I’m an Australian nerdy/sporty/outdoorsy guy living in Boston. I love being surrounded by so many smart people who are deeply passionate about their own fields of expertise. It seems that everywhere you go around here you can bump into a world expert in something and I think that’s exciting. Better still, I get to share a home with one of the most fascinating and loving of the smart people, Amy, who I moved here and married.
Growing up I was always interested in science, machines and computers. My physics teacher Kym Lawry inspired me to enrol in a physics degree at the University of Melbourne, this eventually morphed into a major in applied mathematics.
The Running Part
In the midst of this I got very caught up in long sprinting. Training under Peter Fortune alongside Australia’s best athletes, Olympic champion Cathy Freeman, world indoor champion Tamsyn Llewis and commonwealth bronze medalist Kris Mccarthy. Unfortunately my impressive stablemates didn’t lead me to such stratospheric performances, though I was South Australian mens champion of the 800 meters and competed at the national level. What I found most interesting was the chance to observe the positive psychology of these elite performers. Perhaps not coincidentally the psychology of performance is also a central topic of my wifes book, Presence. Overall, I learned that it’s fun to test your physical limits and be part of a group who love what they do.
The Mountain Part
The next part of the story actually starts much earlier in 1988 when I first visited the French Alps on a summer road trip from england where we were based for my dad’s sabbatical. I was only seven but it’s safe to say it cut deep and I was dragged away in tears. Since then I’ve had a mostly irrational love of the Chamonix valley and being in the mountain landscape in general. So after finishing my degree I spent six months climbing, hiking and washing dishes in Chamonix. So still today the valley keeps calling me back and I’ve spent almost two years living there, and much of the rest being in the mountains and training for climbing. I dream of one day becoming a certified climbing guide but for now my goal is to just climb as well as some of the 12 year olds in the local competition squad.
The Computery Invention Part
One of my earliest memories is drawing elaborate diagrams of machines to travel across the beach on a family holiday. I imagined springs and and suspension arms and all kinds of wheel counts and configurations. I’ve enjoyed inventing new solutions ever since. Moving into new domains of knowledge has given me the chance to invent under several new sets of constraints, axioms and opportunities. Over time built up a toolkit of analytical skills based on a degree in mathematics that taught me how to open my mind, think critically and suspend my assumptions. A degree in computer science got me started on the creative process of developing software and combined with my mathematical background, gave me a sense of power to solve some cutting edge problems.
My honors year took me into the weird world of competitive sorting! When I told people that my honors project was on string sorting they either didn’t get it at all or thought that sorting was a solved problem. On the contrary, the cache hierarchy of real computers means that Cache-efficient algorithms like burstsort could crush their hardware-ignorant competitors. It showed me that there are many layers of abstraction in algorithm design and that optimisation of algorithms is multidimensional. Sorting is a surprisingly useful thing to be able to do efficiently and it was fun to have such a sharply focussed project to really drill down into. I enjoyed inventing a multicore version of the state of the art algorithm that gained advantage by dividing the input data across the cores, sorting, then redistributing the sorted data across the cores and re-merging across all cores. (include honors thesis here).
After that I moved to an astrophysics PhD program creating some cool software to detect signals from radio telescopes. The bulk of my work was around developing a software package to search for signals from undiscovered pulsars (binary pulsars ideally). This is a (semi)periodicity detection problem with a large dataset, weak signals, multidimensional search space and several other nasty tricks to trip you up. By harnessing GPU power it was possible to detect signals of the type that would be emitted from binary pulsar systems. The hope is that these systems can be used to test the limits of relativity. The software is now being developed and maintained by Ewan Barr.
This work showed me the some of the limitations of detecting periodic signals using fourier transforms, (harmonics!!) and amidst the delirium of one isolated observing run the idea of a new fast folding algorithm formed. This would have some of the benefits of recycling precalculated values while having the more easily read (by human or machine) output of folded data. This led my interest away from the work of my PhD and I packed up and moved to Boston.
Since moving to Boston I've been working on several personal software projects across domains ranging from email assistants to auction software to automated gardening. These projects have allowed me to play around more with the AWS cloud computing platform and write more in my favorite non-C++ programming language Python.