Wen ETH 2.0? Doesn’t matter. Layer 2s don’t mind. They are busy working to help ETH achieve 100k transactions per second. As we wait for ETH 2.0, compatible Layer 1s and Layer 2s are competing for market share through usage and TVL. What’s their crypto purpose in the interim? The short answer is scalability, increasing the capacity of the ETH blockchain to handle transaction volume. StarkNet is a Layer 2 solution that’s been a quiet yet highly viable contender to make this happen. We’ll look into StarkWare, its parent company; StarkWare’s first product and original scalability solution, StarkEx; and the progression to StarkNet, the multi-app ETH Layer 2 solution enabling funds from a single wallet to interact across multiple apps on Layer 2. To start, let’s review the technical definitions surrounding zero-knowledge proofs (ZK), STARKs, Layer 2s, ZK-rollups, and permissionless decentralization; all of which enforce the maintenance and decentralization integrity of ETH mainnet. It’s bullish to note Vitalik feels ZK-rollups are the closest option to ETH 2.0 sharding. As you can see, Detective Decent is following the money and Vitalik on this one.
StarkWare develops STARK-based solutions on the blockchain. Capital is flowing into StarkWare following its $162M in total funding from their recent Series C raise of $50M at a $2B valuation. The founding team is Eli Ben-Sasson and Uri Kolodny, academic and entrepreneur respectively. Eli’s inspiration for StarkWare evolved from theoretical computer science and the desire to make proofs shorter and quicker. With the help of Bitcoin, Eli realized “blockchains are very good for proof systems—proof systems are very good for privacy and scalability.” This sentiment is much aligned with crypto and the decentralized finance ethos. What’s bullish about StarkWare is the transparency and exponential verification speedup it offers that previous ZK-based protocols failed to embody.
What are ZK proofs?
“Zero-Knowledge Protocol (or Zero-Knowledge Password Proof) is a way of doing authentication where no passwords are exchanged, which means they cannot be stolen. This is cool because it makes your communication so secure and protected that nobody else can find out what you’re communicating about or what files you are sharing with each other.” - Cossack Labs
We need to understand the premise of ZK because ZK-STARK is the base technology used by StarkNet. STARK’s competitive advantage is permissionless usage and trustless certification of proofs. Security is derived from its trustless nature, making it feel more decentralized than other Layer 1s. STARKs are important because they maintain computation integrity. Blockchains achieve computational integrity via trust in numbers, inclusive accountability (open sourced), and decentralized validators and checkers. Scalability works if the chain keeps the property of inclusive accountability—but this is challenging. The reason crypto can’t 10x block size or the gas limit is that inclusive accountability is hard to maintain while scaling quickly. There must be incentives. Proper scaling and inclusivity is the pulse of DeFi.
"ZK-STARKs (Zero-Knowledge Scalable Transparent Arguments of Knowledge) are a type of cryptographic proof technology that enables users to share validated data or perform computations with a third party without the data or computation being revealed to the third-party, also known as a zero-knowledge proof, in a way that is publicly verifiable. In simpler terms, a zero-knowledge proof can prove something is true without having to reveal what exactly it is proving. For example, ZK-STARKs would allow Alice to verify Bob's banking information using a zero-knowledge cryptographic proof instead of revealing the confidential information to Alice.” - [EthHub](https://docs.ethhub.io/ethereum-roadmap/layer-2-scaling/zk-starks/#:\~:text=ZK%2DSTARKs%20(Zero%2DKnowledge,proof%2C%20in%20a%20way%20that)
What are Layer 2s?
L2s are chains built on top of ETH mainnet that alleviate the mainnet load, essentially spreading out users and consequently lowering costs. StarkNet, Optimism, Arbitrum, and MATIC are the most well-known examples. Circling back to the first point, ETH’s problem right now is that the network is too popular and too busy, which means fees are too high and its throughput is easily fully tested. StarkWare uses ZK-rollups for their scaling solution; ZK-rollups have ZK-STARKs proof of knowledge.
As described on EthHub, “ZK-rollups are one of the options being developed for layer 2 construction that increases scalability through mass transfer processing rolled into a single transaction. Where Plasma creates one transaction per transfer, ZK-rollups bundle hundreds of transfers into a single transaction. The smart contract will deconstruct and verify all of the transfers held in a single transaction.”
StarkWare has been scaling with ZK-rollups either with Rollup Mode or Validium Mode. A good case study is dYdX, a decentralized perpetual trading platform using StarkEx Rollup mode for its scaling solution. Rollup mode implies the transaction and state changes roll up to ETH mainnet in a validity proof. This has saved 100x in gas fees. It would be too expensive computationally to run dYdX on an L1 because there are tradeoffs. L1s achieve immediate blockchain finality and security, but the user pays a higher cost. However, with StarkEx there are no changes in the L1, meaning it maintains inclusive accountability, but it temporarily loses time to finality with added data. Longer time to finality (minutes to hours) means lower cost and, therefore, higher scale. Security and cost comparative analyses are constant thought challenges in blockchain scaling.
Time to upgrade to StarkNet, a network. StarkNet was released on public test net June 21, 2021. Its Alpha launch was revealed on November 29. While StarkEx was like a single giant scaling computer for clients’ processing, StarkNet is like the internet or cloud or blockchain, with open-sourced deployability features. Eli Ben-Sasson says it’s like ETH itself with more scale. StarkNet takes the execution layer and compresses it, retaining the major benefit of StarkEx but sharing the execution layer. Unlike StarkEx, which catered to ubiquitous cases like massive payments, trading, and minting, StarkNet can cater to new general-purpose applications such as generative art, evolving, and metaverse gaming. We at Decent deem this bullish for Web3 and DeFi.
It’s funny that ETH 2.0 and StarkNet launch in stages and not feature complete—like a rocket building itself as it's launching into space. StarkNet’s development stages are usability, improved performance, and decentralization. Thankfully, usability is already complete with an ETH-like computation state having been delivered. This occurred quickly because StarkNet launched without an optimized sequencer. It was confident in the batcher from StarkEx as a central component of sequencing. After all, the batcher was battle-tested via dYdX and DiversiFi. Now to improve performance, StarkNet pivots its development efforts to transaction throughput, cost, and latency. But there are current bottlenecks, including necessary improvements for its sequencer node (StarkNet’s mining machine). Come mid-2022, we should see updates here in substantial speed and cost improvements. The prized jewel of decentralization occurs when StarkNet achieves leader election and governance mechanisms. It wants to decentralize the sequencer and prover, giving anyone access to participation. Full node deployment is underway by three teams: Erigon, Nethermind, and Equilibrium. StarkWare is even scaffolding out of its development involvement efforts, which is perhaps a signal of an incoming DAO. Overall, the transition to StarkNet focuses on the permissionless nature of independent smart contract deployment.
What’s the Decent take on StarkNet? We appreciate how focused StarkWare and StarkNet are on decentralization. It’s an open network that wants to open up more. StarkNet aligns with core DeFi principles as its development goals align with open-source and decentralized technology. StarkNet is a permissionless decentralized rollup operating as an L2, offering an unlimited dApp scaling solution for computation without compromising ETH fundamentals. Applications can be currently built and will hopefully eventually be built with utmost decentralization. StarkNet is open for smart contracts onboarding deployment now. As mentioned earlier, if StarkNet achieves ETH decentralization with its ability to create general-purpose dapps—all at low costs to the end-user—that will be the Giga bullish moment for EVM. We know true scalability means inclusive accountability with verification costs staying constant or growing slower than the cost of performing computation. Think of how dope the EVM ecosystem will be if we can retain permissionless verified transactions and incentivized inclusive accountability while spreading the computation. StarkNet is signaling that the ETH 2.0 transition can chill and take its time.