We've been through some pretty tough sledding in the last two parts. Both the project- management view and the human-element chapters asked that you think hard about some of the thornier parts of our craft. In this section, there's a bit of a respite. Before I work up to the crescendo of the symphony in Part 6, "Advanced Topics," I have a quiet interlude for you.
The theme of this section is that software people think differently. Some of their ideas are speculative, maybe a little off the beaten track. It's helpful to gain the insight of seeing some of the various ways in which they look at problems. Much as sports have been referred to as the "toy department of life," you can take this section as the toy department of this book.
In Chapter 17, "History Lesson," I discuss an interesting episode in naval architecture. The sinking of the Swedish warship Vasa on its maiden voyage centuries ago can provide some valuable lessons for software developers.
Chapter 18, "Bad Analogies," confronts the issue of pseudoscience. Using bad scientific analogies to "explain" software to non-practitioners gets us into trouble, so this topic is worth exploring.
In Chapter 19, "The Refresh Problem," I look at an interesting problem in a new lighthow should we go about updating software in an installed base? The answer I propose may surprise you.
Finally, in Chapter 20, "Not So Random Numbers," we have our last visit with Roscoe Leroy. Therein he relates to us some calculations he once made on a desert island, without the benefit of any computers. This tale shows us how to analyze a problem that starts out simple but turns out to be more complex than we thought.
