An article discussing the EigenLayer white paper: What are the potential problems?
Compilation of the original text: Deep Tide TechFlow
Compilation of the original text: Deep Tide TechFlow
Editor's note: The re-staking (Re-staking) scheme proposed by Eigenlayer allows users to pledge ETH, lsdETH and LP tokens on other public chains, oracles, middleware, etc., as nodes and get verification rewards, so that third-party projects can Borrowing the security of the ETH main network, the security of the ETH consensus layer is released.
EigenLayer is a cool concept. Crypto analyst kydo raised several questions/thoughts after reading its white paper, including non-native staking and its security, NFT positions, AVS activity, etc. The following is a detailed introduction.
1. Non-native restaking
Native restaking is changing `withdrawCredential` to EL's smart contract.
Native Restaking: Deposit your LSD, LP of LSD, or any kind of token to contribute to EL's economic security layer.
What is the security model for these different things? Given the composability of the underlying assets, how can security be proven?
2. NFT positions
Unlike Lido, when you deposit EL, you don't get FT (aka stETH), you get an NFT representing your position.
This makes sense because as a funder you may have a different appetite for risk and require operator nodes to engage in different services.
However, NFT will greatly reduce the liquidity of the entire system.
Will a Lido-like entity spawn on top of EL (maybe Lido itself?)? Or will a marketplace/aggregator be born on top of it and convert to ERC 20 by combining these NFTs together? What are the trade-offs of the different approaches for EL, the protocol, and Ethereum? Many questions remain to be answered.
3. Ensure activity
While slashing is a direct measure of security, how the system ensures liveness is less explored in the white paper.
You can use an Optimistic + bond + anti-fraud model, or you can actively track validators in EL smart contracts. However, the two systems operate on completely different assumptions and require completely different smart contract architectures.
What will EL do?
4. Cryptoeconomic security is sometimes a meme
Chainlink does not have security economics, it still keeps the entire DeFi safe. So, is it feasible to actively monitor economic security for a given AVS?
Furthermore, the amount of security guaranteed by a given AVS is not determined by the TVL it guarantees, but by how much it costs and the risk of running the service. Therefore, it will be interesting to see how EL alleviates these problems.
5. Decentralized sorter
The idea presented in the white paper is that when faced with an ordering problem, L2 can offload it to a set of EL validators running the ordering service. It sounds like using EL can bypass the "run PoS-L2 for sorting" problem, but can you really?
6. Event-driven MEV
Event-driven MEV allows a validator to make an agreement with a MEV party such as (the builder) that if a liquidation event occurs at the validator's proposed slot, it will provide a "strong guarantee" that the validator will include the MEV party's transaction .
This is in a sense very similar to a private relay, but there are differences.
What about doing this kind of behavior? Maybe it wants to get a fee/bribe upfront? Could ETH blockspace even become a subscription service?
7. Parallel Ethereum
I don't know what to call it, so I'll call it Parallel Ethereum for now. I may be confusing this concept, but it can refer to the following two situations:
1) Restaking => another L1, but with security from ETH tokens.
2) Change Casper FFG/LMD GHOST to something else.
1 is reasonable, but 2 is pretty funny, to say the least. My hunch is that 2 is impossible, since you'd still agree at the end to the chain finalized by Casper FFG, so your parachain would still have the same finality as the original ETH chain.
8. EL leads to centralization?
The argument for decentralization is: these AVS are cheap/light to run, so independent stakers can still run them compared to centralized solutions.
I think this is an optimistic future, if we look at the current bandwidth requirement of a validator running Mev-boost, it is already ridiculously high, around 10 TB/month.
So these add-ons, while not going to be a huge centralizing force, do little for decentralization.
9. Dual pledge utility
Simple question: how do you reconcile the weights of the two?
Are you using another EL AVS as a price oracle? Or use native AMM LP weights? Or do you hardcode and use governance to take a chance on a regular basis?
Original link
