Not accidental

I was reading a defense of the Canadian system of postal codes, which is part alphabetic and part numeric. One example is X0E 0R8 for a region in Northwest Territories. The writer claimed that part-alpha is easier to remember.

Nope. When our post office established Zip Codes, they chose 5 numbers. As far as I can tell the system evolved from a WW2-era system of internal three-digit codes. Possibly through natural influence, they arrived at the same pattern as Bell’s long-lasting 5 digit phone system.

BELL KNOWS PERCEPTION. Bell Labs was THE center of perception research for 60 years until it was killed by Share Value and LBOs like every other good thing in the world. Research started with speech and hearing, for a simple economic reason. Circuitry is easier to design and cheaper to build if you can restrict the range of needed frequencies.

For instance, Bell found out that speech can be easily understood when the phone only carries a range from 300 to 3000 cycles. This was not intuitively obvious. Actual speech runs from about 100 to 10000 cycles, and a healthy ear responds from 30 to 20000 or so. The basic pitch of an adult larynx is 100 to 200. Intuitively you’d think that a system which excludes the main pitch couldn’t work at all. Bell understood that our perception doesn’t use the main pitch, only the harmonics or formants.

Every physical object works best in a preferred range of frequencies. If you can restrict the needed input, you have a better chance of staying within the preferred range of the cables and insulators and capacitors and tubes. Lower frequencies require bigger components, so ‘pulling up the floor’ on the range saves materials and space. (Superhet radio also relies on this type of efficiency.)

After Bell established a pool of researchers and technicians and experimental methods, they used the talent to design other aspects of the system so it would work best with human muscles and senses. Everything was checked against human needs. The shape of a handset, the size and friction of dials, and the layout of numbers. Earlier phone exchanges often had an irregular mix of numbers and letters like 307X and 1163R.

Letters and numbers trigger different parts of the brain, and ‘words’ of different lengths also trigger different types of responses. Memory works best for a fixed-length series of one type, and Bell found that five was optimal. Not accidentally, a fixed-length sequence is ALSO easier for the mechanical parts of the dial system to handle. The system knows it can stop watching after 5 inputs.

The earlier Strowger dial system, without benefit of perception research, had a variable number of digits plus a button on the phone to say “done”, similar to the Enter key on a computer.