The standalone calculator may seem like a quaint holdover from the pre-PC era, but that's not the case at all. Two recent developments linked the past to the present, and the future. First, Hewlett-Packard kicked off a 40th-anniversay celebration of the introduction of the HP-35 scientific calculator ( http://h20331.www2.hp.com/Hpsub/cache/457008-0-0-225-121.html). This unit, still revered and regularly used by “old timers”, revolutionized the industry in many ways, to invoke that overused clich, and soon killed the slide rule industry.
Technically, it was the first pocket-size scientific calculator to implement transcendental functions, using a very skillful and clever implementation of the CORDIC algorithms (relies on shift and add operations only, no multiplications needed) supported by brilliant circuitry and packaging design. Even at $400 (real money in those days!) it was a bargain. (When my time to buy came, I preferred the TI SR-50 calculator of 1974, ( http://smithsonianchips.si.edu/texas/t_025.htm), which was only $200 and did not use Reverse Polish Notation for data entry.)
You might think that for today's students and engineers–surrounded by PCs and multimedia, multifunction cell phones loaded with basic arithmetic functions as well as spreadsheets and analysis software–the calculator may seem obsolete. But that's not the case. Evidence for this is the otherwise tepid second-quarter 2007 financial results from Texas Instruments, which contained an interesting sentence: “[orders saw] a $50 million increase in demand for calculators.” Note that since today's units retail for far, far less than that HP-35, we're talking about a very large volume of units.
Why the continued, strong demand for calculators? It's another case of an often-repeated but usually ignored lesson: end-users want simplicity, ease of use, instant turn-on, and operating consistency. Calculators do a few things, they do them well, they don't have battery-life issues, they don't have interface and connectivity issues, they don't crash, they have few or no bugs, and they don't try to be too much. They are basic, focused, and friendly. You turn them on, and they are ready to go. They don't try to get ahead of you, by anticipating what you are trying to do, and then doing for you what they (incorrectly) infer you trying to do.
Unfortunately, the tide of software- and flash-memory-based products runs counter to the calculator model. It's just so darn easy to add more almost free, software-based features which bewilder users. Worse, we can let unwary users be the beta testers of our products. After all, they can download updates and so-called enhancements which will cover any misgiving we have about releasing products with bugs. Then we can rest easy with a soothed conscience.
-x x x-