Summary and Info
problem 1 is that the authors obviously aren't developers. they're a couple of guys who like sql puzzles. much of the code in this book has little or no practical value. you should not use it in production systems because it will not perform well and will be difficult to maintain. it's like they're trying to show off some way of getting a complex result set with a single select and not even thinking about the performance impact of the code or its maintainability. that kind of stuff is cool to newbies, but real engineers know better.
problem 2 is the astoundingly bad advice throughout. the complex numbers discussion doesn't belong in the book and you should not try to implement complex number handling in t-sql unless you want to be out of a job. much of the advice in the book revolves around cool coding tricks that aren't applicable to any real world problem. who cares if you can do something with a single select that most people would take two statements to do? does the code perform better? have you considered what it will be like to tune and maintain? how readable is it?
bottom line here is that the authors are amateurs who really shouldn't be writing a book to instruct others in t-sql. they need to learn some other languages and get beyond the "cool coding tricks" stage to true engineering.