Why Layer 2 Rollups is the only way for the development of Ethereum?

Original author: Polynya

Original compilation: 0x137

I am very happy that the current currency circle has transitioned from the stage of "Rollups are not good" to "Rollups are good, but only if I hold Some L1 public chains". However, the whole point of a modular architecture is that it makes the concept of L1s disappear.

Rollups will always be the safest, most decentralized, most robust and most affordable The actual battle-tested SL layer (Settlement Layer) is verified, and it is backed by the most stable funds and the most extensive liquidity and depth. Its cost is negligible and not worth the risk of bridging the less secure SL layer. I believe Bitcoin would be the best SL layer, but since it doesn't support rollups, Ethereum is the next best option. For a PLONK rollup that executes 1,000 TPS (Transfer Per Second), the gas fee per transaction is only $0.00003, which is basically negligible. Therefore, in order to compete with Ethereum and Bitcoin, the only option for other settlement layers is to become safer, more decentralized, and more robust to coordinate the liquidity of more assets. As mentioned above, gas cost is an irrelevant argument because it is negligible.

But actually, Data Availability is a more interesting area. Currently, data availability in Ethereum is definitely very expensive, so it makes sense for ZKR (ZK Rollups) to take advantage of the alt-DA (Alternative Data Availability) option. But, so far, projects like StarkNet and zkSync 2.0 are building their own solutions instead of using "L1s". Data Sharding and Data Availability Sampling have turned this dilemma on its head: the more distributed your network is, the greater the DA throughput. So, in any case, the current L1 chain without DAS (Data Availability Solution) is completely obsolete. In the future, projects like Celestia and Polygon Avail will be more capable of capturing DA's market share.

In short, Rollups are inevitable. With the SL layer, you have to be more decentralized, more secure and more robust, there is no other way out. When using the DA layer, DAS must be implemented, and the decentralization should be as low as possible, because higher decentralization equals higher Rollups TPS. I know that people's confusion comes from the deep-rooted thinking of a single public chain. In this kind of thinking, decentralization and throughput can only be a trade-off. But in the modular world, this is no longer the case. If you don't adopt rollups, you have to compete on decentralization. In addition, since OR (Optimistic Rollups) is based on the operating assumption of "universal honesty", it does not make sense to use a DA layer with high operating costs. But in the future there will be some very low-value ORs on Celestia or Polygon Avail, because on a high TPS L1 chain, it doesn't make much sense.

Finally, something like StarkNet or A project like zkSync 2.0 will definitely choose the best settlement layer, providing DA with multiple options, allowing users and developers to make their own choices among them. So, the trade-off between decentralization and throughput does not exist at all. To clarify, some traditional L1s are building data shards with data availability sampling, notably Ethereum and Tezos, and I believe NEAR has similar plans. But it is likely that new DA layers like Celestia, zkPorter, and Polygon Avail will gain the upper hand in the short term.

Welcome to join the official BlockBeats community:

Telegram Subscription Group: https://t.me/theblockbeats

Telegram Discussion Group: https://t.me/BlockBeats_App

Official Twitter Account: https://twitter.com/BlockBeatsAsia