Season One: Digital Physics
Letter from the Editor: vera
A few months ago, before joining Lattice, I joked to ludens that the team should publish an article called Why Your Protocol Needs Physics. Part philosophical treatise, part sales-pitch (and perhaps, part-shitpost) the piece would espouse why automated market-makers, decentralized lending markets, and L1’s and L2’s should make their protocols physics-aware by implementing notions of spatial coordinates, the speed of light, energy conservation, and other tenets of physics into their base smart contracts and architecture.
The comment was inspired by a remark ludens made to me (and has also expressed on podcasts and in talks) about the computational limits of blockchains today, and their irreconcilability with universal physics. Ethereum has a single dimension: time, and as a result, EVM-compatible chains process transactions sequentially. The universe has four dimensions: three-dimensional Euclidean space (x,y,z), and time, which means events happen in parallel. Blockchains events are ordered, and the rate of information propagation is determined by the rate at which validators agree on ordering. Universe events are subject to relativity and the rate of information propagation is bounded only by the speed of light.
Could we emulate this phenomenon of universal physics and relativity by “giving” contracts coordinates onchain, and limiting interaction with specific contracts to conscribed locations? This would eliminate notions of global state (because information travels at the speed of light), and essentially parallelize the EVM by cheating how state propagates in the network. If Alice trades with Bob in one region of the universe, it has no impact on whether Charlie can trade with Dave a million miles away. The vanilla EVM needs to order this transaction regardless, but physics-enforced smart contracts would not, because they would have a notion of spatial coordinates. At the network level, you’d no longer need global block producers, just local ones that eventually reconcile transactions on a global scale.
Other phenomena could be unlocked by spatial coordinates as well – not just hacky implementations of EVM parallelization. Such an example can be found in designs for zkDungeon, the game that predated MUD, Lattice’s operating system for onchain applications (as well as OPCraft and Sky Strife, our first two games built on MUD). zkDungeon was a cross between a board game and an onchain battle royale, where players could build on and mine territory on a map, summon creatures, and trade resources like gold and souls.
Like the hypothetical EVM described above, contracts had a defined coordinate on the map. Unlike the hypothetical EVM, the coordinates weren’t enforced for parallelization – they were there to encourage emergent behavior, like players building trade routes, maritime republics, and special economic zones, all of which would emerge from the 'physical', locational constraints of AMMs. With contract-locality, we could insert market inefficiencies into the game, and incentivize new enterprising player behavior to overcome them.
Something simple – like defining contract coordinates in metric space – can have dramatic consequences, impacting everything from creating new kinds of mercenary player behavior, to helping the EVM transition from the serial computer it is today, to a more performant machine. We’ll call these simple stipulations “digital physics.” I like to think of digital physics as fundamental laws of an onchain system that have the potential for resonance across the entire stack – from the application layer to the infrastructure layer.
There are other examples of digital physics in well-known onchain games. In Dark Forest, players had options for how to uncover the game’s fog-of-war map. They could use Dark Forest’s standard in-browser single-threaded Javascript miner to compute hashes that would reveal the map. Or they could devote more CPU cores to the effort, by using a simple drop-down module. There were also custom miners written in Rust which could uncover the map even faster, or if players wanted to outsource the mining effort entirely, they could use a remote miner that operated on a server or Raspberry Pi. One player drew attention for implementing this on a 96-core AWS instance (which probably cost somewhere in the range of $3.00 an hour).
Such behavior might seem irrational if you forget about other plugins that players built for Dark Forest. As a game that hosted dozens of permissionless plugins, there were also markets where players could buy and sell artifacts, planets, or even coordinates of the planets themselves (in a fog-of-war world with incomplete information, the information itself can become the most valuable commodity). Suddenly, mining the Dark Forest universe is perfectly rational economic behavior – akin to mining for any kind of valuable resource in the physical world.
What you choose to accept as a valid input has dramatic impacts, and directly guides a system’s digital physics. Imagine Dark Forest with a static rate of map exploration, with no way for players to customize how quickly they wanted to uncover the map. This would make the total universe size in Dark Forest a function of cumulative player count and time spent playing, rather than a function of those two factors, plus cumulative resources expended mining. The game between players would be simpler: the strongest players would be ones that spent more time in-game, or spent more real-world money to buy map coordinates. The version of Dark Forest that actually exists permits for a third variable, based on the compute resources a player is willing to expend to uncover the universe. In other words, by having hash rate as an input, users were given more control over how large they wanted the universe to become, and increased the likelihood for more dynamic behavior down the line.
Autonomous worlds are the ideal petri dishes for digital physics, because they allow players to create consequences and points of absorption for a game’s emergent systems. There are no guidelines on what “strong” digital physics looks like – this will depend on the onchain world you are devising. Not every world needs a limit on the actions that can be performed within the confines of a grid, or a universe that expands at the same pace as your hashrate. What matters the most with digital physics is the resonance it can create.
At Lattice, we’ve been exploring these concepts, and want to share what we’ve discovered. Today, we’re announcing the launch of The New World, a series on Mirror that we’ll be using to probe into onchain life. For the next few weeks, during Season One, we’ll be exploring Digital Physics in all of its forms. We’ll study tick rates and the notion of onchain time. We will investigate anthropocentric digital physics, and the kind of digital physics that might be more hospitable to AI’s and aliens. We will itemize current EVM-compatible verbs (spoiler: there’s only one), and propose ways for Ethereum to support more verbs (spoiler: we can do this using MUD). Along the way, we hope to finesse our understanding of digital physics, and how to build autonomous worlds with the same degree of resilient, enduring, immutable physics that exists in the universe.
We believe that autonomous worlds are emerging from primordial state. And like the universe we inhabit, they will require a deep investigation that’s complementary to product-level experimentation and technical documentation. We would like to catalog the thoughts, intuitions, mistakes, and wisdom we’ve acquired while building autonomous worlds, to make the terrain more accessible for anyone exploring the space with us.
If this sounds interesting to you, stay tuned and subscribe to our Mirror. If you’d like to contribute, email me at vera@lattice.xyz – we’ll be accepting submissions from developers and researchers who are interested in these questions. To get you started, we’ve brainstormed a list of words and concepts that might be helpful as you explore.
Digital Physics: Conservation Of Energy, Lindy, Modifying The Execution Layer, Stephen Wolfram’s A New Kind of Science, Jump Crypto Is Battling The Speed Of Light, Hidden Information Games, Onchain Cellular Automata, Entropy vs Persistence, The Anthropic Principle, MEV Is Digital Physics, Quantum Mechanics, Things That Should Never Be Rewritten In Rust, Concrete, Your Blockchain’s State Transition Function Is Its Original Sin, Giving Ethereum Eyes, Smart Contracts Are Fascist
To learn more about Lattice and MUD, visit our website, read our documentation, or join our Discord server. *
*Thank you to ludens and biscaryn for feedback and edits on this post