My friend told me:
Peter thiel’s argument is that we have a lot of Innovation in bits but not in atoms. So stagnation in the broader economy instead of progress (as measured by median wages). One way to look at the lack of innovation in atoms is that when innovation in bits provides easier financial return (as shown by the internet riches) what’s the incentive fir trying the hard stuff in atoms
Won’t name him hereRelated Articles
Innovation in bits is currently happening without Trial and error. Although the early semiconductors were discovered through trial and error, and couldn’t even be modeled using quantum mechanics, we have since then built useful abstractions which ignore QM mostly (e.g. electrons vs holes and band gaps) and help us build things with less trial and error, and then top of that we have algorithms which are trivial to develop without trial and error.
Innovation in atoms is lagging because there is a lot of trial and error needed. e.g. even now the best way to make a high temperature superconductor is trial and error. We are trying to have theories about how to predict if a material is a high temperature superconductor or not, but what works well are rules of thumb. Batteries are trial and error.
Despite there being plenty of interpretations of quantum mechanics, and many theoretical unification between General Relativity and Quantum Mechanics, i think too much effort in science is spent on fundamentals of physics (with no new results for the last 40 years), and very few work is done in say condensed matter physics (for super-conductors), or plasma physics (for fusion reactors). Thankfully solid state physics (for semiconductors) has enjoyed a lot of trial and error, and very little theoretical speculation. That is why we are enjoying transistors the same size as the width of a DNA strand, and transistors so fast it can be used to count changes happening at the speed of light in physics.
Despite having an AI that can predict how a protein folds we still cannot simulate the higher abstractions (e.g. the interactome) well enough to say develop a COVID-19 vaccine in silico. This shows our “beautiful” theoretical models are perhaps just limiting cases and platonic abstractions which have no basis in reality.
This post first appeared on Me In Words | A Compendium Of Plagiarized Ideas, please read the originial post: here
