Early in my academic career, John H. Anderson (JHA) was a pivotal figure who shaped my intellectual trajectory. I first took his undergraduate control systems class at the University of Melbourne, probably in 1976 — an experience that ignited my fascination with the field. Shortly thereafter, I worked for him as a research assistant (1981-2, see note below) and later stepped in as a lecturer (1983) to teach the courses supported by his research grants. His rigorous, foundational approach to dynamic systems and modern control left an enduring imprint on how I think about engineering and served as an important catalyst for my subsequent career.
Anderson brought to academia the sharp pragmatism of someone who understood real physical machinery, having worked in the power generation industry before pursuing doctoral research. That industrial grounding was evident throughout his postgraduate work in the mid-1960s at the University of Manchester. His doctoral research—completed around 1967 and awarded through the University of Manchester—focused on the high-order dynamic modelling and simulation of large natural-circulation drum boilers and turboalternator units.
During that intensely creative period, Anderson was situated at the very epicentre of the British multivariable control revolution. Based at the University of Manchester Institute of Science and Technology (UMIST), he joined the newly founded Control Systems Centre in the Ferranti Building under Howard H. Rosenbrock. While there, Anderson tackled the challenge of bridging theoretical state-space formulations with large-scale industrial reality, authoring pioneering papers on dynamic optimisation and model order reduction. He was also closely linked to the vibrant control group at the University of Sheffield, co-authoring key work on optimal linear control with H. Nicholson—the same department where David H. Owens would soon establish Sheffield as another major powerhouse of British multivariable control.
By late 1968, Anderson moved to Canada, joining the Department of Electrical Engineering at the University of Waterloo in Ontario. Over the next decade, he built an active and highly productive research program, making major contributions to deadbeat control, synchronous machine dynamics, and high-speed excitation systems. His papers appeared regularly across leading IEEE and IEE transactions, cementing his reputation as an international authority on power plant dynamics and multivariable frequency-domain methods.
In the late 1970s, Anderson moved to Australia to take up the Chair of Electrical Engineering at the University of Melbourne. He brought with him deep international perspectives and the cutting-edge multivariable control philosophies that had emerged from Manchester, Sheffield, and Waterloo. For those of us fortunate enough to learn from him during those years, his command of dynamic systems and uncompromising analytical rigour provided an extraordinary masterclass in the power and elegance of modern control.
How I got that research assistant job. I was hanging out in the lab which had (for the time) an awesome computer — a PDP11/60 running RSX-11M. All the documentation was in the lab in huge binders, and I pored over all that, particularly system calls. I wrote a simple trojan horse (probably in Fortran) that mimicked the login dialog, and ran it on the terminal that JHA habitually used. I got his password easily, became a privileged user, got the password file (plain text!), printed it out and stuck it under his office door. He figured it was me and offered me a job.