This observation comes from Tony Hoare’s paper “How Did Software Get So Reliable Without Proof?”. It reframes the purpose of testing. On the usual view, a test succeeds when it finds a fault in the code; Hoare suggests its deeper value lies in what a fault reveals about how the code was made.

A bug, on this account, is a symptom. It points to a weakness in the methods used, a lapse of concentration, or a gap in the skills of the people who designed and wrote the program. Fixing the individual defect treats the symptom; attending to what the test exposes addresses the cause.

The point connects to Hoare’s wider interest in how reliable software is produced. By treating tests as feedback on the development process and its practitioners, rather than only on the product, teams can improve the way they work and prevent whole classes of error.