In computational seismology we've long had our own creation story for the FFT: "The first machine computation with this algorithm known to the author was done

In computational seismology we've long had our own creation story for the FFT: "The first machine computation with this algorithm known to the author was done by Vern Herbert, who used it extensively in the interpretation of reflection seismic data. He programmed it on an IBM 1401 computer at Chevron Standard Ltd., Calgary, Canada in 1962. Herbert never published the method. It was rediscovered and widely publicized by Cooley and Tukey in 1965." [1] [1] Claerbout, J., 1985, Fundamentals of Geophysical Data Processing, p. 12.