Colloquia

My Ph.D. in Physics followed an applied mathematics undergraduate degree, and my interest in computing started with summer undergraduate experiences punching holes in cards and carrying the daily jobs from Cambridge to London to the IBM 7094 data center. My Ph.D. adviser wisely told me, “Geoffrey, nobody looks at data.” In those days, High Energy Physics was the Big Data leader, and I taught myself statistics from DOE laboratory memos and Kolmogorov. Some of my papers, such as “Veni, Vidi, Vici Regge Theory,”  described large-scale Regge+Resonance data analyses. I learned modeling from Feynman, worked with the particle data group at LBL, and participated in experiments to get more reliable data. Meanwhile, working with Wolfram on the early version, SMP, of Mathematica, I stumbled into computer science through parallel hardware, software, and algorithms for Quantum Chromodynamics and other simulations. However, I was always termed the physicist, and my training in this field was very useful as it taught me about systems. Computing is itself a complex system that produces discoveries in other complex systems. Since then, computing has continued to advance through the Internet and distributed systems and, most recently, Artificial Intelligence, AI, and a situation where “Geoffrey, everybody only looks at data.” Every ten years, I have noted that computing has been successful, but the next ten years look even more promising. Today, AI for Science has incredible promise, and I finish with some of these opportunities.