Author: terence.eth, Arbitrum developer; Translation: Golden Finance xiaozou
In this article, I will summarize the live broadcast of Arbitrum’s permissionless verification protocol BOLD by Ed Felten, co-founder of Arbitrum development company OffChain Labs, and Raul Jordan, co-founder of Prysmatic Labs, on the Bankless podcast, because it was really wonderful.
First, it’s an exciting time right now as we build on L2 (particularly Arbitrum) with Stylus, Orbit, and BOLD. Meanwhile, Offchain Labs is also making significant progress on the L2 decentralization frontier.
Arbitrum is already the most decentralized L2, and BOLD is the next step in decentralization, enabling fully permissionless verification. From a security perspective, anyone can force the chain to achieve the correct result, even if everyone else is malicious.
Decentralization has been criticized in the past for being a very subjective term. L2beat provides a great perspective for objective measurement, such as the different stages of decentralization. The pie chart risk makes it easy to spot which parts of the L2 chain are more decentralized than others.
The live broadcast provides a comprehensive introduction to all aspects of the pie chart.
- DA is shown in green because it uses Ethereum.
- Proposer and sorter failures are also green due to design decisions, e.g. the sorter cannot enforce invalid state transitions. It can only front-run you.
- Upgradeability is in yellow for security reasons, but one could argue that given the early stages this is a feature, not a bug.
- Finally, state validation is indicated in yellow because the current validator set is permissioned. If the DAO chooses to accept, BOLD state validation will be green.
The livestream details the BOLD protocol, with Raul giving his thoughts on Ethereum scaling and why teams are moving from sidechains to L2 given the ultimate goal of reducing trust as much as possible. Ultimately, what users want is a secure validation bridge to support activity on L2, and fraud proofs are key to achieving this. BOLD makes it more secure by addressing latency attacks.
Before the release of BOLD: The Game style, to win the game you had to defeat all of your malicious opponents one by one.
After BOLD is released: A large-scale battle scene, you have the superpower to knock down all malicious opponents at once.
To participate in the BOLD challenge, validators need to run the same node software as today. There are roles such as active validators and defensive validators. The software automatically detects invalid state transitions and helps to make correct settlements on the chain.
When a fork occurs, every honest validator will know that your fork is correct and use that fork. Disagreements will happen behind the scenes and will eventually be resolved.
Finally, why would anyone want to run it? First, if you are already running an L2 node, then it is cheap. Second, if you run a dapp on L2, you may be taking risks. The live broadcast also talked about sorters and staking, which I will not introduce one by one here.
