> As theories go I like it, but I don't think the headline and conclusion are supported by the article.
Right. What the author doesn't really pick up on is how engineering developed.
Before 1800 or so, science and math were not routinely used by people who built stuff. Edison tended to deride mathematical theorists, although he hired some.
Early navigation and astronomy required math, but that was mostly measurement, not design. The first thing that really required math in the engineering was the steam engine. It's possible to build a steam engine with no understanding of thermodynamics, but it will be a really crappy steam engine. Watt's early engines were about 2% efficient. Carnot, in 1824, published "Reflections on the Motive Power of Fire", which, for the first time, addressed engine design in terms of heat, pressure, efficiency, heat transfer, etc. Around the middle of the 19th century, Joule and Kelvin finally got a decent mathematical theory for heat engines. Engine designs improved substantially and efficiencies went up. At last, direct use of math in engineering was really paying off.
Early electrical work was trial and error. Sort of. Edison publicly derided theorists, but he had some on his payroll. When AC came along, more theory was needed. Now calculus was required. Westinghouse and Tesla got involved, and generators, motors, and transformers started working much better. (I've pointed out before that Tesla's real contribution was that he figured out how to get AC electric motors started.) Good motors require electromagnetic fields in strange shapes, and intuition fails for that. You need math.