89 internautes sur 92 ont trouvé ce commentaire utile
- Publié sur Amazon.com
In his role as leader of the End-Use Forecasting Group in the Environmental Energy Technologies Division of Lawrence Berkeley National Laboratory, Jonathan Koomey is professionally concerned with using numbers from many different sources to gain knowledge about where markets for energy-efficient technologies may be headed. It's an approach called quantitative problem solving.
"Although the technical aspects of this process are taught at many universities, the art of problem solving is rarely discussed and even more rarely written down," Koomey notes in the preface. His goal is to amend that lack, helping the reader "to become a first-rate analyst in your chosen field."
"Your chosen field" covers a lot of ground, and while the book delivers on Koomey's promise, much of the pleasure of reading it comes from his eccentric definition of both his topic and his audience. Not a textbook -- or not just a textbook -- "Turning Numbers into Knowledge" is aimed at students and professors alike; at problem solvers in business, government, and research; at middle managers and potential investors; and even at journalists.
Beginning with a chapter titled "Beginner's Mind" and including others titled "Question Authority" and "Reflect," one might think Koomey's book could have been named "Zen and the Art of Statistical Analysis." But it's also full of technical advice, in chapters like "Let Tables and Graphs Do the Work," "Use the Internet," and many more.
In fact Koomey has organized his topics thoughtfully, beginning with considerations of why anyone, professional or amateur, would undertake quantitative analysis. Koomey's bottom-line answer is that it helps us manage a runaway information explosion, which otherwise can lead to paralysis through overload or the opposite, a refusal to learn anything new. He then moves on to the mental preparation needed for problem solving, consideration of common pitfalls, critical thinking, and finally to nitty-gritty techniques.
In the first chapter, "Beginner's Mind," Koomey reminds the reader that "experience is a two-edged sword. It eliminates unnecessary detail," but it can also "lead you astray when a new problem is sufficiently outside your experience." What's needed, he says, is a combination of curiosity -- the nonjudgmental observation of the beginner -- with the experience of a senior analyst.
He makes his point with Bruce Lee's amusing parable of the Zen master and the Western university professor who came to inquire about Zen but never stopped talking about what he already knew. During a pause in the monologue, the master suggested tea. He poured his visitor's cup full and kept on pouring and pouring until finally the professor protested: "No more will go in!" Noting that professor's mind was like his tea cup, the master asked "How can I show you Zen unless you empty your cup?"
Later, in the chapter "Question Authority," Koomey allows that sometimes following authority is essential -- in military maneuvers, for example -- but that most life situations aren't like that. Appeals to anonymous authority are immediately suspect; so are expert pronouncements by experts in an unrelated field, or those with a vested interest in the outcome. Koomey cites the "expert" testimony of researchers funded by tobacco companies who found no evidence that smoking had adverse health effects. He warns that while information on the internet is particularly susceptible to a lack of institutional validity, "any source can propagate nonsense."
In an exercise from a subsequent chapter, "How guesses become facts," Koomey suggests that the reader "Find an official statistic that sounds plausible and explore its origins. Do you still find it plausible after you've investigated?" It's an exercise Koomey himself has undertaken more than once, notably with the 1998 claim that the internet was consuming eight percent of total U.S. electricity production and was well on its way to consuming half. Koomey discovered that the numbers were based on bad measurements, bad assumptions, and outright guesses -- none of which prevented them from becoming enshrined as fact before Koomey's team did their best to set things straight.
In summing up, Koomey lists a dozen lessons to remember, among them "Don't be intimidated," "Get organized," "Document, document, document," and "Remember that others don't care as much about your work as you do" -- an invitation to know and persuade one's audience. It's sound advice, cheerfully and colorfully delivered by a man who knows what he's talking about, and of value to all sorts of readers -- whether or not they intend to ever crunch a single number.