Aren't two layers not enough for Ethereum? As Layer 2 platforms like Polygon and Arbitrum are thriving, developers in the Ethereum space are already thinking about additional chains to help the Ethereum ecosystem grow.

In his article What kind of Layer 3s make sense? Vitalik Buterin talks about a few ways in which Layer 3's could work out as a useful addition. We'll talk about two here: 

1. Scaling
2. Customized functionality such as privacy

So why Layer 3's? Let's remind ourselves of the goal of Ethereum and competing blockchains. They want to become a 'world computer', processing gigantic amounts of transactions. Not just financial transactions, but data in general. As the idea of blockchain tech is decentralization, people that run nodes can't be burdened with terabytes upon terabytes of data. That's why the data size of the Layer 1 blockchain (Ethereum in this case) has to be limited. 

And that's were Layer 2's come in: they offload transaction and data load. They compress ('roll up') data, so that the Layer 1 is less burdened - and thus less expensive. The first solution Vitalik proposes, sees Layer 3 as taking this scaling one level further. A company called Starkware is already working on this so-called hyper scaling. And with success: StarkWare has already rolled up as many as 600,000 NFT mints into a single transaction on Layer 2 ImmutableX. More on that below. 

Use case 1: scaling - a proof of a proof of a proof...

Even though the way they do this sounds alarmingly technical - recursive zero-knowledge proofs - it is interesting to have a look. Everything in crypto is about signing transactions, about proving that you own coins, that such and such transaction is valid. But ideally, you can proof something without revealing too much about yourself. That is where zero-knowledge proofs come in. They allow you to proof that you have, let's say, enough funds to pay someone on-chain, without revealing how much you own. 

In other words, these zero-knowledge proofs or zk-proofs combine cryptographic proof with privacy. They are used on Layer 2. The computation is done on Layer 2 and then posted back to Layer 1 as what is called a validity proof. This means someone went through all the transactions in a rollup, did some work and found that they were not fraudulent. They then submit their list transactions and their proof to layer-1, for example Ethereum.

So how did Starkware succeed in rolling up the transaction equivalent of 600 k NFT mints? By making the above zk-proofs recursive. They could take 10 zero-knowledge proofs — each representing a thousand transactions — and generate a new zero-knowledge proof verifying that those other 10 proofs are correct. 

An example from real life. Suppose I want to prove to you that I spent 7 successive days in a certain park. On the first day, I take a selfie with the newspaper of that day. On day two, I make a selfie of myself holding both that day’s newspaper AND the selfie from day one… etcetera. All in one photo, instead of seven. Which makes my photographic evidence take up much less space.

 Recursive proof. Source: CoinGeek

 In this case, the aggregation of proofs happens on Layer 3, and the recursive proof is verified by a single proof on Layer 2. And that's how Starkware could roll up 600 k transactions with only using one transaction on Layer 2. According to the company, the limit is not reached. 

Use case 2: customized functionality such as privacy

Apart from scaling, according to Vitalik, Layer 3 could help specific applications to develop without having to connect to the expensive Layer 1. Privacy applications are an example.

Let's say a new application is slowly gaining traction but not super widely used yet. The number of transactions in such a case would be too low to be profitable for Layer 2 rollups - which have to use the expensive Layer 1. See, that's the catch with roll-ups: they pay a high fixed cost every time they submit a batch of transactions to Layer 1. Therefore, in the current set-up, they are profitable only at high transaction volumes.

Conclusion

More than 4.000 Ethereum developers across the globe keep working on scaling this 'world computer' so that it can be widely used for everyday use cases. In the meantime, more and more consumer brands start jumping in web 3 and NFT's. The ETH price may have dropped, but the fundamental developments are looking good.