pull down to refresh
1 sat \ 1 reply \ @BlueSlime 16 Feb 2022 freebie \ on: Some thoughts on PoW vs PoS bitcoin
Any fork of Ethereum chain could easily be led by this one node: https://infura.io
Aside from Infura, Metamask also has a big say on which Ethereum chain is the 'real' one. OpenSea is another contender.
Jeff Bezos could easily shut down over 1/3 of the network and evaporate the stakes of all validators hosted on AWS (then whitelist approved peers at the firewall, that choose the 'right' fork of course).
I'm not sure why the Ethereum network has become so centralized over the years. It may have something to with the fact that a full (i.e 'archive') node on Ethereum now requires 10 terabytes of storage in order to store the full blockchain history:
https://etherscan.io/chartsync/chainarchive
I think Vitalik is a brilliant engineer, and I do appreciate what the EVM can do as a developer. I hope the core developers at Ethereum can figure out how to better scale the behemoth which they created.
Ethereum 2.0 is an interesting concept, but I don't think it addresses the real elephant in the room. I wish them the best of luck for the activation though. Hopefully it buys them more time while the Ethereum team pursues additional solutions.
Networks trend towards more decentralization or more centralization. The decisions made early on in Bitcoin, like the block size limit and 10 minute block times, are what help it trend towards decentralization. Ethereum's faster block times, unbounded block size (I say this because the devs and miners can increase this arbitrarily) are what force it to trend towards centralization. Even without PoS, Ethereum will eventually run into the problem of no one being able to run a node unless you can make a business out of it, which is what Infura does. Looking at BSC is like viewing Ethereum at 10x speed, where it's already impossible to run a full node for the average person, not that it provides them any benefits anyway.
There really isn't much weight behind the archival node's storage problem. In Ethereum, an archival node can be created out of a full node using its own data. An archive node isn't necessary for the average user, since a full node performs all the same full validation as well, it just doesn't store the extra bloat created in transactions, only the result. The real problem is the block size, which in ETH's space is the gas limit. Miners can increase this on their own, as long as they don't increase it by more than something like 1% per block. This is laughably bad for users because it makes it harder to run a full node, contributing further to the centralizing effect.
reply