The Acme code consists of one hundred thousand five letter codes each intended to stand in for a phrase. It was designed to be tolerant of transposition errors; the author claims that "no transposition of any two adjoining letters will make another word in the book". However, as later discovered by J. Reeds, the code did not provide this level of error correction, containing at least eleven pairs of words differing only by the transposition of two letters.3 Despite these errors, this code is a precursor to more modern error correction codes.4
Newton, David E. (1997). Encyclopedia of Cryptology. Santa Barbara California: Instructional Horizons, Inc. p. 4. ↩
Kahn, David (1996). The Codebreakers. Simon and Schuster. p. 274. ISBN 978-0-684-83130-5. 978-0-684-83130-5 ↩
Bellovin, Steven M. (27 July 2011). "Compression, Correction, Confidentiality, and Comprehension: A Look at Telegraph Codes" (PDF). Retrieved 17 November 2018. /wiki/Steven_M._Bellovin ↩
Simmons, G.J. (May 1988). "A survey of information authentication". Proceedings of the IEEE. 76 (5): 603–620. doi:10.1109/5.4445. /wiki/Gustavus_Simmons ↩