The future on L2s
Welcome!
In the following blog I’ll try to express what are my expectations about the future of L2s and their integrations (specifically speaking of rollup technologies).
Another thing to mention is that my main focus is on L2 solutions for Ethereum, although some experts/knowledgeable people in the space have said that it is viable to deploy this solutions in other L1 chains and that it’ll depend on how those L1 manage data availability and security/decentralization.
I want to clarify that I’m not a developer so maybe some of the implementations I’d like to see aren’t possible, but I guess time will tell.
First of all, Optimistic or ZK?
In my humble opinion Optimistic Rollups (OR) are a necessary, but temporal, step towards full blown scalability on Ethereum.
Why necessary?
Ethereum is currently unusable for lots of people, including some that started investing in ETH after the 2017 bull market. If your average smart-contract interaction most days costs anywhere from $50 to $150 in gas fees then it’s unpractical to use it with any amount under $1500–2000 (personally I would pay at most 0,5-1% of the transacted value in fees).
That been said, we need something to solve this issue now and that’s where ORs come in.
Even though they aren’t able to get transaction costs below $1 (for now) they are still been optimized and they also benefit from full EVM compatibility, something that’s incredibly hard to do on ZKrollups (to learn more see zkSync 2.0).
But what does it mean for us, end users?
It means that anything currently live on Ethereum mainnet can be “copy-pasted” into Arbitrum or Optimism with very minor changes. This greatly benefits the development of ecosystems inside different OR .
This is great for now, but when greater demand for L2 usage comes in it will also increase demand for L1 blockspace (because rollups need to keep publishing data and proofs to Ethereum) which will result in even higher L1 fees forcing more and more people to make the jump to L2s, provoking a vicious cycle until every user is directly on-boarded to an OR or ZKR.
Taking into account the limitations of OR on how much they can lower the transactions costs, people who want the security of Ethereum will eventually seek even cheaper options.
By that time I expect solutions like zkSync 2.0 (with their zkEVM) and Starknet (with Warp, their Solidity-Cairo transpiler) will kick in and finally enable small users to interact with Ethereum (at least for the foreseeable future).
How would that interaction be?
Currently Ethereum protocols/smart-contracts/projects (DeFi, NFTs marketplaces, DEXs, Bridges, etc.) are fragmented between L1, L2s and Sidechains (most notably Polygon PoS).
We can already see a little bit of the future I expect when using dashboards like Zapper, Zerion, Debank, among others, where you simply connect your wallet of choice and use it with different protocols and even blockchains. The catch is that you need to change the blockchain you’re interacting with in order to use the correct balance of your wallet.
What I’d like to see in the future is full compatibility between zkRollups like zkSync and StarkNet where transaction cost will be extremely low, especially in validium setups (called zkPorter by Matter Labs) where data is not stored on the Ethereum mainchain.
For more information you can read this Ethereum.org article, and you can also check out this video of Alex from zkSync explaining why zkPorter would be safe for the average user and how easy will it be to opt-in and out between zkPorter and rollup mode.
Going back to the full compatibility part, what I meant by that is that users shouldn’t have to manually check the network they’re in in order to use certain protocol, at most they should only need to flip a switch on and off between using validium or rollup mode in their wallet.
As it was said by Alex in the video I mentioned before, zkPorter and zkSync will interact seamlessly between them allowing users to choose if they want a specific transaction to be stored on-chain (safest) or off-chain (riskier).
Why would anyone want to choose validium mode you might ask?
Think of it this way, on a day to day basis while you’re at work, when you go to the park with your family, when you go shopping… You don’t want to spend 10 or 20 cents on a transaction for an ice cream worth 1 or 2 dollars, you’d want the transaction to be free or at most a cent or two. Who cares if you buying an ice cream is stored on-chain?
Now what about receiving your paycheck? Making some degen trading? Staking ETH? That’s 100% what you want to get stored on-chain just in case something wrong happens off-chain and you need to access that data.
Losing details about what ice cream flavour you bought? No problem.
Losing or getting 32 ETH freezed because a malicius actor or what ever screwd up the off-chain data? That’s a big no no.
So summing up, how do I envision the future for zkRollups?
Due to the huge reduction in gas fees this kind of solutions will bring, I hope users will be able to use projects deployed on Starknet mainchain/validiums or zkSync 2.0/zkPorter without noticing any change in gas fees because the only thing keeping them apart is a bridge between zkSync <> zkPorter <> StarkNet <> any validum <> any zkRollup.
Let’s think of an example:
You loaded up your “zk wallet” with some ETH, you find out that Uniswap launched on a zkRollup (you shouldn’t need to know which one), you go to their site with your wallet on rollup mode and inmediatly start making swaps. Now you have some YFI, DPI and UNI on your “zk wallet”, simple enough.
The thing is Uniswap launched on StarkNet and your wallet was deployed initially on zkSync, so it had to bridge your coins to StarkNet and back in order to complete your trades. Ideally the swap + bridging cost shoudn’t be noticeable for the end user, or in the “worst case” scenario you could use validiums in order to do the swap and later make the change to rollup mode so your balance is safely stored on-chain.
This example (I believe) could work out, but it’ll bring up some other interesting consequences, one that always comes to mind is how could one measure the TVL of a certain zkRollup/validum if they all could access liquidity freely between them with no limitations?
Is it all going to be a big “zk-pool” of different coins and tokens versus mainchain protocols TVL?
What I just mentioned isn’t an issue really, just something to think about… We’re blurring the lines between protocols by allowing them to use the same pools without any restrictions. This contrasts with the current L1 situation where projects are valued by their TVL instead of their usability or improvements over other competitors (you could see this with Sushiswap vampire attack, where they basically copy-pasted Uniswap and just created a token to attract liquidity).
You can read a more technical view into this “liquidity sharing” following this link.
And that’s it.
Our example was very basic and we only talked about swaps, but this interoperability between different zkRollups and validiums will open up a ton of applications which are impossible to do on mainnet.
Just by lowering gas fees DEXs will be able to create order books without having to worry about excessive transaction fees, gaming on blockchain could take off using validiums that every now and then save data on-chain, “real world” applications like medication traceability will be affordable for pharmaceutical companies allowing for a level of public transparency never seen before.
There is a lot to do in order to onboard the world into crypto, Ethereum took the first step by creating a secure and decentralized smart-contract blockchain… this is the second step in the right direction.
Thank you for reading!
If you have any comments or thoughts please let me know, I enjoy learning and discussing in a constructive manner!
In case you’re willing to support me you can:
Buy me a coffee! | Comprame un cafecito!
