Over the past couple of days, I helped a friend look into “how the transfer was clearly made on-chain, so why hasn’t it arrived yet,” and after messing around for half a day, I finally found out it wasn’t that he was slow—it was that the data setup he was looking at was a bit “late.” Put simply, the “on-chain” you see in your wallet/browser is, in many cases, a view assembled for you by an RPC node plus an indexing service: the node is still syncing, the RPC has caching, and the indexer hasn’t finished scanning the new blocks yet—so you feel like the chain is “stuck,” when in reality the data was written on-chain long ago.

So sometimes I check two more entry points, or I’ll just switch to a different RPC; don’t immediately start suspecting that the contract has run away… It also reminded me of the recent pile of arguments about NFT royalties, which are pretty similar: everyone focuses on the “display” and “settlement” layer in the secondary market, but how the underlying data is read—and when it’s read—can be just different enough that it leads to nothing but misunderstandings.

What I’ve learned isn’t techniques, but to not take “what I see” as “the on-chain state right away like this.”