So, we've gotten roughly halfway through this book, and not once have we defined any sort of tests, or ensured that the programs that we have written are verifiably correct. Up until this point, you've taken my word that the programs I have shown you do everything that I've said they do. But how can we guarantee that they give us the same results every time regardless of how many changes we make to them?
This is where your testing strategy comes into play.
In professional software development, testing your software to try and limit the bugs is one of the most important things you can do. All great software developers implement a decent testing strategy surrounding the systems they build, and this, actually, enables them to make changes faster and with more confidence.
Say we had a legacy system that had 100,000 lines of code, and had no test suite and no testing strategy implemented. How would you test that what you were doing wasn't breaking something further down the chain? How could you confidently say that a code change you implemented wasn't going to bring down X, Y, and Z other applications in production, and, potentially, cost your business money? The answer is it's next to impossible; every change you make will make you nervous when the change is deployed, and you'll be on support for potentially catastrophic breaks 24x7.
Conversely, say you were in charge of developing a new feature on a legacy system that has 100,000 lines of code. If you made any changes to specific parts of the code base, the suite of tests that your team has built up would catch any potentially catastrophic breaks, and you would be confident that this new feature would be able to go into production without impacting anything existing. This is a huge advantage for development teams that follow an agile methodology, and iteratively implements lots of changes to their software systems. It also means that the chance for your business to be impacted by an issue in production is far lower, and you don't have to worry about being on support all the time.
