Thread by @anthonykrose

1/ The L2 landscape has changed a ton over the past few years. I definitely didn't use to have to scroll down on @l2beat.

I also know it is not easy to understand the differences between these systems.

While I have a coffee, here is a TL;DR on what we're building @zksync.

@l2beat @zksync 2/ This space is complex, and even for developers working inside the Ethereum ecosystem I *know* it isn't easy to understand the differences between rollup projects.

You then mix in hype/marketing and it can be difficult to see the signal from the noise.

@l2beat @zksync 3/ Optimistic vs ZK is still not well understood, let alone the more exotic creatures in the blockchain zoo (optimiums, volitions, validiums).

I'll try and keep this as simple as I can, because this system is genuinely really exciting and I want people to understand better.

@l2beat @zksync 4/ The vision is a system that is the infra under an application layer that adds value to the lives of billions of people.

Applications that make people's lives better, much like the internet today, but supercharged because of what blockchains can add for value and ownership.

@l2beat @zksync 5/ To realize this vision, it is obvious to me that this system needs to scale like the internet. Engineers call this "horizontally" - essentially we can just add hardware and expand the network.

We need to do this with no compromises.

@l2beat @zksync 6/ This is why ZK is a superpower. You can prove the correctness of a huge system that performed a ton of computational work with a tiny proof that can be verified on any smartphone.

This is why ZK systems can scale horizontally (sensibly), and optimistic systems can't.

@l2beat @zksync 7/ To offer a bit more context, this horizontal scalability really just means: deploy more rollups into this protocol layer. They can be app specific, or general, but all of this can be abstracted away from the end users.

Here is a really scientific image that hopefully helps:

@l2beat @zksync 8/ At the protocol layer, these systems all need to trustlessly interact. If they don't, the entire user experience is fragmented (just as today it feels fragmented when interacting with different L2s, or L1s for that matter).

@l2beat @zksync 9/ Optimistic systems can't implement this interaction properly.

The options are basically:
- wait 7 days for the interaction (the challenge period)
- every L2 to run a full node for every L2 any of its users want to interact with
- use a middleman, who charges users.

All bad.

@l2beat @zksync 10/ Using ZK lets these systems interact trustlessly, any rollup to any rollup. They can just interact and _know_ that the state of the other system is correct.

The math proves it.

@l2beat @zksync 11/ Users won't even need to know where the game they are playing is hosted, or which rollup their friend is using when they want to send them $20 for lunch.

They shouldn't need to know.

@l2beat @zksync 12/ Optimistic rollups also need to post all of their transaction inputs to the L1, because ... this is how they work! The data is needed to resolve a potential challenge to the state transition.

At the scale we're talking, this is a huge amount of data.

@l2beat @zksync 13/ ZK systems also don't have this problem.

You can prove the correctness of the system without posting all the state to L1 (you can post somewhere else, or build a private system). This is a validium. You know it is correct, but you don't have the state on Ethereum.

@l2beat @zksync 14/ Did I mention that proofs can be combined across the whole @zksync ecosystem, and you could verify the _entire_ network in a single proof? This means the onchain costs trend to zero as the system scales.

Oh and you can verify it on any smartphone.

@l2beat @zksync 15/ The system we're building @zksync is highly scalable, highly performant, highly secure, and completely different to some of these other systems that are unfortunately all lumped together as "L2s".

These things are not the same. Many will be gone one year from now.

@l2beat @zksync 16/ The ecosystem will be made up of many ZK chains, often configured differently to match specific use-cases, but with the same strong security guarantees and fast, trustless interop.

@l2beat @zksync 17/ Anyone will be able to permissionlessly deploy a new ZK chain into the ecosystem, and interact across the network instantly.

@l2beat @zksync 18/ There is a huge amount more to say, but I've finished that coffee and this is the key intuition I wanted to share today.

This is why I'm excited about @zksync, why ZK systems really will change the world, and why ZK is the endgame.

@l2beat @zksync 19/ Please send questions if things aren't clear. I plan to add more context soon about the v24 upgrade (which is related to this "shared bridge" in my beautiful picture above).