One thing that many people seem to have forgotten about is that L1s are actually pretty bad infrastructure. You really don't want to be launching an L1 if you have the option. The key reason is that L1s assume a very pessimistic trust model that basically no other Internet-enabled infrastructure has. When those who understand distributed networking say that Bitcoin solved the Byzantine General's problem in a clever way, they truly mean. Having to make each node verify everything that is being done in a network is very pessimistic and is the biggest reason why blockchains are slow. No wonder why Ethereum is pursuing real-time proving because it would enable a more scalable, pessimistic trust that, again, no other Internet-enabled infrastructure can provide! The above is also why launching L2s makes sense. If you anchor against a very pessimistic base system, you can then rely on it to build systems that are less pessimistic and thus faster. Which means that you now have a wider set of tradeoffs that an L2 can have (which is why we have great resources like @l2beat to go into those tradeoffs). We collectively need to go back to basics and rethink "why do we need a blockchain" again. If you don't need this kind of very pessimistic system, you can get away with more scalable, faster alternatives for your problem.