| timezone | UTC-7 |
|---|
- 自我介绍 大家好，我是一位UNC at Chapel Hill在讀的CS and Math學生。對加密學，區塊鏈，以及科技對社會的改變有興趣。希望可以通過這個共學，更好地瞭解以太坊。
- 你认为你会完成本次残酷学习吗？ 可以。
- 你的联系方式（推荐 Telegram） TG: @matseoi
Study Plan:
- Finish reading Vitalik's 6 passages on the future of ethreum protocol in the first half week 6.16~6.21
- Then, read the EPFsg Wiki and the Geth
Today, I read the first passages on the future of ETH protocol. The first passage is about the merge and the second passage is about the surge.
In the first passage, Vitalik discusses the pursuable future of more decentralized staking, reducing the time to finality and improving resilience against network attacks and failures.
To promote more decentralized staking, Vitalik proposes lowering the minimum validator stake from 32 ETH to 1 ETH, making solo staking more accessible and viable.
The goal of reducing finality time is to enable block finalization within each 12-second slot, rather than waiting approximately 15 minutes as is currently the case.
-
Brute-force signature aggregation – Use advanced schemes (e.g., ZK-SNARKs) to handle millions of validator signatures per slot.
-
Orbit committees – Select medium-sized committees that retain economic security while allowing low-overhead validation.
-
Two-tier staking – Have a high-deposit tier (for finality) and a lower-deposit tier (delegating or contributing in lighter ways).
-
Maintain status quo – Stay with current model, foregoing SSF and staking democratization.
A mix of these proposals are also proposed by Vitalik.
Single Secret Leader Election (SSLE) mechanisms is proposed to hide which validator will propose until the block appears. They use cryptographic shuffling or ring signatures but add complexity. Alternatives like network-layer defenses are also being considered
Note:
What is Single Slot Finality?
While finality is a crypto-economic terms pointing to the iead that the blockchain is economically costly to alter, SSF refers to that blocks getting proposed and finalized in the same slot.
The Surge refers to a phase in Ethereum's development roadmap focused on scaling the network to handle a massive number of transactions per second (TPS), while maintaining decentralization and security.
Vitalik also discusses the scalability trilemma, which highlights the difficulty of simultaneously achieving decentralization, scalability, and security. This was long considered a hard tradeoff, but recent innovations like SNARKs and data availability sampling (DAS) have made significant progress in addressing it.
In summary, DAS allows nodes to verify that data is available without downloading all of it, data compression reduces the space each transaction takes on-chain, and Plasma improves scalability by posting only Merkle roots of its state to Ethereum, greatly reducing data requirements.
Also, Ethereum wants the L2 ecosystem to feel like one unified chain, not many separate blockchains. Solutions include chain-specific addresses, shared bridges, and standardized cross-chain transactions.
Note:
State channel: A small group of participants stake ETH on smart contract and only the final state is being recorded on level 1 chain, allowing extreme low fee but fixed within a fixed group of people and limited flexibility.
Plasma: running a separate blockchain (a “child chain”) and post only periodic summaries of its state to Ethereum.
Rollups: bundling large numbers of transactions and submit either fraud proofs or validity proofs to Ethereumsupport complex smart contracts and enable a wide range of decentralized applications.
This passage discusses the current centralization risks in Ethereum caused by PoS and the current block construction mechanism. Specifically, smaller stakers tend to join larger staking pools to maximize MEV (Maximal Extractable Value), which increases efficiency but risks centralization. If two dominant builders control around 80% of blocks and decide to censor transactions, inclusion could be delayed by an average of 9 slots. which would allow market manipulation.
- FOCIL + APS:
- Inclusion lists ensure key transactions can’t be censored.
- Attester-Proposer Separation (APS) auctions off block production while proposers ensure inclusion.
- BRAID (Multiple Concurrent Proposers):
- Distributes the block building task to multiple proposers.
- More decentralized but needs encrypted mempools for privacy.
- Encrypted Mempools:
- Protect user transactions from front-running.
- Requires delay encryption or threshold decryption to ensure post-inclusion transparency.
In this passage, Vitalik discusses The Verge, which refers to making full verification of the Ethereum chain computationally affordable for marginal devices. Originally, The Verge focused on restructuring Ethereum's state representation into Verkle Trees, but it has since evolved into a broader goal: enabling efficient, low-cost verification for all users and devices.
There are two main approaches to achieving The Verge's goals: Verkle Trees and STARKed Binary Hash Trees. Verkle Trees are currently the most mature and time-efficient solution. However, they rely on elliptic curve cryptography, which is not considered post-quantum secure. On the other hand, STARKs are post-quantum safe but either suffer from longer proving times (if using conservative hash functions) or lack maturity and widespread trust (if using newer, aggressive hash functions).
| Approach | Proof Size | Security | Prover Time |
|---|---|---|---|
| Verkle Trees | Data + 0.1–2 MB | EC-based (not quantum-safe) | <1s |
| STARK w/ Conservative Hashes (SHA256, BLAKE) | Data + 100–300 kB | Trusted | >10s |
| STARK w/ Aggressive Hashes (Poseidon, Ajtai) | Data + 100–300 kB | Less mature | 1–2s |
Today, I started reading the Purge, which is about reducing data storage burden on clients whileing preserving ETH's permanence of on-chain data. However, I haven't finished reading this passage. Hence, I'll summarize tmr.
請假
In The Purge, Vitalik addresses the growing complexity and data storage burden on Ethereum. He proposes ways to preserve the permanence of critical on-chain data while significantly reducing its overall size.
First, he introduces history expiry, which limits long historical data to most nodes—typically about a year. A torrent-like network would be used for archiving, reducing the burden on each node while preserving data permanence.
Second, he explores state expiry, focusing on two main approaches. Partial state expiry would archive rarely accessed data, allowing it to be restored when needed. Address-period-based expiry would eliminate indefinite state growth by requiring periodic renewal of address state, though this would entail changes to the current address format.
Lastly, Vitalik also talks about the removing and reforming outdated and complex features including transitioning from RLP to SSZ for data serialization, removing old transaction types, reforming the log system, removing the beacon chain sync committee.
The Splurge is the part that is critical but hard to categorize upgrades. In the passage, Vitalik discusses the following updates.
First, since the current EVM is hard to analyze, inefficient, and inflexible. It would be a good idea to introduce a modular strucrture separating code from data, removing dynamic jumps, and enabling subroutines called EMV Object Format (EOF).
Second, since the current Ethereum only supports ECDSA signature-based accounts, account abstraction would make the net more flexible and secure.
Third, (I'll read the multidimensional gas part again tmr. Did not fully understand)
Lastly, the author discusses the ZK-SNARKs providing private and efficient proofs, FHE providing data processing without revealing, Indistinguishability Obfuscation encrypting programs where logic is hidden but still executable. In addition, in the future Quantum One-Short Signature may also be applied.
Kind of a easy day for me, just summarize the idea of multimensional gas. Currently, MG only supports execution and blobs. In the future, calldata, state reads/writes, and state size expansion will also be supported.