For over a year now, I’ve commended Tezos as the L1 with the best tech roadmap in the industry, and they’ve continued delivering relentlessly. Indeed, my last thread before quitting Twitter was about Tezos’ achievements. Unfortunately, tech matters very little in crypto - it’s first and foremost a socioeconomic movement. So, despite their fully decentralized enshrined rollups, there’s been very little real adoption - nearly everything interesting is happening on Ethereum rollups. But here’s an interesting possibility that gets the best of both worlds - validiums or optimistic chains that settle proofs and bridge to Ethereum, but settle data on Tezos.

Nomadic Labs detailed their approach to their data availability layer last week. Now, I haven’t had the chance to read their in-depth design, so this post is not going to be about that. They are taking a similar approach to Ethereum - V1 without data availability, then V2 gradually introduces data availability sampling. The most impressive thing about Tezos has been their timeliness - they have released ideally decentralized rollups well ahead of anyone else in the industry. Now, a data availability layer (DAL henceforth) is a different kettle of fish, but Tezos has a track record with other things as well - including a consensus protocol swap. I won’t be surprised if Tezos is first to deliver the first complete DAL.

Why not use a different DAL like Avail, Celestia, or L2-specific like zkSync’s zkPorter or Arbitrum’s AnyTrust? It comes down to economic sustainability. Tezos has been around for years now and has established a baseline economic adoption. Granted, it’s not very much - coming in an at less than 1% of Ethereum - but it’s there and it’s proven. As I’ve argued multiple times before, a pure DAL will have an uphill battle finding economic adoption, given data is the relatively orders of magnitude more abundant, scalable and commodified resource than their consumers - execution on rollups. Of course, new DALs can find said sustainable economic sustainability & security, but it’ll have to be through means other than just selling negligible-cost data. On that note, zkSync has an easier path as their execution layer Era is already delivering substantial usage for months now. AnyTrust or StarkEx DACs bring forth an honest-minority assumption, which is very interesting too. While currently they are permissioned, they can be made permissionless through mechanisms like governance onboarding ala Lido. EigenDA is interesting too, which can bypass that uphill challenge by restaking ETH. Even if EigenDA stakes only 1% of ETH, it’ll still offer greater economic security than Tezos, and given it’s on-demand model it’ll be ideally sustainable. EigenDA is a complex solution and there are more unknowns - so it could come before Tezos’ DAL. Honest-minority DA layers like Adamantium are very interesting too, and could be a bonus add on top of existing data layers. Clearly, the DA layer space is wide open for innovation, and we’ll see much to come, but many of these solutions will take 3-5 years to achieve maturity and sustainable economics. Tezos has the unique advantage of being an established, sustainably economically adopted network, for nearly half a decade now.

Then there’s Ethereum, of course. EIP-4844 is deep into development, with devnets being spun up supporting it. I expect Tezos’ DAL V1 to release around the same time as EIP-4844. As I’ve commented previously, I believe EIP-4844 will likely be enough for most valuable financial applications for the foreseeable future. However, there will be low-value, non-financial apps that’ll be data hungry and require more data bandwidth. Ethereum does have a long term solution for this - full danksharding - but based on prior track records, I’d certainly expect Tezos’ DAL V2 with full DAS to come before Ethereum’s full danksharding which IMO is still multiple years away with complex prerequisites like PBS to come first, and more stringent limitations. As a sidenote, from the little I’ve understood, Tezos’ DAL V1 has decentralized block production straight away, which won’t be available on the other data availability layers. Only Ethereum’s post-PBS full danksharding will offer this, but like I mentioned, that’s probably years away. Now, with DAS, the requirement for decentralized block production is less critical, but when we’re talking about massive data, you need to keep block production adequately decentralized as well, and Tezos delivers this requirement from day 1.

An interesting hybrid are enshrined smart rollups that verify and settle to Tezos, but also have a validating bridge to Ethereum that verify the same proofs. This could potentially bring Ethereum’s vast liquidity and audience into the Tezos ecosystem. Then there’s also the possibility of an Ethereum rollup forking off Tezos’ work around rollups to bring a decentralized rollup to Ethereum. To be clear, all of these solutions bring many additional assumptions, each must be assessed individually, and at the end of the day, the ideal solution is going to be a full Ethereum rollup which also settles data on Ethereum for the foreseeable future. But not every chain needs ideal security properties, and an Ethereum-Tezos hybrid validium or optimistic chain is certainly worth considering for lower value applications. On that note, I’d also urge Tezos to consider this angle, build frameworks, documentation and incentives to encourage Ethereum validium/optimistic chain projects to settle data on Tezos - it might be what brings much needed adoption for their industry-leading tech.