My previous posting on engineering vs. science reminded me of a professor who insisted “real engineers” had to be able to reason things back to “first principles.” If you couldn't, in his opinion, you were just a technician pushing buttons on a machine you had no business operating. I've always thought he made a valid point. Yet there are many things we do that are not fully understood from a scientific standpoint. I’m not talking about people who can’t resolve forces with a free body diagram and vector arithmetic, or “gear engineers” who are lost without their favorite computer programs. Our AGMA committees are charged with issuing standards, writing information sheets, and answering questions that arise from them. They don’t have to scratch too far below the surface to find things that are difficult to reason back to “first principles.” The late Don McVittie famously warned committee members “In God we trust, all others bring data,” but he was fifty years too late. Lots of negotiation and commercial concerns have been baked into standards around the world. We have too much invested in the current rating system to completely discard it, and lack the resources to significantly improve it. At one point it was assumed finite element analysis would evolve to the point where a completely new and scientifically “bullet proof” rating system could be introduced and universally accepted. Call me a cynic, because 30 years later we can hardly agree on what constitutes an accurate FEA model of a tooth. Perhaps the FEA gear rating method will be delivered in a flying car. I am not without hope though. A few days ago I got to watch a new parts scanner analyze a bullet to determine if it were fully packed with the right size and shape of gun powder. Who saw that coming?