Jason Robert
Carey Patterson
BCompSc (Hons), PhD
(see FAQ and computing background)
Email: jason@lighterra.com
(with PGP key)
Computer scientist with interests centered around performance and the hardware/
Experience
Dr Patterson is an enthusiastic, natural-born programmer, self-taught from the age of 12. Over the years, he's worked on an amazingly broad range of software, from the lowest, most "nitty-gritty" levels in the system, as close to the metal as you can get, right up to the highest levels.
He has independently written everything from a C++ compiler, a world-leading code optimizer, a processor microarchitecture simulator, a threading library and a data-structures library, through to user-interface libraries, stockmarket charting & analysis software, a chess game, 3D graphics and GIS utilities, mobile apps, and even a web video component.
In addition, he has enhanced GPU and printer drivers, worked on 3D graphics & animation software, tweaked search engines and network daemons, setup and run a CDN with servers spread across 4 continents, and worked in cybersecurity defending against both general and targeted attacks.
All in all, he has had an almost top-to-bottom coverage of the computing world!
Fields of Interest
Dr Patterson's primary area of interest and research is performance and the hardware/
His other main interest is 3D graphics, where his work has mainly been geological data visualization, taking drilling, magnetics and other geophysics data and turning it into 3D models and animations to gauge the size, shape and grade of mineral deposits – from a bunch of spreadsheets to a nice 3D visualization (in just a month!). Mining isn't the most exciting field, but it is Australia's largest industry (Australia is the world's biggest producer of iron, second largest of gold, and third of lead & zinc... and Dr Patterson never works on fossil fuels). The 3D animations really are quite interesting, considered the "best mining videos I've ever seen", leading to quite a few new mines, as well as saving some old mines worth billions of dollars! Of course, they don't have the coolness factor of the visual effects and games that Dr Patterson's brother Dale Patterson gets to teach at university, and they're nowhere near the high production values of the big-budget movies his friend David Gould writes shaders and does other effects work for, like "The Lord of the Rings" and "The Hobbit".
The current movement of graphics processing on-chip into the main CPU opens up interesting possibilities at the microarchitecture level, one of Dr Patterson's main areas of interest. Today, we only see simple designs that place an independent GPU onto the same chip as the CPU, but in time we'll see resource sharing between the two, and ultimately perhaps real fusion of CPU and GPU. When that time comes, there will be some interesting issues, particularly regarding the programmability of such designs.
Other topics, further away from the hardware, that interest Dr Patterson include coding design patterns, teaching programming as a universal skill everyone should learn at school (like reading/
The major progress in AI from deep neural networks both impresses and worries him. Convolutional neural networks for images and transformer large language models have clearly achieved stunning results (and chain-of-thought to a lesser extent, IMHO) far beyond what anyone expected to be possible 15 years ago, but the hype is out of control at this point, clearly an "AI bubble". AI systems are being deployed way before they're ready and well-tested, as the current scourge of prompt-injection exploits shows. Reckless early adoption of AI could lead to serious problems, and over-hyping AI technology will lead to, at the very least, mass disappointment, likely causing an AI bust and yet another AI winter. So Dr Patterson wishes his friend Adam Lynton well in his AI-related work, hoping for the best, but worried. We need to not get ahead of ourselves, or overpromise and underdeliver – we've seen that story before – lest the coming public backlash kill a truly useful, major technological breakthrough.
Aviation and aerospace have always been a major interest of Dr Patterson's, since building models and an RC plane as a boy. He keenly follows every new development in stealth, avionics, drones and rocketry, with the latter two making major progress in recent times. Clearly, drones of every kind, from low-end right up to advanced "wingman" designs like Australia's Ghost Bat, will be the defining feature of "6th generation" air power, as we're already seeing in Ukraine (note the flag flown in the nav bar, above – Slava Ukraine!). And the new era of commercial space has been great to see, with even Australia getting in on the action! I do hope NASA keeps being well-funded though – it's irreplaceable.
Flight simulators have always been a keen interest, again going back to his teenage years, with his current choice being X-Plane due to its accurate aerodynamics. Dr Patterson even partially wrote an (unfinished) semi-flight-sim as a teenager, as well as a simpler 2D scroller fighter-jet game. Speaking of fighter jets, it was sad, if inevitable, to finally see the retirement of the mighty F-111, an awesome plane with which Dr Patterson's family had a special fondness – they gave it a really good sendoff!
Finally, anything scientific or technical usually interests Dr Patterson, especially physics, astronomy and genetics. He was a huge fan of Carl Sagan's COSMOS TV series as a child, and would love to do a similar series on computing technology one day – it's way more interesting than people think!
See Also
- frequently asked questions (tabs vs spaces?)
- computing background (education & history)
- articles & papers (both technical & general)
- and, of course, numerous pieces of software