing schemes are practical with small N’s today? What numbers could become practical with your favorite consensus change?
In this introductory post, we won’t be answering these questions, but as a preview of future posts, we can broach how people have attempted to overcome interactivity:
Covenants: This general bitcoin term tends to mean “more expressive bitcoin script,” including generalized transaction introspection.
Coordinators: Systems with coordinators can reduce the interactivity problem, allowing them to function as long as the high-requirements coordinator synchronizes pairwise with all clients sharing a specific UTXO.
Liquidity: With additional bitcoin liquidity, we can gain blockspace efficiencies.
In later posts, we will explore these research and engineering directions and the ultimate limits of these concepts.
Adversity is the mother of innovation.