Arrgh! I Goofed Again!

We've already said that it is acceptable to make mistakes. If you're a manager reading this, don't rip thischapter out before giving it to your programmers. The fact is that 'trial anderror' is often a good way to learn a new technology – once you've crashed theserver for the third time in 20 minutes, you will probably have worked out whatit is you are doing wrong. OK, if you're writing monitoring software for ahospital intensive care unit, your quality goals might be a little stricter –but on your own, while learning, you will make mistakes. Guaranteed. You willlearn from them; you won't make them again, and you'll go on to write booksthat tell people how to debug applications.

So what this chapter is really about is how to write codethat, while not exactly bulletproof, is a lot more robust than it mightotherwise be. When the worst does happen, we talk about how you track downthose errors quickly and efficiently so that you can get to the root of theproblem and fix it, or at least find sources of information when the errorstill mystifies you.