Suppose your personal computer has 2 megabytes of memory, and you are thinking about increasing its memory to 3 megabytes. The marginal cost of increasing your computer's memory is the cost of the additional megabyte of memory you are thinking about installing. ... The marginal benefit is the benefit you will get from 1 additional megabyte of memory in your computer, not the benefit you'll get from all 3 megabytes that you will have if you add 1 more megabyte.
Now I can recall a day when the decision to add one more megabyte was an interesting one. I even fondly recall the happy childhood days I spent playing on my Commodre 64 (64 kilobytes, 6% of a megabyte).
Today, however, one megabyte more or less is excessively marginal. We wonder about adding one more gigabyte (1000 megabytes) and that is a marginal decision today. In 1995 (BC) when this book was published, buying one gig would have had almost nothing marginal about it. By 2025, will the decision to add one more gig seem as trivial?
But here's the burning question for technophiles: Do real economists make computer memory decisions bit by bit? [Correct answer worth 1 point, incorrect worth 0.]