Look Who's Powering Palm


Is the Palm (PALM ) software that powered the first practical handheld in 1996 and redefined mobile computing on the road to oblivion? It sure looks that way, at least for devices such as Palm's popular Treo, which combines voice and e-mail service with the traditional contacts and calendar functions of a PDA. Palm will soon announce a Treo powered not by Palm OS software but by Microsoft's (MSFT ) Windows Mobile 5. And while Palm will sell both Palm OS and Windows Treos for the indefinite future, Microsoft software is likely to dominate the market over time.

The sale of PalmSource (PSRC ), the software arm spun out of the former Palm Computer in 2002, will also affect the future of Palm OS. PalmSource had been struggling both to find more customers and to get its products, particularly a simpler operating system for cell phones, to market. Earlier this year it sold its share of the Palm trademarks back to the hardware arm and put itself up for sale.

As the only significant licensee of Palm OS, Palm was widely expected to buy back the software business -- and it tried. But major U.S. and European handset makers jumped into the bidding and drove the price sky-high. "There was a point beyond which we didn't think it made sense," Palm CEO Ed Colligan wrote in an e-mail to Palm employees that was obtained by BusinessWeek. After Palm withdrew, Japanese software maker Access, which supplies Web browser technology for the Treo, came up with the winning bid: $324 million in cash, an 83% premium over the market share price.

FOR PALM THE ACCESS DEAL is the least objectionable outcome since it prevented Palm OS from falling into the hands of a competitor. Palm is not likely to shift away from Palm OS as the software behind its Zire, Tungsten, and LifeDrive PDA models. But the market for these nonphone handhelds has been declining for several years as wireless handsets take up more of their functions. And with the acquisition of PalmSource now off the table, Palm can be agnostic about its software choices.

Although I have long been a Palm fan, I have to concede that Microsoft software increasingly makes sense for a converged device like the Treo. The first Pocket PC Phone Editions, which came out three years ago, were ghastly, but the hardware and software have steadily improved. I tried a Hewlett-Packard (HPQ ) iPAQ HW6500, due this fall from Cingular Wireless, and it was nearly as good as a Treo 650, both as a phone and for e-mail -- and it uses an older version of the Windows Mobile software. HP used a Treo-like square display instead of the elongated ones used in other Pocket PCs. This left room for a built-in keyboard while keeping the device compact and not too top-heavy for typing. Between the improved software and Palm's Treo design experience, the Windows Treo, expected to hit the market early next year, should at last be as good as the Palm OS-based Treo 650.

The embrace of Microsoft is bound to cause howls of betrayal among Palm faithful. But the time is ripe for Palm to move to Windows Mobile. The hardware inside Treos and Pocket PC phones is virtually identical. Windows Mobile is popular with corporations, especially those whose mail systems are built on Microsoft Exchange and Outlook. And the fact that programs can be written to run on both Treo-size devices and the smaller, cheaper Windows Mobile Smartphones, such as the Audiovox SMT 5600, appeals to both companies and independent software developers.

Microsoft may be a company of predatory instincts, but its history of competition with Palm is typical of how it so often wins in the end. With the patience conferred by deep pockets and the determination to keep trying, it can eventually penetrate any market it desires. Palm held off the juggernaut for longer than most, but the time for change is rapidly drawing near.

For a collection of past columns and online-only reviews of technology products, go to Tech Maven at www.businessweek.com/technology/wildstrom.htm

By Stephen H. Wildstrom

    Before it's here, it's on the Bloomberg Terminal.