No silver bullets – you can’t radically fix the test cycle

The earliest mass-market book about software engineering to still be in print is probably The Mythical Man-Month, by Fred Brooks. Published in 1975, although the book’s language and tone may seem a little archaic, it contains three ideas that are still relevant today. The first is the idea that adding staff to a late project will make it later. As projects grow, the number of potential communication paths between people grows exponentially, shifting the cost from effort to communication and coordination. We’ll cover how to keep those costs light and cheap in the next section. Secondly, Brooks suggested the idea of the second system effect, which is the idea that we don’t know what will be good for the customer until we’ve made a first, unsuccessful attempt at it. Rapid prototyping and other measures fell out of this. As Brooks put it, “Plan to throw one away – you will...