Learning to Love Data Science

Preface

I first heard the term “data science” in 2011, during a conversation with David Smith of Revolution Analytics. David led me to Drew Conway, whose data science Venn diagram (reproduced with his permission in Figure 1-1) has acquired the legendary status of an ancient rune or hieroglyph.

Like its cousin, “big data,” data science is a fuzzy and imprecise term. But it gets the job done, and there’s something appealing about appending the word “science” to “data.” It takes the sting out of both words. As a bonus, it enables the creation of another wonderful and equally confusing term, “data scientist.”

Confusing is the wrong word. Redundant is a better choice. Science is inseparable from data. There is no science without data. Calling someone a “data scientist” is like calling someone a “professional Major League Baseball player.” All the players in Major League Baseball are paid to play ball. Therefore, they are professionals, no matter how poorly they perform on any given day at the ballpark.

That said, the term “data scientist” suggests a certain raffish quality. Indeed, the early definitions of data science usually included hacking as a foundational element in the process. Maybe that’s why so many writers think the term “data science” is sexy—it conveys a sense of the unorthodox. It requires ingenuity, fearlessness, and deep knowledge of arcane rituals. Like big data, it’s shrouded in mystery.

That’s exactly the sort of thinking that gets writers excited and drives editors ...