A Data Science Central Community
Interesting article published in IEEE Spectrum.
The queen of the sciences may someday lose its royal status
By ROBERT W. LUCKY / MARCH 2012
Illustration: Robert Neubecker
Long ago, when I was a freshman in engineering school, there was a required course in mechanical drawing. “You had better learn this skill,” the instructor said, “because all engineers start their careers at the drafting table.”
This was an ominous beginning to my education, but as it turned out, he was wrong. Neither I nor, I suspect, any of my classmates began our careers at the drafting table.
These days, engineers aren’t routinely taught drawing, but they spend a lot of time learning another skill that may be similarly unnecessary: mathematics. I confess this thought hadn’t occurred to me until recently, when a friend who teaches at a leading university made an off-hand comment. “Is it possible,” he suggested, “that the era of mathematics in electrical engineering is coming to an end?”
When I asked him about this disturbing idea, he said that he had only been trying to be provocative and that his graduate students were now writing theses that were more mathematical than ever. I felt reassured that the mathematical basis of engineering is strong. But still, I wonder to what extent—and for how long—today’s undergraduate engineering students will be using classical mathematics as their careers unfold.
There are several trends that might suggest a diminishing role for mathematics in engineering work. First, there is the rise of software engineering as a separate discipline. It just doesn’t take as much math to write an operating system as it does to design a printed circuit board. Programming is rigidly structured and, at the same time, an evolving art form—neither of which is especially amenable to mathematical analysis.
Another trend veering us away from classical math is the increasing dependence on programs such as Matlab and Maple. The pencil-and-paper calculations with which we evaluated the relative performance of variations in design are now more easily made by simulation software packages—which, with their vast libraries of prepackaged functions and data, are often more powerful. A purist might ask: Is using Matlab doing math? And of course, the answer is that sometimes it is, and sometimes it isn’t.
Read full article at http://spectrum.ieee.org/at-work/education/is-math-still-relevant