Program Result-Checking: A Theory of Testing Meets a Test of Theory

We reuaew the field of result-checking, dascussang simple checkers and self-correctors. We argue that such checkers could profitably be ancorporated an software as an aad to eficaent debuggang and relaable functaonalaty. We consader how to modafy tradataonal checkang methodologaes to make them more appropraate for use an real-tame, real-number computer systems In partacular, we suggest that checkers should be allowed to use stored randomness: a e., that they should be allowed to generate, pre-process, and store random bats praor to run-tame, and then to use thas anformataon repeatedly an a serzes of run-tame checks. In a case study of checkang a general real-number lanear transformataon (for example, a Fouraer Transform), we present a sample checker whach uses stored randomness, and a self-corrector which zs partacularly eficaent af stored randomness as allowed