The test of a good procedure is how well it works, not how well it is understood.
- John Wilder Tukey
The notion that processes must first be understood before being useful is just that notional. This approach seeks the lowest common denominator for those using a procedure or a process. It's the race to the bottom, or it's the dumbing down of our work force.
Tukey reminds us:
No catalog of techniques can convey a willingness to look for what can be seen, whether or not anticipated. Yet this is at the heart of exploratory data analysis. The graph paper - and transparencies - are there, not as a technique, but rather as recognition that the picture-examining eye is the best finder we have of the wholly unanticipated.
My first "real" program I ever wrote from scratch by myself was a the Cooley-Tukey Fast Fourier Transform algorithm. It was in Macro-8, running on a PDP-8M, with 4K Words of Core memory. Yes, that's correct 4K. What the Fast Fourier Transform (FFT) did was sample analog signals, convert that sample stream through the Fourier Transform into a "spectrum" of frequencies, and store this accumulated signal specturm in memory. This is the basis of digital signal processing used in everything from music recording to cell phones. Of course today this is all done at the silicon level.
The message here is that seemigly complex process are effective for the purpose. The FFT algorithm source code has long been lost, but if it were here today, in MACRO-8 format, printed on "green bar" paper, I doubt anyone on the planet would be able to figure out what it did. This algorithm was transferred to the PDP-15 and rewritten in FOCAL.
The result of this effort was a Spectrum Analyzer for signals coming off laser back scattering experiments - Mie Scattering. Which, along with Rayleigh Scattering is why we see rainbows, cloud colors, and other atmospheric colors.
