Correctness in System Engineering

A philosopher has argued that there can be no such thing as a provably correct system. Some software reliability and safety experts believe that no system can have a software reliability of 1. These claims are related, and they are both mistaken, which we prove with a simple counterexample. But is this the only kind of counterexample? An interesting question arises: which types of systems admit of correctness proofs? Setting the boundary becomes important. I commence an analysis of what systems may consist in, and what the logical relations are amongst various concepts concerned with systems, such as speciications, correctness and failure. The goal is to put system engineering on a logically rigorous foundation. Discussing two more examples, I conclude that some useful systems may, not just in principle but in fact, be proved correct.