Vitalik Buterin's latest brain hole: a scalable data blockchain model without committees

On January 6, Ethereum co-founder Vitalik Buterin hosted a discussion at the Ethereum Research Forum. In the discussion, with proof of fraud-free data availability proof, he outlined a scalable data blockchain model that does not require a commission. According to Buterin, this fraud-proof data availability check scheme retains more of the attributes of traditional non-sharded blockchains.


Image source: pexels

He says:

"If the client accepts a block at time T, it will continue to be accepted at any time after T, so a fraudulent proof will not invalidate this fact after it occurs."

The Ethereum co-founder stated that this opens the door to the possibility: If we have a sharded blockchain without a committee, the only mechanism for validating data is data availability checks?

He also proposed a possible design that includes an existing base chain, "similar to the Ethereum-style non-scalable blockchain, anyone can post transactions to that blockchain." Here, users It will be able to pay a fee and send a special type of transaction containing a data commitment, and the block proposer / miner can then perform a data availability check on that transaction to verify that the data is available.

"They need to do this through an anonymous network to avoid attackers being content with just their own checks and not with anyone else."

Buterin has also expanded how to accept only blocks that pass each availability check, and that data availability sampling will rely entirely on "absolute security."

Buterin proposes to do so because it provides consensus on a chain. He also mentioned the fact that the data on this chain is available and can be used as a base layer for building systems like rollups.

"… so [it] will reject any block containing a commitment to that unavailable data. It has no 2/3 online assumptions based on the committee's system."

However, Buterin also provides reasons to oppose such plans, such as scaling calculations, not just data, and avoiding "layer 2 reliance on synchronization assumptions." He also talked about how the committee has important spin-off benefits, such as the stability of ~ 128 randomly selected validators, which guarantees that the data has actually been downloaded and stored.