| timezone |
|---|
Asia/Shanghai |
请在上边的 timezone 添加你的当地时区，这会有助于你的打卡状态的自动化更新，如果没有添加，默认为北京时间 UTC+8 时区 时区请参考以下列表，请移除 # 以后的内容
timezone: Pacific/Honolulu # 夏威夷-阿留申标准时间 (UTC-10)
timezone: America/Anchorage # 阿拉斯加标准时间 (UTC-9)
timezone: America/Los_Angeles # 太平洋标准时间 (UTC-8)
timezone: America/Denver # 山地标准时间 (UTC-7)
timezone: America/Chicago # 中部标准时间 (UTC-6)
timezone: America/New_York # 东部标准时间 (UTC-5)
timezone: America/Halifax # 大西洋标准时间 (UTC-4)
timezone: America/St_Johns # 纽芬兰标准时间 (UTC-3:30)
timezone: America/Sao_Paulo # 巴西利亚时间 (UTC-3)
timezone: Atlantic/Azores # 亚速尔群岛时间 (UTC-1)
timezone: Europe/London # 格林威治标准时间 (UTC+0)
timezone: Europe/Berlin # 中欧标准时间 (UTC+1)
timezone: Europe/Helsinki # 东欧标准时间 (UTC+2)
timezone: Europe/Moscow # 莫斯科标准时间 (UTC+3)
timezone: Asia/Dubai # 海湾标准时间 (UTC+4)
timezone: Asia/Kolkata # 印度标准时间 (UTC+5:30)
timezone: Asia/Dhaka # 孟加拉国标准时间 (UTC+6)
timezone: Asia/Bangkok # 中南半岛时间 (UTC+7)
timezone: Asia/Shanghai # 中国标准时间 (UTC+8)
timezone: Asia/Tokyo # 日本标准时间 (UTC+9)
timezone: Australia/Sydney # 澳大利亚东部标准时间 (UTC+10)
timezone: Pacific/Auckland # 新西兰标准时间 (UTC+12)
YenChih Liao
-
自我介绍 web3 developer, 朋友推坑第一次嘗試參加
-
你认为你会完成本次残酷学习吗？ 50/50
-
你的 Telegram @YenCLooo
Finished week0 and week1
The week1 contains a lot of useful info, despite the vedio recording of somehow hard to focus on. The quiz is extremely helpful
Picked up/strengthened some key concepts
- Pretty Good Privacy (PGP) protocol, Signal protocol (both asymatric)
- Stacking offline panelty
- Proto-Danksharding and Danksharing
- validator --(Beacon APIs)-- Consensus layer --(Engine API)-- Execution layer --(JSON-RPC API)-- User/Web3
Finished week2 Pre-readings
Picked up/strengthened some key concepts
- Patricia Merkle Tries
- Archival nodes (13TB in Feb. 2023)
- Front running attack awareness
- MSTORE/LOAD: 3gas, SSTORE: 20k, 5k, -15k
Further study:
- Transaction Fee Mechanism Design (About EIP1559)
Finished week2
Picked up/strengthened some key concepts
- No more uncle blocks after the Merge
- The withdrawal functionality in the Beacon Chain allows stakers (validators) to receive their staking rewards (interest) automatically when blocks are finalized.
- p2p protocol broadcast tx to square root of peers
- snap sync protocol: contrast to full sync
Finished week3
Picked up/strengthened some key concepts
- Each "slot" has a block unless a validator failed/missed to propose
- Every blocks have a bunch of "attestation"
- Attestations of a block are included into the next block
- Every 32 slots forms a "Epoch" (for mainnet)
- Slashes and rewards are evaluated at the end of every epoch
- The full validator set is randomly shuffled and distributed to one of slots in a epoch
- Each validator only verifies once in a epoch
- Justification: 2/3 of validators attested
- Finialize: Justifying a justified block
Finished week4
It's talking about all the different ways/frameworks of testing. Without some concrete examples, it's really hard to get the merits of each of them
Picked up/strengthened some key concepts
- Execution Spec tests: Generate tests for EL clients
- Fuzzy Differential State Testing: Run a fuzzy smart contract and compares resulting state root of differenet clients
- Blockchain Testing: Similar to the previous test but on the scope of block hashes
- Hive Test: generate CL and EL for specific tests
- Shallow-fork: Early hard-fork from limited nodes
Finished week5
It's clear what is(will be) different with week5's lecture. Exciting.
Picked up/strengthened some key concepts
- Others
- Two types of rollups: Optimistic, and ZK
- Optimistic rollup: Optimistically trust all txs. Slash with fraud proof otherwise.
- ZK rollup: Succint proof verified on L1
- The Surge
- EIP-4844 empowered rollups: Utilize KZG commitment scheme. Verify by sampling partial data
- KZG is not quantum proof and needed a trusted setup
- CL has a verification function. EL has KZG commitment
- The Scourge: mitigate centralization concerns e.g. MEV, liquidity staking and pooling
- Proposer/Builder Separation: PBS
- The Verge: Simplify block verification
- Merkle to Verkle: Proofs now only need the path and the intermediate nodes. Siblings of the intermeidate nodes are no longer needed.
- Shorter proof
- Wider tree enabled: 16->256
- Stateless validators
- Merkle to Verkle: Proofs now only need the path and the intermediate nodes. Siblings of the intermeidate nodes are no longer needed.
- The Splurge: EOF, AA, better EIP-1559, encrypted mempool, etc.