Bridging Islands in the Content Stream

Satoshi Nakajima has a dirty little secret: He's a former Microsoft bigwig who supports an open software platform. To say Nakajima was a bigwig is a bit of understatement. He was the lead software architect on the team that created Windows, and Windows is one of the most closed software platforms known to man. So for Nakajima to leave Microsoft in 2000 and set out to create an open platform for wireless devices was a shocking reversal in philosophy.

Now, the product of that philosophic shift, a Bellevue, Wash.-based company called UIEvolution, is promoting an idea some might call visionary. For the most part, though, it's merely practical: create a programming platform that developers can use to create an app or content just once, and only once, and have it work across all devices.

“I realized that connected devices required a new programming paradigm,” Nakajima said. “Today, if someone builds a PC app, they have to build a Windows application. They can't just build an application that works on any operating system. The same goes for wireless devices. I saw a need in the market for this kind of flexibility.”

In many ways, the situation is worse in wireless because of the fragmentation of the operating system market. While in the PC world Microsoft is the clear leader, multiple vendors designing disparate mobile OS platforms are vying for the lead in an immature market. Microsoft brought Windows to mobile, offering platforms for both wireless PDAs and smartphones. PalmSource, the old king of handheld computing, is pushing the boundaries of its code to encompass wireless. And the current marketing leader, Symbian, built a new modular OS platform from scratch.

Adding more layers of complication are the vendors building off the Symbian platform, creating middleware that further fragments the OS. In the CDMA world, there's Qualcomm's BREW. Then there are the various Linux- and Java-based operating systems, not to mention the carrier proprietary platforms such as NTT DoCoMo's i-mode.

Needless to say, every one of these platforms has its own application programming interface (API) and its own developer program. With the mobile data market still in its fledgling state, the industry has made it almost impossible to develop broad-reaching applications, Nakajima said. Developers are not only forced to launch a separate design effort for each operating system, they must also design for each middleware variation and often each individual phone.

So Nakajima and UIEvolution designed their own middleware — but unlike other middleware that serves only to fragment mobile data, Nakajima's middleware unifies it. Called the UIEngine, the software creates a standard programming interface for developers to build applications and create content. The engine overcomes the inherent differences between platforms and phones by shipping a device-specific client over the air along with any content a wireless user requests. That middleware client acts as the translator between the universal content and that device's particular platform protocols and interfaces. It's akin to a Java machine, forming a virtual processing engine in every device.

“The people who publish content hate to deal with ‘technology,’” said Chris Ruff, UIEvolution's vice president of marketing and product management. “These are creative people who want to focus on their products, not tinker with the mechanics of individual devices. Technology is the most frustrating thing in the world for content developers. We've found a way for them to not have to think about it anymore.”

So far, developers seem to like the thought of not thinking about technology. In fact, one of those developers, Japanese content giant Square Enix, was so enamored of UIEvolution soothing its technology frustrations that it recently agreed to acquire Nakajima's company — vision, technology and all — for $58 million.

“[UIEvolution and Square Enix] both have a vision for delivering content across all forms of media,” said Ichiro Otobe, general manager of Square Enix Co., Ltd., the Los Angeles-based arm of the Tokyo-headquartered company. The digital entertainment company is a supplier of content over NTT DoCoMo's network and publisher of such popular gaming content as Final Fantasy and Dragon Quest. “There is a benefit to owning the technology,” he added. “It gives us a good advantage as this vision happens.”

UIEvolution was created almost by accident. Nakajima never thought he'd start his own company after leaving Microsoft. He just thought he'd be funding one.

By 1998, Nakajima had become one of the most significant programmers at Microsoft. He designed the architecture for Windows 95 desktop, which propelled PCs into the living rooms of millions of people worldwide. He led the design team that created Internet Explorer. He produced the third largest portfolio of intellectual property for Microsoft of any employee (including Bill Gates himself). But he was still frustrated with the direction the platform was going. As Windows began to dominate the desktop world, its architecture wasn't bridging the differences between platforms. It was further isolating itself.

“I tried to solve that problem within Microsoft from '98 onward,” Nakajima said. “But I hit roadblocks. Those ideas broke the existing business model, and I realized I wouldn't be able to do them at Microsoft.”

So Nakajima left, and with fellow Windows guru Brad Silverberg and wireless maven Steve Hooper formed Ignition, a venture capital firm and wireless incubator. Nakajima's original goal was to find a company that was espoused his ideas about open architecture in the real world and then lend his expertise and Ignition's backing to the project. But that company never popped up.

“There were 40 companies trying to become the wireless Yahoo and 20 companies trying to become the wireless Hotmail, but no companies wanting to pursue something original in wireless,” Nakajima said. “Finally I decided to start the company myself.”

With Ignition's help, Nakajima launched UIEvolution, committing his design team to do for wireless what he couldn't do for the PC at Microsoft. The company faced numerous obstacles, but the biggest was the reluctance of the vendor community to see all of their operating systems bridged. So, UIEvolution decided not only to bypass the vendor community, but also bypass the carrier community entirely, going directly to application and content developers — companies like Square Enix — with its solution. The idea was that developers not only wouldn't have to deal with the fragmented vendor community, they also wouldn't have to deal with the competitive politics of carriers, which typically want exclusive deals as well as a share of the revenues.

The UIEngine is essentially an amalgam of separate middleware clients, each tailored for the particular operating system and environment that a given mobile device hosts. It consists of two parts: a software client and a server, which share the computational tasks of any given application. Whenever a customer downloads a piece of software, he downloads the UIEngine client specifically designed for his device.

For example, Disney, another of UIEvolution's customers, wants to distribute screensavers based on its popular cartoon icons. Disney designs one set of screensavers using the UI development environment, and then places it on a server with the UIEngine platform. When a customer downloads a screensaver of, say, Cinderella, the UIEngine detects the make, model, operating system and processing capabilities of the customer's device, and then sends the screensaver along with the appropriate middleware client to run that screensaver on that particular device. Disney only creates the content, while UIEvolution manages the complex task of keeping a steady catalog of clients for every mobile device released.

To avoid cluttering the already taxed memory capabilities of smartphones, UIEvolution developed a very thin device client, leaving most of the number crunching to the server itself. This presents a problem because an application running from a server has to push its results over the air to the phone, facing the latency limitations of today's 2.5G networks. Nakajima admitted that the platform is not ready for the most robust real-time applications, but due to the technological limitations of today's smartphones, neither are most mobile devices.

“Middleware and operating systems take up so many resources on the phone — to add another bulky layer of software would be pointless,” Nakajima said. “Our client is so small, it only takes up 30 kilobits on a handset. Our middleware was designed with network latency in mind. In fact, we specifically designed the engine to work at speeds of 14 kb/s. Feasibly it will work fine at 7 kb/s. It doesn't require a hefty connection.”

UIEvolution was just starting to make its way into the wireless consciousness when Square Enix made its move to buy the company. In addition to having signed Square Enix as a customer and early stage investor, Nakajima's team also signed a deal with Disney to distribute content for its animated films and ESPN divisions.

Square Enix' Otobe said the company doesn't want the acquisition to hurt UIEvolution's chances of winning more fans in the content developer community.

“We don't want to monopolize the technology,” Otobe said, though he added that he expects Square Enix to become UIEvolution's most important customer.

UIEvolution's Ruff modified that stance by pointing out that being owned by a larger entity means that more resources could potentially be dedicated to other UIEvolution customers.

“Companies that go after media dollars are not always direct competitors,” Ruff said. “UIEvolution's other customers will benefit from the acquisition. It will give us the R&D budget to accelerate our ability to execute our concepts.”

What UIEvolution is accelerating toward is a destination that retains the spirit of Nakajima's original idea, but that has the potential to apply to more than just the mobile industry, mobile devices and mobile users. Square Enix and UIEvolution want to use UIEngine to bridge other islands in the broad stream of multimedia content and consumer electronics devices, such as TVs, computers, MP3/MP4 music and video players, and other delivery venues.

“UIEvolution always has had a vision of any content over any kind of device,” Ruff said. “Its software is the vehicle that will allow users to realize the vision.”

Still, Nakajima, Ruff and company have a long way to go to sell their ideas to the mass developer community, even with the backing of Square Enix. For now, vendors and carriers typically still control the content experience, and in the earliest stages of wireless data experimentation, developers seem all too willing to work with them to ensure that their products find a market.

However, as more and more operating systems are launched and content developers become more and more frustrated with the fragmentation situation, vendors may begin to see some value in a less fragmented market, said John Jackson, wireless data analyst with The Yankee Group.

“Vendors are realizing that their device strategies could use a closer look,” Jackson said. “They might need to adopt a more wholistic approach.”

Other companies, such as Openwave, have designed their own software to bridge the differences between operating platforms, but those solutions are aimed at vendors and carriers, the same companies that created the fragmentation in the first place. UIEvolution is unique in that it is dealing directly with developers. Nakajima admitted the approach has its risks, but he said the rewards would be great not only for UIEvolution but the industry overall.

“If we succeed, it might make vendors reconsider their strategies,” Nakajima said. “Content providers love the fact that they can just deploy their applications over any phone and over any network. Maybe the vendors will see the light.”