Its difficult to have both scalability and decentralization on a blockchain.
If the blockchain community cares about maintaining decentralization, we must prioritize decentralization over scalability.
Ethereum and Decentralization
Yes, decentralization matters. What makes ethereum decentralized?
As Polynya describes on the linked Medium article, in order to ensure decentralization, it is important for a significant number of average users to be able to run a validator node.
A validator node is a someone with a computer within a peer-to-peer network that ensures consensus and validates transactions on the blockchain network. This prevents falsification and fraudulent activity, ensuring the blockchain remains trustworthy and tamper-proof.
To increase decentralization, blockchains must not create barriers or restrict / limit anyone’s ability to serve the community as a validator node.
Blockchains also need a way to incentivize users to become validator nodes. Today, ethereum does this by something called “proof of stake”, where validator nodes earn interest for their work.
Today, ethereum does a pretty good job of this by keeping block size low, meaning less computing power is needed to validate each block.
And while validator nodes on ethereum technically require 32 ETH, users can join a staking pool to participate with a small group.
How is a decentralized community of validator nodes structured?
To ensure that ethereum continues to have as many validator nodes as possible, it needs to be easy for people to serve as nodes.
Users with consumer laptops/hardware and novice technical acumen should be able to participate in the community as a validator.
Compare the two possible blockchain architectures below and how they might affect the average person’s ability to serve as a node:
- Scenario A: Low computing power is required for validation. Someone with a regular laptop (the kind you can buy from best buy or costco or apple) is able to run a node on the network and act as a validator.
- Scenario B: High computing power is required for validation. A network full of a high-performance servers, or million-dollar supercomputers acting as nodes.
The model described above in Scenario A is decentralized because there can be a LOT of validators and nodes ensuring that no bad actors try to take over.
In Scenario A, while maintaining decentralization, the tradeoff with having a culture of small validator nodes means the blockchain is unfortunately inefficient and less scalable than a centralized system can be.
Scenario B is way more centralized because there are fewer people that will be able to serve as a node on the network.
The small number of nodes in Scenario B could in theory work together to take advantage of some portion of users at any given time.
Although Scenario B will be highly scalable, lower cost, and will enable the network to run faster, having such a structure defeats the purpose because the whole reason blockchains are valuable in the first place is to enable decentralization.
Scenario B could be less-favorable for end users because you cannot trust the few large-scale validator nodes from acting out of self interest.
To maximize decentralization, as many people as possible should be able to serve as a validator node on the network.
With present technology, true decentralized systems are great for establishing trust and avoiding but not so great for scalability.
The good news is that Ethereum will release data sharding which will increase scalability, but the risk is that, by Polynya’s estimates, it may take until 2023.
This is the world we’re living in today with Ethereum.
As Polynya describes below, based on the current state of the internet, I’m not so sure whether or not users truly do care about decentralization.
“there’s plenty of evidence from the web2 world, where the most adopted services are the ones that are most convenient, despite their overwhelming problems with privacy.” Polynya on MediumPolynya on Medium
Perhaps some people simply want to use the most convenient blockchain with the lowest fees.
However, this partially misses the point.
In traditional models, as in Apple’s app-store, users experience fees on the backend, where the centralized entity takes a percentage cut of all app-generated revenue.
This idea came from BelugaWilliams’ tweet below.
The advantage of decentralized cryptocurrencies, albeit stressed with high-gas fees, is that creators and developers can maintain ownership of whatever they create.
What does this mean for ethereum?
Given that some users inarguably do prefer networks with lower fees, one possible outcome is that ethereum is overtaken by another, more centralized blockchain.
Although this may feel unlikely, it would be naive to ignore it as a hypothetical future scenario.
Today, ethereum does feels like the dominant chain.
More projects are built on top of ethereum than anything else.
Ethereum has created a rapidly growing application ecosystem.
From NFTs to layer 2 blockchains to lending and borrowing apps, as of right now Ethereum has more developer usage and user adoption than any other blockchain.
- NFT’s, for example, while largely containing hype infused projects, also contain projects building a genuine community for their members. Surf Punks NFT, for example, is a community centered around surfing, and is even throwing a few wave pool surfing events.
- Decentralized apps like Sorare enable enable fantasy futbol, powered by ethereum. What’s amazing is that all fees, transactions, and blockchain jargon is abstracted behind the scenes, so that all people experience is a user friendly fantasy sports game with their friends.
Ethereum’s current state is not without risks. The fact that sharding will take until 2023 to implement means that scalability will continue to be a challenge, although layer 2 chains may help segway some of the demand off-chain.
Given Ethereum’s strength, an unknown and unforeseen up-and-coming chain is unlikely to takeover without anyone realizing.
Sure, it is technically possible, but probably unlikely as the network effects of ethereum are growing rapidly.
However, while the new dapp ecosystem is focused on Ethereum, perhaps someone in the world is working to build a true Ethereum killer that will arise from out of nowhere and take us all by surprise.
No one knows for sure.