program devs, do you utilize the irreversibility of PDAs? ignoring the technical limitations, the ability to derive non-sensitive seeds (e.g. owner, mint, etc) from the PDAs would make historical indexing 100x easier