The thesis is simple: Interoperability first is good for maintaining a cohesive ecosystem while leaving interoperability for later risks heavy ecosystem fragmentation.
Cosmos and Polkadot are the two best known interoperability chains out there. They were designed from the ground up with interoperability first in mind. It means that since the early days interoperability among their members is present. Polkadot has XCMP and Cosmos has the IBC.
Both allow for native interoperability across blockchains and parachains connected to them.
On the other hand we have Ethereum, that has abandoned for the most part native sharding efforts which were once claimed as the final piece of Ethereum 2 scalability in favor of layer 2 solutions (L2), rollups and sidechains mainly, that have been proposed as the solution for Ethereum’s scalability concerns. This came after the publication of the Rollup-Centric plan for Ethereum 2. It means that competing L2s on Ethereum will fight for users, applications and volume. There are three reasons for this:
1.- L2s are not interoperable now.
3.- Without interoperability and composability among L2s, the ecosystem becomes fragmented with layers 2 fighting for users, volume and applications to keep them inside their L2 or inside a single rollup.
Non-interoperable L2s make the ecosystem fragmented.
L2s are great, they allow for repeated use of applications with less fees. Inside rollups you can have interoperabiliy and even composability. But not all applications will exist inside the same rollup or L2. But without interoperability and no ability to communicate to other L2s, users still rely on L1 user transactions and, in Ethereum’s case high fees, to move to-and-from diverse L2s and fast and cheap app operation is limited to a single layer 2 or rollup.
Let’s imagine a user who wants to move his L1 ETH to a L2 rollup. That transaction will incur L1 fees. Let’s say the user wants to withdraw his ETH living in a rollup inside the newly announced Optmism L2 to Uniswap V3 who will most likely live in another rollup. That will also incur additional fees. What about Polygon? Moving funds there will also incur L1 fees.
But fees are not the only issue, it’s the user experience. Switching L2s at the moment is another hop users need to go through. Many won’t even realize that different applications are located in different rollups causing funds to be sent to addresses in L2s they are not part of. This will cause that L2s will try to attract users, applications and volume to their L2 or rollups and other L2s and rollups and their corresponding applications will become their competitors.
The proposed solution for L2 interoperability and composability at the moment is Connext. It’s a third party application that theoretically will allow communication between L2s and it’s still undergoing heavy research and there are no clear answers on how they will achieve this objective so it’s still uncertain. Moreover, trusting interoperability to third parties will naturally decrease decentralization and will include a centralized point of failure on interoperability.
Interoperability First on Cosmos and Polkadot.
Cosmos and Polkadot have created their ecosystems with interoperability first in mind. So that from the moment applications connect to their blockchain, they provide native interoperability. So applications deployed in Cosmos or Polkadot don’t need to trust third party applications with interoperability and can be sure they will be able to communicate with other applications inside the same Hub or applications sharing the relay chain respectively.
The consequences are evident, in Cosmos and Polkadot there are no ecosystem fragmentation due to standardized use of interoperability. Not as an afterthought but as a foundational piece of their ecosystem.