Mini Windows: If At First You Don't Succeed...

Can Microsoft gain ground on Palm with Pocket PC?

In last November's James Bond movie, The World Is Not Enough, the beautiful nuclear physicist Christmas Jones races against time to defuse a warhead as it whistles through an oil pipeline at 70 miles per hour. Jones whips out a Hewlett-Packard Co. palm-sized computer, running on Microsoft Corp.'s Windows CE operating system, to determine if the bomb is nuclear before it detonates. After deciding it isn't, Christmas and 007 get out of the way and let the bomb explode.

That's some computer. Unfortunately, the $4 million product placement underscores the fundamental flaw that has plagued Windows CE since the software giant first launched its pint-size operating system in 1996: Windows CE tries to do too much--and winds up accomplishing too little. Although the operating system can't actually detect nuclear bombs, Microsoft built it to power everything from pocket computers and television set-top boxes to digital gas pumps. But that everything-to-everybody approach made Windows CE bulky, inefficient, and unpopular. The operating system "was not successful. It was not well-received. It was not well-liked," says analyst Jill House of market researcher International Data Corp. Instead, Palm Inc. has dominated the handheld computer market with a 79% share, and other companies have beaten out Microsoft in other key markets.

POWERHOUSE ALLIES. Now Microsoft is trying again. On Apr. 19, it will launch Pocket PC, the new name for its Windows CE software designed strictly for small devices. On the same day, powerhouses such as Hewlett-Packard, Casio Computer, and Compaq Computer will roll out new palm-size models running Pocket PC--all with one goal in mind: to snatch share from Palm. Chairman William H. Gates III doesn't think it's too late for Microsoft to grab a big chunk of the market. These are still emerging markets, he argues, and Microsoft has made a mint by swooping into businesses that others have pioneered. "Were we the first mover in PC word processing? No. Were we the first mover in PC spreadsheet? No. Were we the first mover in PC networking? No," says Gates. "As long as there is still room for fundamental innovation, then there is still an opportunity."

He has reason to be optimistic. The new Pocket PC devices have a host of features that should help Microsoft gain ground on Palm over the next few years. For starters, the handheld computers are finally as small as a Palm. Microsoft also bundles in Pocket Word and Pocket Excel, so users can work on word-processing and spreadsheet files while they are away from their PCs. What's more, a pocket version of e-mail program Outlook can connect wirelessly to corporate e-mail accounts. That should help Microsoft boost its share of the handheld market from 15% this year to 40% in 2003, according to IDC. And Palm? IDC predicts its share will slide from 79% to 58%.

IDC's Microsoft market-share projections have been overly optimistic before. The research firm rejiggered estimates from two years ago suggesting that Windows CE would run 62% of handheld computers by 2002, compared with Palm's 29% share. This time, though, there seems little doubt that Windows CE will be able to take a bigger piece of the growing market. "It's a lot more friendly to use than ever before," says IDC's House. "People will say this has more features, more functionality, and I can grow with it."

The diminutive software couldn't loom larger in Microsoft's future. That's because the number of Internet devices such as pocket computers, set-top boxes, and gaming consoles is expected to boom over the next few years. IDC expects the device market to soar from 11 million in 1999 to 89 million in 2004, surpassing the number of PCs sold that year by a wide margin. Today, Windows CE barely registers on Microsoft's top line. But the company wants to be ready when the market takes off. "Nobody has made more than a rounding error of money in this category," Gates says. "We believe the category will explode, and we're taking this long-term bet."

Short term, however, Windows CE has been a disappointment. Consider set-top boxes. Liberate Technologies, a company spun out of Microsoft archrival Oracle Corp., is one of the leading players in the market. It's selling its operating system to cable operators such as Comcast Corp. in the U.S. and Telewest Communications PLC abroad. And that's after Microsoft agreed to invest more than $3 billion in those two companies. Where are the Windows CE set-top boxes? They haven't found their way out of the development labs.

GRAND DESIGNS. Microsoft hasn't fared much better in smart phones, which can receive e-mail and browse the Web. Microsoft has grand designs for that business as well. But it is Symbian Ltd., a consortium of mobile-phone makers and British handheld computer maker Psion PLC, that has lined up the key contracts in this emerging market. The best Microsoft has done so far: a deal last December with Ericsson, a Symbian partner, to build phones with Microsoft's Mobile Explorer Web browser inside.

Gates thinks his chance will come as each of the devices' tasks become more complex. Portable organizers, for example, could well evolve into gizmos that handle phone calls and play music. Already, consumers carry pocket computers, cell phones, and MP3 music players. Gates sees a future where all those devices become one. Set-top boxes could soon become digital gateways into homes, ushering in Internet access while connecting computers to printers. If that happens, Windows CE's ability to handle disparate tasks would make it the operating system of choice. "Eventually, these other guys will have to rev up their platforms to take advantage of what's possible there," Gates says.

It's an argument Microsoft has been making for years--and it has yet to pan out. Even Microsoft's own research suggests that the market for a personal information manager that can also play music or run games is limited. Last summer, Microsoft hired a Philadelphia market-research firm to see if there was demand for multipurpose devices. The company asked nearly 500 people who either used palm-size computers or intended to buy one if they wanted a device that could handle multiple tasks. While 72% did, the tasks that were most important to them--such as reading e-mail or jotting down notes--are also in the Palm. What's more, the features that distinguish the Pocket PC barely rated. Only 14% said it was important that the device "plays MP3 or other music downloaded from the Internet." And just 11% said it was important that the device "allows you to play games like Trivial Pursuit and Solitaire."

That's been the Windows CE bugaboo for years. "They took the PC metaphor and tried to extend it to a handheld device. It didn't work," says Merrill Lynch & Co. analyst William Crawford. Windows CE offered more than consumers wanted but asked them to pay for it anyway. Windows CE devices typically run from about $350 to $600, while Palms range from $250 to $500. Add the fact that the clunky Windows CE needs a larger microprocessor and a bigger battery than the Palm, and it's easy to see why consumers aren't buying.

FOCUS ON SIMPLICITY. That's where Palm has made hay. The Santa Clara (Calif.)-based company recognized that consumers aren't interested in a device that weighs more than seven ounces, no matter what its functions. Instead of adding music or games, Palm has focused on size and simplicity. "That doesn't mean we don't add new features. But we don't put those things in until we make them simple and wearable," says Michael Mace, Palm's chief competitive officer. Palm's first device with a color screen didn't make its debut until February, even though color has been a staple feature of Windows CE devices. While color improved the readability of the small screens, it also taxed the microprocessors and drained the batteries. Palm only added color after chip speeds and battery power improved. "We understood our customers better than the competition did," Mace says.

And Palm continues to up the ante. Last fall it licensed its operating system to Handspring Inc., a startup founded by three former Palm executives. The idea: to build an industry around the Palm operating system the same way Microsoft has built an industry around full-blown Windows. "Clearly, you don't want to compete with Microsoft as an island," Mace says. Handspring created the Visor, a palm-size device that lets users plug in modules, allowing it to double as an MP3 player or a cell phone. It's a broadside at Windows CE's bid to be the only device that can handle multiple tasks.

Such complexity continues to be the undoing of Windows CE in other markets such as set-top boxes. Three years after buying interactive TV pioneer WebTV Networks Inc., which figured out a way to let consumers surf the Web from their couches, Microsoft has yet to fit Windows CE into set-top boxes. WebTV devices still run on the homegrown operating system its founders created. And Microsoft still hasn't completed building the set-top box software for such partners as AT&T. As part of the AT&T deal, which calls for Microsoft to develop interactive-TV software for as many as 10 million set-top boxes, the software giant invested $5 billion last year in AT&T. "There's a lot of work to get the software on the box working," says Microsoft Vice-President Jon S. DeVaan. "We're in the year of making it real." Microsoft expects to roll out its boxes this summer. Again, the problem stems from the fact that the operating system wasn't built specifically for set-top boxes so it's not ideal to run them. "There is nothing about Windows that makes it well-suited to these devices," says Forrester Research Inc. analyst Josh Bernoff.

Perhaps CE's best showing so far is in markets where companies have historically built their own operating systems. Increasingly, everything from industrial machinery to medical equipment needs computer power. And software developers have found that using familiar Windows technology to manage complex tasks has its advantages. Windows CE has found a niche running such equipment as gas pumps and electronic kiosks. The software has even found its way inside feeding machines for dairy, pig, and sheep farming made by Furster-Technik of Germany.

Don't smirk, it's a huge opportunity. Although hardly as sexy as pocket computers or interactive TV, industrial automation software is a market that will top $1 billion by 2002, according to market researcher Dataquest Inc. And for corporations that also use other Microsoft products, such as its industrial-strength Windows NT operating system for big computing tasks, embedded Windows CE has a lure. The two operating systems share software hooks that developers use to make the programs work with one another. That way, developers need only write one program for a range of industrial applications.

Gates is determined to make progress in other device markets too. He estimates the company has spent "not quite a billion, but many, many hundreds of millions of dollars" on Windows CE and software for such devices as set-top boxes, pocket computers, and tablet PCs. And he won't stop spending anytime soon. The company is investing millions on voice-recognition software, so one day executives will be able to dictate notes into their pocket computers. Microsoft has created ClearType, a breakthrough technology that smoothes the jagged edges of electronic type, making pocket computers easier to read. So far, ClearType only can be found on Windows devices. And Microsoft will continue to tap its $18 billion cash hoard to invest in companies that agree to use Windows CE.

More than anything, though, that cash has bought Microsoft time. The company has worked at slimmed-down operating systems for a decade and finally seems to be turning the corner in the pocket-computer business. But that's only one market. Time may be running out in other markets such as the set-top box business. That may be a race against time that even Christmas Jones and 007 can't win.