Research Converts Money to Knowledge; Engineering Converts Knowledge to Money

The piece opens with a viral article questioning why a professor's motorcycle engine only exists in PowerPoint, highlighting a tension between academic research and practical engineering.

Jeffrey Nicholson of 3M frames the distinction sharply: "Research is the process of turning money into knowledge; innovation is the process of turning knowledge into money." Research spends money to gain knowledge, while engineering applies knowledge to generate returns.

Engineering is described as pragmatic, short-sighted, and return-driven. It solves concrete problems with the most mature technology available, rarely requiring theoretical breakthroughs — only context-specific adaptations and trade-offs. Research culture, by contrast, should be idealistic, targeting fundamental, long-horizon problems such as deep learning algorithms or quantum computing.

The author observes a role reversal: traditional research institutions are becoming more engineering-oriented, while companies like Google and Huawei are investing heavily in basic research. Excluding pure science (mathematics, physics), most engineering disciplines cannot advance without practice; they cannot thrive in an ivory tower.

Citing Antifragile (referenced as "Asymmetric Risk"), the article likens knowledge detached from practice to Antaeus lifted from the earth — powerless. Most "knowledge" we attribute to scholars was actually forged by craftsmen through trial and error; scholars merely recorded and named it. Knowledge gained through exploration, failure, and direct experience is more valuable, reliable, and sophisticated than knowledge acquired by reasoning, memorization, or study alone, because the latter severs us from the ground of practice.

Software engineering is explicitly called a practice-oriented discipline. Philosophical debates about its values are deemed worthless; the core requirement is to get your hands dirty, run experiments, and practice. The warning against "engineering scientism" (工科理科化) — substituting mathematics and models for real-world practice — concludes the argument: working in an idealized world of symbols ultimately produces useless abstractions.