Tinkered with it for half a day and finally got that Subgraph indexing working—only to find the data still occasionally stalls. I asked a few node guys, and they all said RPC rate limiting is just normal, especially during peak periods on popular chains. No matter how fast the indexer runs, it won’t help—those nodes will just hit you with a 429.

Plainly speaking, whether on-chain data is fast isn’t often about how small the blocks are. A lot of the time, it’s whether the infrastructure behind you is willing to “make way” for you. Big node providers have strategies; small retail users can only rely on luck, or set up a lightweight node of their own and wait slowly.

Lately, people have been comparing RWA and on-chain yield products—gets me every time. One side is steady US Treasury yields, and the other is on-chain machine-gun pools rolling over and over while still being afraid the contract might go sideways. Anyway, I don’t believe any claim that “on-chain yield is always higher than traditional.” A fast pace doesn’t necessarily mean better returns; sometimes being half a step late is actually protection.

Oh, and someone asked, “Then why don’t you just use on-chain-native RPC to build the index yourself”… (silence)