Is Tungsten too bright for Handspring?

Will Palm's new Tungsten burn Treo? Will Handspring ride it out or find a buyer? In the war against Microsoft, the stakes for Palm-based devices have never been higher.
Written by David Berlind, Inactive
As of yesterday, Palm officially entered the converged PDA/phone market with its GPRS-enabled Tungsten W handheld. There's also a Bluetooth-enabled non-GPRS version called the Tungsten T that targets mobile pros who don't want a converged device. It's one of the first devices to run Palm OS 5.

If the Tungsten W's future burns brightly, its primary position in the market will be as a RIM BlackBerry 5810-killer. Like the BlackBerry, the Tungsten W is capable of voice communications via a hands-free headset--its emphasis will be on data communications like e-mail. (It will also take on other Palm-based converged devices like those from Samsung and Kyocera). But the Palm-based device that could be most threatened is Handspring's Treo.

Except for the Treo's clamshell cover, a Tungsten W could be easily mistaken for a Treo. Both have color displays and BlackBerry-like thumbboards. Like BlackBerry, the Tungsten's thumbboard does not illuminate for use in low-light situations; the Treo's does. Dialing can be tricky without this sort of illumination. Both operate as Palm OS-based PDAs and cell phones. Whereas the initial Tungsten W models will support GSM/GPRS-based wireless service providers (T-Mobile and Cingular), the Treo now has support from both the GPRS crowd and the CDMA 1xRTT providers (e.g. SprintPCS).
ZDNet readers and other users report that the CDMA 1xRTT networks are working faster and better than their GPRS-based competitors. I expect the Tungsten W will support CDMA 1xRTT in the not-to-distant future.

Handspring recently laid off 20 percent of its staff; its sub-US$1 trading price could result in a delisting; and it has been operating in the red for over a year. Should Handspring be worried that this market altercation will affect your next purchase? Probably. But not simply because Palm has entered the "Palm-based converged devices with thumbboards" niche that Handspring had to itself. The new operating system, which is more closely tuned to the idea of converging a phone and PDA into one device, also raises the stakes in the war between the PalmOS and Microsoft's PocketPC operating system.

Because Microsoft has the determination and the resources to win this battle, survival of the Palms (Palm for the hardware and PalmSource for the software) cannot be left in the hands of PalmOS licensees such as Handspring, Sony, Kyocera, or Samsung. Palm must take its destiny into its own hands, especially now that competition in the PocketPC market will result in remarkably lower prices for devices based on that operating system. Palm can't afford to do what it has done in the past---wait for PalmOS-licensees to innovate around the platform as Handspring has done.

It was that lack of execution on innovative ideas within Palm that gave birth to Handspring in the first place. According to former Palm executive and now Handspring president Ed Colligan, "3Com's control of Palm prevented us from moving the platform forward the way it needed to be moved forward."

Now, with the ugly 3Com history behind it, and to keep the Microsoft threat at bay, Palm wants to set the example for what a PalmOS-enabled device can do. Enter the Tungstens. Although PalmSource CEO David Nagel vehemently denies that Palm will have the sort of behind-closed-doors advantage over its competitors that Microsoft's Office team was said to have over companies like Lotus and WordPerfect, Palm will have to be the first to capitalize on PalmSource's innovations in order for both companies to deal with the Microsoft threat.

The Tungsten introduction is simply the first stake in the ground to marking Palm's obsession with Microsoft and devices like T-Mobile's PocketPC Phone edition.

Handspring: Ride it out or find a buyer?
What market carnage will be left in the wake of the Palm vs. Microsoft battle? And how will a beleaguered Palm OS-licensee like Handspring fare in the bloodletting? It could be trouble. The company is hoping to break even by the end of the year, but I wonder how much of that optimism is based on some good holiday results. Already, the expectations for holiday spending are gloomy at best. In the context of that bad news, the Palm OS-based market can only pray that it can survive the price war that Dell has already started on the PocketPC side.

However, in my conversation with Colligan, he downplays the threat from either Palm or Microsoft. "The Tungsten is an organizer first, and a phone second," says Colligan. "The Treo is a phone first and designed from the point of view of someone who needs a great phone. I don't know anyone who wants a phone that requires an earpiece. In terms of Microsoft, PalmOS has 50 percent of the market. We're coming at this from a position of strength."

I agree with Colligan on the usefulness of a phone that requires an earpiece. RIM's 5810 didn't fly in my experience. But regarding the market share issue, 50 percent share is down from close to 100 percent, thanks in large part to Microsoft.

What are the next steps for Handspring? The company can try to ride it out or it can find a buyer. If Handspring chooses to ride it out, I wonder if its chances for survival could be improved if the company unhitched its wagon from the PalmSource horse. To do battle with Palm and Microsoft, Handspring should at the very least be equipped with a legion of developers. What operating system has such a following, appeals to businesses, and is a constant thorn in Microsoft's side? Java.

Java virtual machines (JVM) exist for handhelds today. But, developers I've talked to complain that the applications they've prototyped on those VMs are slow compared to the ones they've developed to run directly on the underlying OSes like PalmOS and PocketPC. Performance issues have been somewhat ameliorated through brute force (processors like Intel's XScale) and the use of the smaller-footprint-but-less-functional (compared to the native OS) versions of Java.

Even so, developers don't seem to be flocking to trimmed down JVMs like J2ME and Personal Java. If Handspring could reproduce the end-user functionality and developer pliability of a PalmOS device with Java instead of the PalmOS, it might have a winner on its hands. Not only would Handspring be able to tap into a developer base that dwarfs that of the PalmOS, it would increase the appeal of Handspring's products to business users who have the freedom to choose a mobile device as long as it supports Java.

Handspring isn't the only company that could reverse direction, go the Java route, and have something more compelling with which to battle Microsoft. PalmSource could do it too. But when I suggested this to David Nagel, he said it wasn't going to happen any time soon. Nagel said that the company is sticking by the PalmOS and that its partnership with JVM-provider Insignia should satisfy the needs of anyone with Java requirements.

Colligan agreed, adding that support for Intel's higher performance XScale architecture will overcome any JVM performance issues. "Name me 10 breakthrough applications for Java," asked Colligan rhetorically. "We don't need to capitulate to Microsoft. PalmOS has 100,000 developers, and because of things like its user interface APIs, the PalmOS is a much better lower common denominator to develop for than Java. Since Java wasn't designed with a particular platform in mind, it's just not as good a common denominator for developers as the PalmOS is."

I disagree. In the war against Microsoft, Palm needs to help unite the opposition, and the only viable game in town for that is Java. By leaving PalmOS out there, there's no need for Microsoft to divide and conquer. The mobile community is already divided.Now, all Microsoft has to do is conquer. The trend speaks well for Microsoft's odds. If, as Colligan says, the Java platform isn't robust enough (compared to the PalmOS), then perhaps that can be fixed.

Nagel's commitment to PalmOS paves the way for Handspring to innovate once again. If I ran Handspring, I'd carve out the PalmOS and replace it with something like SavaJe's high performance J2SE-based operating system. Yes, that's J2SE-- the standard edition version of Java that usually runs on desktop computers.

SavaJe has J2SE running natively on XScale-based handhelds, and it's faster and more capable than anything else I've seen in such a small footprint. Moving to J2SE would allow Handspring (or Palm, if Nagel changes his mind) to tap an even larger base of developers and applications. Of course, if Handspring does this, the engine overhaul would have to be transparent to end users. The core PIM applications and synchronization software would have to be comparable to that previously offered on the Palm platform. It's something that I'm relatively certain Handspring and SavaJe could work out.

Not having seen SavaJe's technology, Colligan wouldn't venture a guess if it has the sort of potential I described to him. In the mean time, Colligan and Handspring's management seem determined to ride it out. "Our biggest challenge is still in working with carriers because they handle the distribution. We're totally focused on the communicator space. We still have the Treo90 for basic organization, but the focus of our R&D is on the communicator side of the equation. We're convinced it's the best thing on the market and I know that once you try it you'll agree." Handspring is sending me a Treo. Stay tuned for the results.

As for being acquired, imagine if Handspring made the Java transition. I can name at least two three-letter companies that promote Java to death and would do anything to kill Windows, would like to sell hardware, and don't currently have a handheld in their portfolio. Talk about the Sun (or IBM), the moon, and the stars lining up!

Did David forget to take his medication, or does he have it right? What do you think the wireless handheld market will look like in the next three to five years? Be a market analyst: Share your own handheld market prognostications with your fellow ZDNet readers by voicing your opinion with TalkBack below.

Editorial standards