Hardware is advancing to be so user-friendly that I sometimes feel more like a child with an advanced Lego set than a designer. If innovation is a step forward, this feels, at best, like a shuffle in a slightly forward direction. Effective software, however, seems to be moving deeper into higher levels of math. While engineers aren’t mathematicians, it is often these equations that make the world around us work. I was speaking with Stefanie Bernosky recently, a geo and data scientist, and she talked about using statistic analysis in her models. I realized that even in my engineering career, I have mainly stuck to pre-derived, or more static-like, equations. Worse yet, I simply open the Excel or MathCAD file I used previously and switch some numbers around.
It seems as though after being largely a mainstay, math is now being removed or minimized in many career paths. But it should be remembered that math is a way to understand patterns in nature. In the digital age in which we live, it seems we are not only ready, but that it is necessary, to go to the next level—finding patterns or structure in our math. When dealing with statistical learning, data structure, and FEA-type applications, classical calculus can seem limiting at times. Some professors in Europe are moving toward encouraging matrix theory and linear algebra along with calculus. When working in simulations or numerical analysis, having a range of mathematics can offer dynamic and perhaps more accurate models.
If engineering is reduced to just learning software and engineers don’t understand the theory behind it, we can’t be sure of the risk we are putting in our models. If we have some experience, we can say it passes the “smell test”—it seems right—but the software could have compounded uncertainties that make it more guesswork than engineering. Experience is good, but not if it doesn’t make you question what’s behind the button because that’s the way the company has always done it. Or maybe you think you will make a profit, so it’s good enough. If the United States continues down such a path while Europe advances its math program, we might be left wondering why we can’t seem to reduce cost or be as accurate as other countries.
Personally, I’m still 100% pro-hands-on. However, I feel I must say that there is a time to tinker and a time to hone. Tinkering can get you started, but to have staying power in the market, you need those math skills. In addition, when your new algorithm creates a model that allows your startup to outperform the competition, go to a local school and tell the students that you don’t have to know this math stuff to exist. But if you want to thrive, you should pay attention.