Dilemma of the Day: predicting the future.

I am reading Blue Ocean Strategy, by W.C. Kim and Renee Mauborgne. It's a good book, full of interesting business intuitions and anecdotes, but contradictory statements, like the two below, leave me sometimes puzzled:

We're not talking about predicting the future, something that is inherently impossible. p.76.

Visualizing strategy can also help managers responsible for corporate strategy predict and plan the company's future growth and profit. p.96.

On one hand, the future is inherently unpredictable; on the other, here's a visualization tool that helps you do exactly that. Huh?

I guess they mean that future is probabilistic. Some predictions can be made with a greater degree of confidence than others. For example, I can rather confidently predict that 50 years from now, unless humanity perishes in a cataclysmic event, we will use electricity to satisfy our basic energy needs. We will use money in commercial transactions. We will have developed genetic techniques to treat diseases that seem to be untreatable today. Etc, etc.

On the other hand, I can't predict with sufficient accuracy whether Apple as a company will survive these 50 years; who who will be its new CEO, and what the change in leadership is going bring. My hunch, though, is that he or she will still be human and make contradictory statements like the ones in the Blue Ocean book.

Fernando J. Corbato in his 1991 Turing Lecture On Building Systems That Will Fail said:

What I am really trying to address is the class of systems that for want of a better phrase, I will call "ambitious systems." It almost goes without saying that ambitious systems never quite work as expected. Things usually go wrong and sometimes in dramatic ways. And this leads me to my main thesis, namely, that the question to ask when designing such systems is not: "if something will go wrong, but when will it?"

All inventions are, to follow his definition, seeds of ambitious systems.

need to work through this paper to extract good complexity/scalability problems