To read or not to read?

I’m definitely not the best person to chime in here, but hey, posting is free!

A friend of mine used to say that all problems in software engineering are solved by introducing another level of indirection. Would it make sense to have another circuit that handles the “glue” between the app and the world state? This circuit could receive a large number of read requests from the app and expose them as a single old_private_data_tree_root to the kernel circuit. And it would be easy to swap out in the event of a fork that changes the internal state representation, while exposing the same “API” to the app circuits.

I can’t imagine many devs would want to opt for an app which only works until the next hard fork!!!

Jokes aside, I understand that hard forks would be designed to be backwards compatible with app circuits, right? Unless the internal state representation needs to change in ways like the ones you mentioned (eg removing a hash function that was shown to be insecure).

59 Likes