Soft skills are important, but everyone knew that already. Each field will have certain skills that are fairly permanent that are things that distinguish it fro

Soft skills are important, but everyone knew that already. Each field will have certain skills that are fairly permanent that are things that distinguish it from other fields. In computing, these things are understanding computer architecture, knowledge of common algorithms, etc. However there was one item that stood out for me in the article: "The ability to distinguish “temporarily out of favor” from “wrong." The only way to do this is with technical insight born of study and practice of engineering fundamentals and also knowledge of history. History is something you can study and can learn a great deal from. Everything that people say is the new hotness came from somewhere and nearly always is a modern reflection of something that has happened before. It doesn't give you the power to predict how things will happen, but you can often predict the range of possibilities, and it's usually not what is being hyped. The hype is usually driven by profit seeking and careerism, and so the predictions are partly hollow. I knew a guy that was working on CPU algorithms when GPUs were becoming extremely hot. I was concerned that he was working on something that would become outdated too quickly. He told me that these things go in and out of fashion over a period of years, and to a certain degree he was right. Around that time Intel came out with Knight's Landing, Knight's Corner, etc and AVX-512. While it seems like that architecture is not exactly panning out popularly, it did show that general purpose CPU architectures were able to produce the kind of FLOPs that were competitive. It makes me think the cycle will continue though GPUs continue to race ahead.