The word "modularization" sounds very grand, but for us end users, the most intuitive changes are just two: First, after "settlement location" and "execution location" are separated, the trading experience feels more like using different front-end routes, which usually means cheaper and faster; second, there are more cross-layer/cross-chain actions, and assets are moving around more frequently, essentially increasing the risk weight of bridges and message passing layers.

Recently, there's been a lot of back-and-forth about TPS/fees/subsidies in L2, but what I care more about is: where does your chain send data, who takes the blame if something goes wrong, and how are assets returned to the mainnet during downtime or rollbacks... My own approach is pretty simple: try small amounts casually, for large amounts only use routes that have been audited and patched with a normal update rhythm, and before crossing back and forth, check the bridge's upgrade permissions and withdrawal delay rules to avoid losing a few bucks in fees and ending up stuck on the bridge for a week.