The Man Who Founded Palm As A 'Side Project' Now Wants To Build Machines That Learn Like Humans

When Jeff Hawkins ventured into the mobile computing industry in the late 1980s, he intended to stay only four years to pick up business skills before returning to his passion of neuroscience.

He was about 30 years old at the time.

Just over ten years later, Hawkins’ company Palm was a mainstream brand after introducing the Palm Pilot PDA, enduring a number of ownership changes and listing on the NASDAQ.

Hawkins left Palm – and the mobile computing industry – in about 2001, as it struggled with the fallout that followed the dot com bubble.

He said Palm might have turned out differently had he stayed at the helm, but he felt that the time was right to return to neuroscience.

“Leaving was not easy; it was not easy for other people too. I was the heart and soul of the product of those companies [Palm and Handspring],” he told Business Insider Australia.

“But you only have one life to live, and I felt that my work in neuroscience was far more important than mobile computing.”

Learning from a side project

Speaking to Business Insider Australia in the lead up to his keynote presentations at the YOW! Conference in Sydney, Brisbane and Melbourne next month, Hawkins said that while he didn’t want to belittle his efforts with Palm, neuroscience had always been his passion.

Hawkins initially sought to study neuroscience in the mid-1980s, as a graduate student in Berkeley, but could not drum up enough support for his PhD proposal because there weren’t enough experts in the field.

“When I was younger, I was naive in many ways; I think I had to mature,” he said. “I figured I’d do 4 years in the industry to make connections, build up my finances and learn how to manage institutional change.”

Throughout his years at Palm and its successor Handspring, Hawkins said he had learned business skills such as managing people and expectations, marketing ideas, and finance from executives Donna Dubinsky and Ed Colligan.

Hawkins and Dubinsky now run a machine learning company called Grok, which commercialises technology that analyses data by recognising patterns and learning in a similar way to the human brain.

The company works with a community of open source software developers on machine learning algorithms and collaborates with IBM on hardware that can power brain-like computing.

Industry versus academia

Grok aims to be “a catalyst for future machine intelligence” and attempts to model machine learning on the neocortex – a region of the brain that receives and processes sensory information.

It stays away from “wet science”, relying instead on biologists’ published theories and data and putting those findings through “lots of math”.

Hawkins is an electrical engineer by training, having earned a bachelors degree in the subject from Cornell University in 1979.

But he describes himself as a “general purpose scientist and engineer”, arguing that he has learned vastly more in the 34 years since his undergraduate days.

“I am an engineer; I used to be a software programmer too. That’s helpful [with Grok],” he said. “I learned all the neuroscience on my own.

“A number of years ago, I used to think, ‘I’m not an academic, I don’t have a PhD in this field.’ But a number of my neuroscience friends said, ‘Jeff, don’t think like that.’

“The kind of work I do now is challenging on many levels. Especially at the leading edges of scientific endeavours, we’re doing stuff that’s not well-defined.

“There’s a lot of promotion of ideas and influencing people; it’s not just about publishing in a peer-reviewed journal.”

The future, and lessons for entrepreneurs

Hawkins has arguably achieved more with Palm – his “side project” – than most entrepreneurs manage to do with their dreams.

He told Business Insider Australia that entrepreneurship was a “very personal thing”, built on one underlying element: passion.

“The general advice I give people is to figure out what you’re passionate about,” he said. “I had two passions; I had neuroscience first, but I was also passionate about mobile computing.

“[Palm] grew and got very important and big, and I was very excited about it – I was stuck between those two worlds.”

He recommended also that entrepreneurs have a very long-term goal, and make any difficult decisions accordingly, highlighting Palm’s ownership changes and his founding of the Redwood Neuroscience Institute in 2002.

Hawkins says it’s impossible to know how exactly brain-like computing will pan out, but believes that it will be as significant as the rise of computing from the 1930s onwards.

“Alan Turing wrote [in the 1930s] that [the computer] would be a universal machine. They knew it was a big deal, but didn’t know the details,” he said.

“This is like 1943, 1944 was for computing; we’ve got [neuroscience] theories but there are decades more work to do.”

Business Insider Emails & Alerts

Site highlights each day to your inbox.

Follow Business Insider Australia on Facebook, Twitter, LinkedIn, and Instagram.