Richard Hamming

Richard Hamming (1915–1998) was an American mathematician whose work on error-detecting and error-correcting codes laid the practical foundation for reliable digital communication. A contemporary of Claude Shannon at Bell Labs, Hamming transformed Shannon's existential proof that good codes exist into actual constructions that could be implemented on early computing machinery.

Hamming's central contribution, published in 1950, was the invention of Hamming codes — a family of linear error-correcting codes that can detect up to two simultaneous bit errors and correct single-bit errors. The codes use a clever placement of parity bits at positions that are powers of two, with each parity bit covering a distinct overlapping subset of data bits. A 7-bit Hamming code (4 data bits, 3 parity bits) is a perfect code — the spheres of radius one around each valid codeword exactly partition the Hamming space.

Hamming's work extended beyond single codes to a general methodology: the systematic use of algebraic structure to achieve reliability. He introduced the Hamming distance — the number of positions at which two strings of equal length differ — as the fundamental metric of code quality. The minimum Hamming distance between any two codewords in a code determines its error-correction capability: a code with minimum distance d can detect up to d−1 errors and correct up to ⌊(d−1)/2⌋ errors.

Hamming spent much of his career at Bell Labs and later taught at the Naval Postgraduate School. His 1962 paper 'Numerical Methods for Scientists and Engineers' and his 1997 book 'The Art of Doing Science and Engineering' articulated a philosophy of research that emphasized the importance of attacking important problems, changing one's field, and maintaining a computational perspective on theoretical questions. He is remembered for asking scientists: 'What are the most important problems in your field, and why aren't you working on them?'