In 1975, just before the birth of the PC industry, Edward Tufte wrote a review of pocket calculators for the American Journal of Political Science. He summarized the rate of advance of the computer technology of the day (citing Rein Turn, Computers in the 1980s, 1974): “Studies of the historical development of computational devices report that a new ‘generation’ occurs about every six years. On average, each new generation has included an increase in speed by 10 times, an increase in memory capacity by 20 times, an increase in reliability by 10 times, a decrease in component cost by 10 times, and a decrease in system cost by 2.5 times. Not only should the calculator… be depreciated quickly, but also we should remember that the replacement for today’s machine will be twice as good at half the price in a year or two.”   

Tufte’s abstract for the article reads as follows:

“Four of the fancier and newer pocket electronic calculators are evaluated in terms of their utility for doing data analysis, their price, and their quality. Some general principles, perhaps helpful for the consumer of calculators, are derived from the experience–including: (1) Always buy at a discount. (2) Competition has its benefits in this sector of the computer industry. (3) Computational technology has overrun input-output technology. (4) Calculators are designed by engineers for engineers and business people–and not for data analysts. (5) Some quite impressive (and expensive) machines are now available. (6) If in doubt, wait for a good machine that prints.”

HT: Drew Conway @drewconway