Life after Google: Brad Neuberg's HTML5 start-up

LONDON--For someone interested in capitalizing on the new era of advanced Web standards, you'd think Google would be a pretty good employer. After all, it's got an up-and coming browser, some of the world's most influential Web applications, and plenty of money to invest in both.

But in the culture of Silicon Valley, sometimes there's a time to strike off on one's own, and that's what Brad Neuberg, a very visible Web programmer at Google, decided to do. He announced his departure on the eve of a speech last week at the Future of Web Apps conference here.

In an interview with CNET afterward, Neuberg said he plans to launch a San Francisco start-up in November focusing on the same suite of Web technology he's been steeped in at Google. He's cagey on details, but he said he plans to focus on Web applications for consumers.

"I drank the HTML5 Kool-Aid," he said, saying it and other Web standards are fueling a new wave of entrepreneurial activity. " I really believe we're starting to see those start-ups. We'll see that accelerate in the next six months to a year and a half."

Plus, he didn't like spending three hours a day commuting from San Francisco to Google's Mountain View, Calif., offices and back for two years and nine months of his life.

"What am I sacrificing? It didn't all fit," he said. "I should be doing what I would do if I won the lottery," so now he's begun trying to gather a group of like-minded folk for the start-up.

Winning the lottery
He certainly has plenty of deep experience with the technology. Neuberg has been a JavaScript programmer for Google Docs and Google Buzz. While a developer advocate for Google pushing Web technologies, he created a project called SVG Web. It lets the ancient but still widely used Internet Explorer 6 handle the increasingly important SVG (Scalable Vector Graphics) format via a JavaScript library that would hand off commands to Adobe Systems' Flash Player. He was an evangelist for Google's Gears plug-in to extend browsers with abilities such as letting Web applications work offline.

To Web developers, a lot of Neuberg's views will sound very familiar. He's a fan of SVG and Canvas for advanced 2D graphics, WebGL for 3D graphics, HTML5 for video and audio. And behind the scenes, there's CSS for formatting and animated elements, Web workers to let a browser perform background processing tasks without disturbing the user, and geolocation to tell a Web application the physical location of a person using the Web. In short, it's the full list of buzzwords and technologies that have the Web world genuinely excited, especially with the arrival of Microsoft as an ally advancing the frontier rather than holding the rest of the world back.

But Neuberg also is willing to deviate from the orthodox. Indeed, one of his views verges on blasphemy.

"I think the future is going to WebKit," he said, bluntly.

To translate, that means he's a fan of the open-source browser engine that underlies both Apple's Safari and Google's Chrome. It's also the foundation for many mobile device browsers, most notably those built into Apple's iOS and Google's Android.

Sure, people have their favorite browsers. But Neuberg is saying something different: in many ways those WebKit browsers serve as the new standard for Web developers. That view is at odds with the conventional wisdom in the Web world, that after years of being in thrall to the foibles and shortcomings of Internet Explorer, programmers are now able to program to Web standards that exist independent of individual browsers. The promise is that they'll be freed from the pains of reworking their Web sites and applications to get them to work with incompatible browsers.

Leading with WebKit
Neuberg, though, argues that programmers must grapple with browser differences today, especially when trying out cutting-edge features, and that WebKit is the best place for developers to invest their time.

"There are more and more Web standards. You have to know the ins and outs and intricacies," Neuberg said. Not only has WebKit been on the vanguard, he said, but "The reach of WebKit is amazing...If I were to invest major time on the ins and outs of an engine, it would be on WebKit. I'm having the most fun on WebKit now."

He likens the situation to the first browser wars of the 1990s, when Internet Explorer 4 offered compelling features that drew programmers and he spent hours poring through Microsoft developer documents to learn the craft.

Of course, Microsoft's browser clout--no small part of it the powerful Windows distribution channel--gave it the dominance that proved so stifling for much of the decade after IE won that browser war. Does Neuberg fear such an outcome again?

No. "Things are converging," he said, and within a year top browsers won't be so different.

Standards and browsers must develop together, he said.

Standards and browsers interdependent
"It's the description of the system and the system, evolving together. If either part gets too far ahead, it collapses," he said. The implication is that a world where standards groups rule the world is as bad as one where a particular browser does.

But Neuberg is focused on the here and now, and he thinks there's a reason he can overlook some traditional browser compatibility concerns and focus more on targeting WebKit: Apple's mobile devices.

"I might imagine that if you build an amazing app, it gets an installed base who loves it on the iPhone and iPad. You start from that nexus," he said. "If you have a killer iPad app, that's an ideal market of early adopters and people who communicate."

Of course, plenty of programmers have found success through Apple's iOS devices, but with native applications rather than those that run in a browser. But Neuberg clearly is a believer in Web applications.

His mobile-first idea is a reversal from prevailing practice, where Web developers start with a site for personal computers and add support for mobile devices later.

Blowing off backward compatibility
The approach frees programmers from what was Neuberg's specialty, "backporting the future to the present" through herculean efforts such as SVG Web to make new Web sites work on old browsers.

"What happens when you develop to the leading edge of the Web?" Neuberg asked. "You can rapidly iterate and deliver features and not hardcore hacks. You can grow from there."

The result, he predicted, will be Web sites and Web applications that look profoundly different. It'll be a sequel to the Ajax programming methods that opened the doors to a new class of applications a few years ago.

"Google Maps and Flickr legitimized Ajax. I think we'll see that in the next year," but with today's new Web technologies, he said. "Somebody is going to take this stew of new things [and put them] together in a neat combination. Somebody will take some HTML5, and geolocation, and mobile applications, hook into Facebook perhaps, and they're going to do something unexpected."

He didn't say it, but it's clear: Neuberg hopes to be that person.