Author: Crypto Town with Sharvil
Prabal: Avail is a dedicated DA solution, but DA is just something we are focusing on right now. We are building more internally and will be talking about them at scale in the coming months. We want to create a foundational layer around which all the important components are built, so that it can create a very nice unified experience for users. Because as you mentioned, there are a lot of problems with the current way of doing things.
There will be millions of rollups in the future, and in that case, what will the user experience look like? These are the key points we want to improve. So we want to make sure not only that there is a data availability layer, but also how the interactions between rollups work. What will the user experience look like? How to implement asynchronous messaging and so on. We want to treat these issues as API calls between rollups, etc., and that's what we want to improve.
Moderator: What are the differences between Celestia, Eigenlayer, and Avail? Why should developers choose Avail over other options?
Prabal: I think the core details of the three are different. People may talk about other aspects, but I like to focus on our core technology.
Technology has trade-offs. Celestia uses a hash-based scheme that is efficient and easy to implement. We take a more proof-of-validity approach, combining KZG commitments and erasure codes. This means we do not need to rely on underlying fraud proofs, which Celestia's data availability relies on to ensure security.
What does this mean for developers? Why should they care? If you are running a rollup and have an execution proof, you don't want to wait for a DA fraud proof. That's why they should choose a solution like Avail, where they can get immediate DA guarantees without having to wait for fraud proofs.
As for the Eigen Layer, I think there are still many details to be worked out, and their specific operation and performance benchmarks remain to be seen. But roughly speaking, they are a data committee.
You can have two solutions: either have a data committee or have a dedicated DA L1. Eigenlayer offers more preference towards data committees, depending on how many people are on the committee. While cryptoeconomic guarantees are good, data availability is non-attributable, meaning there are no attributable errors, so cryptoeconomic guarantees can only help to a limited extent. This is why having a dedicated DA is very important.