3 ms·
The problem is, we need confidence that the software is correct in order to trust the scientific results.
by lalaithion 6y ago
The problem is, we need confidence that the software is correct in order to trust the scientific results.
- chrisseaton 6y agoDo you need tests to do that? Are tests the only way? Are you doing formal verification of your software? Why not? If you're not bothering to do that why are you criticising researchers for not bothering to use tests?