" When the genetic code was solved, in the early 1960s, it turned out to be full of redundancy. Much of the mapping from nucleotides to amino acids seemed arbitrary—not as neatly patterned as any of Gamow’s proposals. Some amino acids correspond to just one codon, others to two, four, or six. Particles called ribosomes ratchet along the RNA strand and translate it, three bases at a time. Some codons are redundant; some actually serve as start signals and stop signals. The redundancy serves exactly the purpose that an information theorist would expect. It provides tolerance for errors. Noise affects biological messages like any other. Errors in DNA—misprints—are mutations. "

― James Gleick , The Information: A History, a Theory, a Flood