Chainfeeds Summary:
An ideal state that approaches the "holy grail" in the long run is one where the block generation process is no longer concentrated on any single node, and complete blocks are not assembled from any single location.
Article source:
https://x.com/VitalikButerin/status/2007559523528233041
Article Author:
Vitalik Buterin
Opinion:
Vitalik Buterin: Now that ZK-EVM has entered the Alpha stage (performance has reached production-grade levels, with the remaining main work focused on security) and PeerDAS is live on the Ethereum mainnet, it's time to seriously discuss what the combination of these two means for Ethereum. This isn't just a minor improvement; it's pushing Ethereum towards a completely new and more powerful form of decentralized network. To understand this, let's review two major types of peer-to-peer networks in history: 1) BitTorrent (2000): extremely high total bandwidth, very high decentralization, but no consensus mechanism. 2) Bitcoin (2009): highly decentralized, with consensus, but extremely low bandwidth—because it's not distributed computing, but rather repeats the same work on every node. Now, with the arrival of PeerDAS (2025) and ZK-EVM (expected to begin small-scale use in the network in 2026), Ethereum, for the first time, simultaneously possesses: decentralization + consensus + high bandwidth. This means that the blockchain Blockchain Trilemma has been solved, not just in papers, but implemented through real-world code. Half of it (Data Availability Sampling, DAS) is already live on the mainnet; the other half (ZK-EVM) has reached production-ready performance, with only security requiring further refinement. This has been a 10-year technical journey: my earliest research submission on DAS dates back to here (link omitted), while attempts with ZK-EVM began around 2020. Now, this vision is finally beginning to materialize. Over the next four years, we can expect this blueprint to unfold gradually: 1) 2026: Due to the introduction of BALs and ePBS, the gas limit without relying on ZK-EVM will significantly increase, and the opportunity to run ZK-EVM nodes will also appear for the first time. 2) 2026–2028: Gas repricing, state structure adjustments, moving the execution payload into blobs, and other systemic modifications that make higher gas limits more secure. 3) 2027–2030: As ZK-EVM becomes the primary method for validating blocks, the gas limit will see a further, larger-scale increase. [Original text in English]
Content source 
