3 ms·
Dijkstra for example didn't think so, I think. He had a din view of testing as compared with proofs for ensuring software correctness.
by billfruit 2y ago
Dijkstra for example didn't think so, I think. He had a din view of testing as compared with proofs for ensuring software correctness.
- bluGill 2y agoThey both have their place. If you get a requirement wrong you can prove an incorrect system. In a complex system I doubt you can understand it from the requirements and so we should assume they are wrong. A test by constrast can (but need not!) be obviously correct but it is correct one for that case and so doesn't say anything about the rest.