According to co-founder Vitalik Buterin, Ethereum can afford to lose its finality infrequently without seriously endangering the network, even after a recent client error almost disrupted the blockchain's confirmation mechanism.
Following a recent bug within the Prysm Ethereum client, Buterin said in an X post that there was “nothing improper with losing finalization every so often.” He added that the finalization means the network is “really protected” that a ban won’t be reversed.
Buterin argued that it’s okay if finality is occasionally delayed for hours because of a fatal error, and that the blockchain continues to operate during this. The real problem is something else, he said: “The thing to avoid is bringing the improper thing to a conclusion.”
Source: Vitalik Buterin
Experts fear a lack of finalization
Fabrizio Romano Genovese, Ph.D. in computer science from the University of Oxford, England, a partner at blockchain research firm 20squares and an authority on the Ethereum protocol, Buterin agreed.
He said that if finality is lost, Ethereum will change into more like Bitcoin (BTC), stating that Bitcoin “hasn't had finality since 2009 and nobody is complaining.”
A proof-of-work blockchain like Bitcoin's can branch into multiple chains, with the one which receives essentially the most work (often the longest) being considered valid. However, if a secondary branch grows a lot that it overtakes the foremost branch, the foremost branch and the transactions inside it change into invalid – this is named reorganization.
This is how Bitcoin works: Its finality is probabilistic relatively than deterministic, because although this is sort of inconceivable after enough blocks have been added to the foremost branch, a reorganization can still theoretically occur. Genovese explained the difference between Ethereum and the foundations that designate blocks as “final.” He added:
“Ethereum has a finalization mechanism: if a block receives greater than 66% of validator votes, it becomes 'justified'. If greater than two epochs (64 blocks) pass at that time, the block is finalized.”
This isn't just theoretical; it happened in May 2023 because of an incident just like the recent one involving the Prysm customer. Genovese said these incidents didn’t make the chain unsafe; Instead, “it just implies that our guarantees around Reorg are temporarily probabilistic again and never deterministic.”
Consequences for L2s and Bridges
Still, Genovese noted that a scarcity of finality would impact the infrastructure that relies on it, including some inter-blockchain or Layer 2 (L2) bridges.
A representative from Ethereum sidechain Polygon told Cointelegraph that Polygon will proceed normal operations, but transfers from Ethereum to the sidechain “could also be delayed while awaiting finality.”
Additionally, the Polygon spokesperson said that the AggLayer cross-chain settlement layer would delay transactions from Ethereum to L2 until finality is reached again. Still, they said that there’s “no scenario where users experience a rollback or an invalid message because of lack of finality”:
“The practical impact of a delayed finality event is solely that it might take longer for deposits to seem. Users won’t be exposed to reorganization-related withdrawals beyond this delay.”
Genovese blamed such delays on developers demanding finality. “If a bridge builder decides to not implement a fallback mechanism within the event of lack of finality, that’s their decision,” he concluded.
