Error detection: parity, checksums and CRC

In 1947 Richard Hamming had weekend access to a relay computer at Bell Labs. He left a long job running on Friday evening, and on Monday found that the machine had detected an error early on and simply abandoned the work. If it could detect that something was wrong, he asked himself, why could it not work out where, and fix it?

The answer, published in 1950, founded coding theory. Networks mostly use the cheaper half of it: detect the damage, throw the frame away and let someone resend.