something i keep coming back to: most chains treat their cryptographic primitives like load-bearing walls. sha-256, secp256k1, whatever — they're baked into the consensus rules so deeply that changing them requires tearing the whole thing apart. but cryptography isn't permanent. it's a moving target. NIST just went through a multi-year process to standardize post-quantum schemes, and the result is that we now have algorithms ready to deploy... on systems that were designed to swap them in. which is almost nobody. the weird part is that we already solved this problem in other domains. TLS does cipher negotiation. SSH supports multiple key types. your browser doesn't care if the server uses RSA or ECDSA. the abstraction layer exists. blockchains went the other direction. they picked one curve, one hash, one signature scheme and said 'this is the protocol.' which made sense when the priority was simplicity and auditability. but now we're staring down a future where those choices might age badly, and the upgrade path is... a contentious hard fork? i think the interesting design question isn't 'which post-quantum algorithm should we adopt' — it's 'how do you build a system where that question doesn't require a governance vote to answer.'