Ethereum’s execution layer team lead, Péter Szilágyi, has voiced concerns about the blockchain’s potential shift towards centralization.
PeerDAS is the direction #Ethereum is being taken into with the next forks, with the aim to raise (not immediately) the blobs up to a staggering 32MB.
That’s fun, but are we aware that local block production is collateral damage that the PeerDAS authors are willing to take?
— Péter Szilágyi (karalabe.eth) (@peter_szilagyi) July 26, 2024
He highlighted the proposal to implement the PeerDAS protocol, which among other changes, will gradually increase the size of binary large objects (BLOB) to 32 MB. Szilágyi believes this will negatively impact home stakers, describing it as their “public execution.”
“I didn’t sign up for this when joining Ethereum,” Szilágyi stated.
The developer explained that due to strict network timings, node operators will need to transmit several BLOBs to other nodes within four seconds, considering their distribution. This will require an internet connection with a speed of at least 2.5 Gbps for reliable operation, he noted.
“I would love to prove otherwise, but it seems to me that Ethereum is losing its way. The research team has fully embraced the idea of centralizing everything as long as it can be verified. It’s a funny charade: decentralized validation, but centralized control,” Szilágyi emphasized.
The introduction of BLOB objects in the network occurred after the implementation of EIP-4844 during the Dencun hard fork in March 2024. It is anticipated that PeerDAS, as EIP-7594, will be included in the next Pectra update, which is likely to be prepared for the first quarter of 2025.
One user responded to Szilágyi, stating that Ethereum’s move towards centralization began with the network’s shift from the Proof-of-Work consensus algorithm to Proof-of-Stake following The Merge upgrade in September 2022. Another commentator noted that a home node under current conditions resembles “a hobby or games like Ethereum Classic.”
Vitalik Buterin, co-founder of the second-largest cryptocurrency by market capitalization, called Szilágyi’s suggestion that the research team supports full centralization false. According to him, on the contrary, there are discussions on numerous measures to minimize this process. Regarding PeerDAS, they propose distributed block building and protocol bandwidth optimization.
> The research team fully embraced the idea to centralise everything as long as it can be verified
Having just come out of EF research workshops going on for the past week, I can confirm this is false, we had all kinds of discussions on minimizing centralization. This includes:…
— vitalik.eth (@VitalikButerin) July 27, 2024
In May, Buterin agreed with Szilágyi’s concerns about network centralization, citing MEV, liquid staking, and the cost of running a full node as major threats.