I find this rather hard to believe, given that calculators predated cash registers by some 260 years. Wilhelm Schickard’s mechanical calculator (1623) could add, subtract, multiply, and divide six-digit numbers. Blaise Pascal’s more famous Pascaline (1642) could add and subtract 8-digit numbers (though not in base 10). The Pascaline attracted a great deal of attention and within a few years, everyone and his dog was building mechanical calculators. As far as I know, though, all of these early designs used dials, not buttons, to input digits.
Incidentally, the first cash register, invented by James Ritty circa 1880, also did not use columns of buttons in the manner described in the Staff Report. Registers with a column of buttons for each decimal digit did not appear until much later. Calculators, meanwhile, were already being mass-produced.
Regarding : "Why would “0” be on the bottom? Probably because the dialing mechanism was pulse, not tone. Since they couldn’t do zero pulses for 0, they did ten pulses, and hence put the 0 at the end. (Thanks to Radu Serban for this suggestion.) "
I heard another explanation for this. “0” by itself is an entire valid telephone number (the operator). The early mechanical exchanges could not respond fast enough to such a valid call. (Consider that there may have been several such calls coming into the exchange at the same time.) By making it 10 pulses it gave the exchange time to react and process the call.
I admit it sound a lot like the story about the QWERTY keyboard being designed to slow down typists and not jam the machine, but its a plausible explanation.
Also, with regards to not being able to represent ‘0’ with zero pulses; there would be nothing fundementally wrong with representing ‘0’ with 1 pulse, ‘1’ with two pulses, and so on. This isn’t so far off from today’s programming where we may have a list of 8 entries ( ‘1 thru 8’) ‘pointed to’ by the binary numbers 0 through 7.
I suppose I should have been careful to designated “the modern calculator” or “the modern popular and inexpensive electronic calculator.” We weren’t trying to trace the history of the calculator – the abacus, for instance, is certainly an ancient calculator that doesn’t use a keypad system at all. We were just to indicate why calculator keypads took one approach (based on adding machines) and telephone keypads took another.
The answer is clear if you’ve ever seen an old-fashioned calculator.
These had a button for each number 1-9 for each digit (that is, if you could use display 6 digits, there were 6 x 9 keys). Each key was set on a shaft that moved a gear. The longer the shaft, the more the gear was moved. Thus a “1” key moved the gear one place and was on a short shaft and the “9” key moved the gear nine places and was on a long shaft.
This meant the “9” stuck up more than the “1” and had to be depressed further.
Now, if 9s were at the bottom of the keyboard, they mechanism would slope away from the user. Very awkward – your wrist would touch the 9 key when you were trying to press 1. So the obvious ergonomic setup was to have the 1 at the bottom.
Calculators kept this arrangement when the started using only nine keys.
as to the zero next to the nine at the lower end of the phone:
I have an ancient rotary phone here in Sweden, and the zero is located at the upper end preceding the one. I checked out a local antique (junk) shop, and, indeed, all of the old rotary phones I found had the same feature. Now, why would the Swedish system have been so different way-back-when. Can Radu Serban help me with this? Or Obi-Wan, I suppose, he’s pretty good in a pinch.
I freely confess that I only researched U.S. telephones.
In this thread telehone dials, someone suggests that the pulse was not N pulses per digit, but N+1. That was not true of the U.S. system, but might well have been true in Sweden or elsewhere.
There was very little coordination of telephone systems in those rotary dial days – you couldn’t direct dial an international call, for instance, and one reason may have been (wild-arsed speculation here) that different countries had different pulse systems. Presumably when tone systems and touch-pads came into play, there were international treaties or conventions that allowed coordination of systems.
To call an abacus a calculator is stretching it… an abacus is really an aid to keep track of numbers; you still have to do the calculations in your head. Blind children learning addition and subtraction, for example, are taught to use an abacus rather than pencil and paper.
I suppose if we wanted to stretch things a bit further, we could call Pascal’s adding machine an early cash register, since it was designed to work with the non-decimal system French currency in use at the time. (There were 20 sols in a livre and 12 deniers in a sol.) The only thing it was missing was the actual cash box. About fifty of the machines were built; many were used by merchants and tax collectors.
The reason modern calculators have “0” (and “1”, “2” and “3”) at the bottom is because they are the most commonly used numbers (especially in accounting or banking) and require less effort to reach than the higher numbers. I am the manager of the proof encoding department in a bank, and the machines have 10-key keyboards, with a “00” key next to the “0” since it is very commonly used, too.
I tend to agree with the last statement.
Zeroes are very common.
As an aside, I’ve heard from human factors people that the telephone key orientation yeilds faster and more accurate number punching than the calculator.
The aviation products that my company sells also have the “phone” oriented numbers, for the reason above.