Thread by @_KtorZ_

Does #Cardano need rollups (optimistic or zk)?

Probably not.

Rollups are a technical solution to a particular problem: cost of on-chain computation.

Rollups move on-chain computation off-chain, while keeping most of the traffic on chain.

1/n

This is crucial for Ethereum for which on-chain contracts have turned into large and complex programs that are very resource demanding.

Blockchains are *not* general-purpose programming platforms. They do poorly if one tries to execute its entire app logic on it.

2/n

Cardano addressed that from the start by taking a very opinionated, albeit unusual, road when designing the EUTXO model.

The on-chain logic is only about validation and is a lot less ressource demanding. Most of the DApp logic *already* happens off-chain, by design.

3/n

Plus, due to the current size of scripts and capacity limitations, execution times won't be an issue before a while.

Putting efforts in optimizing the compiler or, working on alternative compilers are much more interesting than rollups (for Cardano!) to me.

4/n

So, when looking at scalability issues to tackle first on Cardano, throughput and settlement time is what comes first to my mind, and these aren't solved by rollups.

We hope Hydra heads can help with some of that. And we have many ideas on top of heads already!

5/n

Yet Hydra isn't a unicorn either and it has limitations. But it's not like rollups are addressing any of them because, once again, they solve different problems.

So I'd rather keep the efforts on building on top of Hydra, than start a new effort in the other direction.

n/n