In our article discussing the road to Hybrid Rollup technology, we talked about the evolution of Ethereum’s scaling struggle. Within that discussion, we saw how an elegant and practical solution finally emerged: a layered blockchain design, with Layer 1 as the main chain dealing with only consensus and data availability, and Layer 2 as a Rollup to provide offloaded computation validation for all blockchain users, thus creating a scalable blockchain solution.
Ethereum’s Rollup scalability solution is so potent that it is starting to guide Ethereum’s evolution to a Rollup-centric future.
How would this Rollup-centric future shape the Ethereum ecosystem? If we compare this ecosystem to a living thing or living economy, then we need to investigate both the supply side (technology) and the demand side (application, use cases).
This article will serve as a warm-up before we further explore ZKM’s vision and solution.
It’s been close to a decade since Ethereum mined its first block. Almost immediately, the network struggled to scale. This struggle isn’t unique to Ethereum; it’s plagued many other blockchains, including Bitcoin. Ethereum co-founder Vitalik Buterin coined the term blockchain trilemma to describe this phenomenon.
The blockchain trilemma spelled out the difficulty of scaling a decentralized system. Decentralization and scalability are almost counterposing properties of a blockchain system such as Ethereum. This phenomenon lies in stark contrast to traditional distributed systems, where scaling the throughput of such a system is typically achieved by centralization and simply scaling the size of the system. Fortunately thanks to the unrelenting evolution of Ethereum, the end of the tunnel is in sight.
To break free from the trilemma, Ethereum began to evolve into a Rollup-centric, layered architecture: the Layer 1 main chain responsible for consensus (security in the trilemma) and data availability, and a Rollup-based Layer 2 responsible for execution (computation validation).
Since the trilemma describes a situation where Ethereum can only optimize at most two of the three properties, we need to pick one property to make sure its optimization won’t weaken the other two. We pick scalability.
First, we separate consensus and execution validation to Layer 2 through a mechanism called Rollup that maintains consensus on Layer 1 with data availability. This is Stage 1 of Ethereum scaling. With it, we can process more transactions, with cheaper gas costs. However, this is not enough. We need to scale Ethereum even further. That leads us to Stage 2.
We have already achieved the Stage 1 goal of Ethereum scaling, with numerous Layer 2 Rollups emerging in the past few years. However at this stage, Rollups have only achieved incremental throughput improvement within the confines of the base layer of Ethereum, by bundling multiple transactions into a single proof. Stage 1 Rollups have some limitations.
One notable Stage 1 Rollup limitation: withdrawing funds using an Optimistic Rollup requires a 7-day fraud proof window before the funds hit your wallet on Layer 1 Ethereum. We’ve had to endure this delay because at the time this was the only Rollup solution that could be implemented to be EVM-equivalent. Another Rollup solution, ZK Rollup, had struggled with implementing EVM-equivalency for a long time.
Fast-forward to today. ZKM’s Hybrid Rollup plans to solve most irking Stage 1 Rollup problems in an efficient and elegant way: keeping the best features of Optimistic Rollups (fast, cheap, and easy) in place, while providing ZK proofs to enable near-instant withdrawals back to Layer 1.
To further scale Ethereum, Stage 2 of the Ethereum scalability evolution will introduce several more innovations. One is sharding. However, the concept of sharding conflict so strongly with decentralization, it becomes hard to apply to blockchain system design without weakening it.
An innovation called erasure coding could fundamentally change the game. An erasure coding-enabled sharding solution allows blockchains to scale via Rollups without sacrificing security and decentralization — breaking free from the blockchain trilemma. The ramifications here are that for the first time in Ethereum history, by adopting erasure coding-based sharding for data availability, unlimited scalability could be achieved without the restriction of the blockchain trilemma.
To match DA (data availability) scalability on Layer 1, we could implement what we call functionality scaling on Layer 2 with Rollups, which is to let every dApp (Decentralized Application) operate its own Rollup operation (some call that parallel Rollup). We will discuss functionality scaling later.
Of course the most challenging part of Stage 2 is not the supply side but the demand side, if we use an economic analogy. The supply side is the scalability of the new Ethereum, but the demand side is more complicated. To facilitate a broader demand side for Ethereum, we must upgrade Stage 1 Rollups to a new breed called Hybrid Rollup.
Ethereum is the most proliferated blockchain technology right now. Bitcoin might have a bigger market valuation, but Ethereum reigns in terms of vibrant applications built on its network. However, the demand side of Ethereum (or blockchain technology as a whole) remains a touchy topic. Many lamented its lack of penetration in the mainstream applications market and the economy at large. To the blockchain novice, Ethereum seems to largely concentrate on fringe financial applications and is mostly driven by speculation rather than organic growth.
Interestingly the technological evolution of Ethereum feels like a hammer looking for nails, which contributes to the regrettable situation that this technology spends almost a decade still searching for a market fit. That is why you see cyclic hype that on one hand drove Ethereum to its top position among all blockchains, and on the other gave us an unclear end goal for Ethereum — ICO, DAO, DeFi, NFT, or whichever direction the speculative wind blows next year. Will Ethereum ever have broad, mass-adopted, real-world use?
Let’s do some soul searching. Remember when Ethereum was dubbed the World Computer by its founder? In the early days of Ethereum, the narrative was that as with Bitcoin, Ethereum was money — with added smart contract facilities to manipulate money, because Bitcoin’s smart contract capability was (and is) very limited. Bitcoin’s script system is not even Turing-complete, a drawback that eventually motivated Vitalik and friends to create Ethereum in the first place.
Smart contracts initiated the imaginary space for blockchain applications, and people can’t stop comparing blockchain applications to web applications. Smart contracts usher in a new kind of web application, hence the birth of the term Web3 (our ordinary web is thus called Web2). Ethereum would thus become the World Computer to bring us Web3. The scaling struggle of Ethereum is to fulfill the promise of being the world computer that can support an entire Web3 ecosystem, with thousands of or even millions of Web3 applications running on top of it. As things stand, this world computer is quite weak in comparison to the computing infrastructure of Web2.
When most people think of crypto and/or blockchain, they think of it as a new kind of money, or a new device of money (e.g., Internet of Money). Being associated with money is both a blessing and a curse. Not being able to dethrone Bitcoin as the de facto currency of the blockchain world, Ethereum pivoted to be the next best thing — a device to facilitate activities around this new kind of money. That’s how DeFi (decentralized finance) took off, coincidentally just as real money overflowing in the real world, thanks to the US government’s runaway money printing during the COVID-19 pandemic.
The wave of DeFi has since subsided, after the Feds slowed the money printer and started to raise interest rates, and crypto fell into a bear market. Beyond that downturn, DeFi may have also delayed hard-needed innovation. Hunger drives innovation, not wealth. Speculation as well as fraud abound when money is everywhere for easy picking.
Imagine what it would be like if Stage 2 of Ethereum scalability arrived for the DeFi boom. We know that the real world increasingly sees blockchain/crypto as a threat rather than a benevolent ally (let alone a savior, as many die-hard crypto libertarians believe). But even so, had DeFi been equipped with performant Stage 2 Ethereum scalability, more external money would depend on this superior DeFi technology, not the other way around. Now that Stage 2 Ethereum scalability is here, what would DeFi do with all that power? As a project vying to become a leader in the Stage 2 Rollup solution provider, ZKM has a big vision for Ethereum.
Ethereum has been explored for plenty of non-financial applications too: governance, social welfare, alt-economy, to name a few. These subjects are warrant their own discussions; we’ll come back to them as they are indeed very relevant. Here we are mainly interested in what these applications mean for the end goal of Ethereum.
Ethereum scalability is a non-starter in the traditional computing world. No matter how performant it could become, it will never match the performance of traditional distributed systems. The value of Ethereum and blockchain lies in their use case: they exist for computing in a totally different world — the decentralized world.
Both DeFi and non-financial applications deal with decentralized social constructs. Decentralized social constructs are where innovation should be. The end goal for Ethereum should be servicing those truly decentralized use cases, instead of competing with traditional social constructs — centralized money and centralized social institutions. Serving decentralized social constructs is the reason we took the hard way to solve the Ethereum scalability problem.
So now we can answer the question: What is the end goal of Ethereum? We want Ethereum to be the base layer to provide a solid foundation for all kinds of decentralized social constructs, no matter if they are on Ethereum or any other blockchain.
We call this base layer the Universal Settlement Layer. And that’s where the next stage of our journey begins. Join us again soon in this space and we’ll explore.
— -
This article’s author, Ming Guo, is Chief Scientist for ZKM.
Have you read the ZKM whitepaper yet? Check it out at https://ethresear.ch/t/zkmips-a-zero-knowledge-zk-vm-based-on-mips-architecture/16156
Want to discuss this and other ZK-related topics with our core ZKM team? Join our Discord channel: discord.gg/Bck7hdGcns
