Good question. In these posts, I generally ignored computational constraints—i.e. effectively assumed infinite compute. I expect that one could substitute a computationally limited version of probability theory (like logical induction, for instance) and get a qualitatively-similar notion of abstraction which sufficiently-computationally-”scrambled” info is no longer an abstraction, even if the scrambling is reversible in principle.
Does abstraction also need to make answering your queries computationally easier?
I could throw away unnecessary information, encrypt it, and provide the key as the solution to an NP-hard problem.
Is this still an abstraction?
Good question. In these posts, I generally ignored computational constraints—i.e. effectively assumed infinite compute. I expect that one could substitute a computationally limited version of probability theory (like logical induction, for instance) and get a qualitatively-similar notion of abstraction which sufficiently-computationally-”scrambled” info is no longer an abstraction, even if the scrambling is reversible in principle.