Why I am excited about the Ethereum merge upgrade
To commemorate the new Ethereum upgrade I bought my first NFT from the Foundation collection. The artwork represents everything I feel about the blockchain (special thanks to the talent behind the creation Sachin Rajeev). There is not a single day that goes by without me learning something new in web3, and it makes me happy & excited for the future.
I see Ethereum as a single computer where everyone can write their contracts as programs. The computer ensures it’s executed correctly. It reminds me slightly of the central AI VIKI from “I, Robot” movie. Though I am sure Ethereum won’t turn into an evil computer that brings doom upon humanity and it is certainly not central either. It’s built and maintained by a decentralized community through collective effort. But that doesn’t mean we have built the perfect open computer for writing records or logs. In fact, we are only at 40% as Vitalik said. There are imperfections and mistakes of the past that have to be corrected and improved upon to move on to the ideal blockchain that we yearn for.
If you have watched Vitalik's presentation during ETHCC, you might have seen his ideal long-term goal about how the Ethereum development should reduce its complexity over time rather than making it increasingly difficult for one person to understand everything. In his speech, he says “Part of Ethereum being a trustless protocol should actually be (down to) Ethereum being simple enough protocol that if you really want to wrap your head around the entire thing you should be able to”. And I strongly agree with him, because simplicity is the ultimate sophistication. One of the qualities I admire a lot about Vitalik is his ability to explain really difficult topics in the simplest way possible. And I am glad that he is applying that philosophy on the protocol level.
A few weeks ago, the web3 community went into panic mode. No, I’m not talking about the regular retail investors who panic over every FUD news out there. Rather, the panic was amongst actual developers and researchers who build & maintain these decentralized products. Tornado cash a seemingly “harmless” smart contract was banned by the US gov agency called OFAC. And this scared the 💩 out of many prominent dapps and devs. Though I understand the pressure to remain compliant to protect their business. But after that incident, it was clear to anyone in the community that there is some strong centralization in some parts of the network and it’s important for the protocol to remain censorship-resistant in the future.
But even if the protocol is censorship resistant, the developers and users should be aware of the choice the web3 is presenting to them. By depending only on a few products we users are creating a shadow of centralization over an innocent decentralized protocol. Metamask for example relies on Infura to provide the RPC endpoint that eventually submits your transaction to the blockchain. But what if Infura never allows your transaction to reach the blockchain? You can look for alternative solutions but sometimes the options are limited by region and features so we end up having no choice at all.
So what stops the web3 world from having so many Infura’s? Well, It’s because, in order to submit transactions to the network, you have to be part of the network as a node. And that means you have to install the Ethereum client software and sync up to the current chain head which requires at least 2 TB+ storage and 16 GB+ memory. Full nodes only give you the privilege to submit transactions on behalf of the user, to actually compete with Infura a Full node is not enough, You have to be an “Archive node”. Archive nodes require at least 12TB+ disk space and have all the querying capability from the chain’s history. So the playing field is only leveled enough for capable cloud infra experts.
This is why the merge and the following upgrades (aka The surge, verge, purge, and splurge) set in motion for the future are important. It separates each task into different roles in the protocol. A node can be validating or submit transactions without downloading the entire chain. And those who validate the transactions are not the same nodes that arrange the transactions into a block which prevents censorship at the node level. This is roughly how the transaction lifecycle may look after the purge upgrade.
As you can see users themselves are part of the network now, I don’t think it can get any more inclusive than this. The important difference is the responsibility of building, executing, and storing the block which used to be done by almost every node or miner is now separated into different roles and each node can choose to play any part it likes. This is probably an oversimplified way of saying things, but I believe that’s the gist of it. I hope now you are also excited about the upgrades that will make a more inclusive, secure, and decentralized blockchain possible.
In this blog, I have only talked about centralization over blockchain writing (submitting transactions). This can happen for blockchain reading too. For example what if the node(Infura-like archive nodes) that we depend on to get the latest blocks turns out to be malice? Although this doesn’t have to be solved on the protocol level, applications can just depend on multiple providers to verify the truth or use projects like graph protocol . It runs a different decentralized network of indexers outside the Ethereum network whose purpose is to serve the latest blockchain data to applications that need it in real time.
