It is occasionally useful to reexamine the motivating questions and paradigms behind our research activities. In this student-oriented seminar, we'll fly above the methods' details and discuss the issues that compel us to address uncertainties in computational modeling. Think: less math and more stories. I won't promise that we'll answer any questions satisfactorily. In fact, if all goes well, you'll leave the 50 minute discussion pondering, "How much can I trust computer models of the so-called real world to make accurate predictions and inform critical decisions?" This seminar will combine the first two lectures from a graduate course on uncertainty quantification I taught at Stanford. We'll have some open discussion, and I'll show an end-to-end example of predicting the time it takes a ball dropped from the top of a tower to hit the ground.