Vitalik Buterin published an article today titled "How do layer 2s really differ from execution sharding?", in which he mentioned that in public discourse, "Layer 2" and "sharding" are often described as two opposite strategies for how to scale blockchains. But when you look at the underlying technology, you'll find a puzzle: the actual underlying scaling methods are exactly the same, including some data sharding, fraud proofs (fraud prover) or ZK-SNARK proofs, and cross-(rollup, shard) communication solutions. The main difference is: who is responsible for building and updating these parts, and how much autonomy do they have?
An L2-centric ecosystem is a very real sharding in the technical sense, but in this sharding, you can create your own shards with your own rules. This is powerful and supports a lot of creativity and autonomous innovation. But it also faces key challenges, especially in terms of coordination. For an L2-centric ecosystem like Ethereum to succeed, it needs to understand these challenges and address them head-on in order to get as many benefits as possible from an L1-centric ecosystem and get as close to the best of both worlds as possible.