Jeff Nelson: Chromebook Inventor

Image result for images jeff nelson

Right-click to download the podcast episode.

Jeff Nelson had a problem.

“In 2006 I was working on the Google Accounts team. I was writing a browser extension, and what I was finding over and over again was that restarting Firefox in Ubuntu was a particularly slow operation–like 45 seconds. At the time, there wasn’t a way to reload an extension and dynamically start the extension without resetting the browser, and this was really slowing me down as a developer.”

“What I did was find a way to restart Firefox faster.”

Jeff forked a Linux distribution called Puppy Linux and began using it for his development.

“Puppy Linux is designed from the ground up to run in RAM entirely, and this is different than other Linux extensions. That has enormous advantages from a performance perspective, because you eliminate I/O as a bottleneck. The Firefox restart went from 45 seconds to less than 2 seconds.”

As with many innovations, what began as a narrow solution to the inventor’s problem showed broader promise over time. ¬†Starting in 2006, Jeff led a campaign for his vision of a webtop. Repeatedly spurned by management, he left the company in frustration before his ideas reached fruition.

His vision has been carried out by other people. A modern Chromebook is a laptop running Chrome OS as its operating system. The device is selling exceptionally well.

“It’s more of a system than a device. When you are using a Chromebook, you shouldn’t need an offline Unix environment. That was never the intent. It is this online environment that allows you to access all of the services on the network. It is the hardware in your hands, plus all of the services we are delivering to you.”

Jeff feels he has been written out of history. The most vocal Googlers who have addressed Jeff’s complaints claim he deserves no credit. They believe that his thread of innovation was disjoint from that which birthed an actual consumer product.

“A lot of this controversy where some people are saying that I shouldn’t get credit comes from the fact that people feel like I am taking credit away from them. That’s just not the case. I’m putting an accomplishment that I did on my resume.”

“After I had built the initial prototype, I was so impressed that I brought it in and showed it around to some of my coworkers, as well as the director of consumer products I was reporting to, David Jeske. He kind of resisted it.”

“It was developed as a raw prototype, that had to be online all the time.” You couldn’t use Chromebook on an airplane.

“Because the team didn’t want to pursue this as project, I ran it by myself.” After several more months of solo pursuit, he sent a prototype of his operating system around the company.

“At that point product management got involved, because there was a powerful response. It did become a project. The main guy on the Chrome team, Mike Jazayeri got involved [as a PM]. Eventually we presented to Larry Page on the whole concept.”

This momentum proved temporary.

“It was December 2007 where I made the decision that it wasn’t happening. I was getting too much pushback from my own boss.”

“My own boss didn’t like the concept and didn’t want me to keep working on it. We had presented to Larry Page, Jeff Huber, Brian Rakowski. We had even taken it outside of Google, to HP and ASUS. It was going extremely well, except that the person who didn’t support the idea was my own boss.”

“I did attempt to change teams, and that soured my relationship with my direct manager even more. He wanted me to stay on the team and keep doing what I was doing. My own manager didn’t want to let me go.”

“It was company politics. People have this amazingly high regard for Google, but in the end, it’s a company. The same kind of corporate interplay nonsense that you have to deal with at any other company still occurs at Google.”

Michael O. Church’s description of “software politics” echoed in the words of Jeff Nelson.

“Some of that amazingly high regard for Google as a perfect company is not entirely well-placed. It’s still a real company with real problems and real people working there.”

“I had exhausted all of my options. I knew that without going over the head of my manager, I couldn’t change teams. It came down to the fact that I wasn’t having fun anymore, so I left Google.”

Since Jeff Nelson left Google, Chrome has taken off as a browser and an operating system. In the meantime, Jeff has taken criticism for trying to claim credit for his work.

Were the Chrome team’s efforts completely agnostic of Jeff Nelson’s work? That is unlikely. Mike Jazayeri worked closely with Jeff Nelson when the product had positive internal momentum at Google. Jazayeri ¬†also worked with the Chrome OS team. At a minimum, the cross-pollination between Nelson and Jazayeri has impacted the trajectory of Chrome.

“I don’t want to claim that when I invented Chromebook I wrote every line of code.”

“I worked on it for a year and half, I wrote two patents on it. I left Google in 2008, and those patents were actually filed in 2009, a year after I left the company.”

“If there is any animosity coming from the Chrome team about why I have this accomplishment on my resume, all it would take is two words from Larry Page and it would clear this up. I am getting slammed by these one or two guys that seem to think they know what happened, when they really don’t know what happened.”

“To some extent, the only reason we are even talking about this is that I happened to write the patents and Google decided to file them. If not for that, we wouldn’t even be talking–everyone would think I am this crazy guy who claimed to create this project. The purpose of a patent isn’t to back up your resume, but in this case it has become that.”

“People in software don’t think about intellectual property ownership as the guy who created technology, they think of the guy who is working on it right now.”

“The guy who worked on it eight years ago–they ignore that. There is that disconnect because people in software don’t view innovation in terms of the guy who started the idea, who wrote the first version.”

Jeff emphasized that he is not looking for a zero-sum form of credit. He doesn’t want to subtract from the claims of anyone else. “The other innovations–the people that created them deserve credit as well.”

A crucial question is whether Chrome would be as successful if not for Jeff Nelson.

“I don’t think so. I was the first person to build this project. We had meetings up and down management.”

“I called my initial prototype Guppy. When the Chrome team got involved, we created a new project called Google OS. In sometime around March 2008, there was a new project created called Penguin, which was similar to Google OS, but didn’t use a window manager. It ran entirely in a browser.”

“Around July 2009, they launched Polar Bear. It brought back a window manager, and looked more like Google OS. Offline mode came around 2009 as well.”

“My theory is that Google may be concerned with something like an intellectual property battle. I think they are worried that they will lose ground if they give me credit for this.”

“They should just clarify exactly why these patents are in my name, why they filed them. It would be a non-issue at that point.”

Since leaving Google, Jeff Nelson has recalibrated his career goals toward starting a machine learning company.

“I’ve been working for the man for a long time, and I want to go back to my roots of trying to start my own company, take an idea to fruition, and know that I’m building my own company and can take it to the next level. I’m trying to build a team of really exceptional people.”

“The market as a whole has started to appreciate machine learning. What a lot of companies are doing now is kind of naive. I think that there’s a next level that a lot of companies–even Fortune 500 companies–haven’t appreciated at this point. Over the past month, I’ve been trying to meet with as many companies as I can to identify the big problems that aren’t being solved yet.”

“A lot of companies jumped into this Big Data hype without understanding that they actually had to do something with that data.”

“After leaving Google, I didn’t want to just go get a job. I wanted to try to build my own company. That’s a very big decision to make. It’s easy to get a job in Silicon Valley. It’s not easy to start a company.”

“Google has this extraordinary mythology of being the best company to work for ever. When you make the decision to leave, it’s to some extent difficult to recover and decide where to go next. You are leaving this company that has been perceived as Charlie and the Chocolate Factory.”