The sound of one Palm clapping

If you're charged with setting corporate standards for handhelds, it's important to understand who really controls the PDA-phone convergence space: the wireless carriers. But Palm's acquisition of Handspring this week adds a new twist to the plot.

Handspring's line of Palm OS-based Treos makes it one of the key players in the convergence space. Reminiscent of how Apple brought back Steve Jobs, Palm's acquisition of Handspring has the makings of bringing back Palm's glory days. Handspring's top brass, Jeff Hawkins and Donna Dubinsky, were the founders of Palm back in 1992. Eventually, they left the company under acrimonious conditions and formed Palm OS-licensee Handspring. Now, they--along with the Treo--will be rejoining the ranks of Palm.

Prior to the acquisition, both companies were struggling in the face of stiffening competition from Microsoft, Symbian, and RIM-based alternatives. My question is: Can the companies do any better together than they were doing when they were apart?

The answer is important because deciding on a corporate standard for any platform---especially if it requires custom application development---often involves a confidence check on the competing platforms' futures. My other question isn't only whether the Palm platform will survive, but whether it's the best choice.

Palm's long-term survival depends largely on two aspects of its strategy. First, can it regain its role as the company we once depended on to push the envelope on the very definition of a handheld device? Second, can it accomplish one of two things in the developer arena: Can it grow the number of application developers (and subsequently, the number of applications) that ultimately determine a platform's popularity, or can it become the best platform for powering another application environment like Java?

Palm ascended to its prior heights on the basis of the KISS (keep it simple stupid) principle When other offerings were trying to be all things to all people, Palm recognized the limitations of handheld technologies and focused on doing the few things that handhelds could do, and then got its handhelds to do those few things extremely well-far better than any other handheld offering. But, while the company stuck to the KISS principle, another principle --- Moore's Law --- caught up with and overtook Palm, despite what Palm's chief competitive officer Michael Mace was claiming in 2001.

Mace told me that Pocket PC was too big, too complex, and required too many pit stops at the recharging station to be appreciated by anybody but the techno-elite. "Handheld technology is immune to Moore's Law," he said. "The situation is not about to improve for Pocket PC any time soon."

Indeed, PocketPC is big, complex, and can require frequent pit stops for power if you own a device that doesn't have a replaceable battery (a no-no for any handheld design). But even with those attributes, the devices appealed to more than just the techno-elite. (Especially once Dell entered the fray and drove the entry cost below the $200 price point.)

Two years later, Palm is eating Mace's words as PocketPC-based devices continue to eat away at Palm's share of the handheld OS pie. While handheld technologies were becoming more capable and Microsoft's PocketPC OS in particular took advantage of those capabilities, Palm remained true to its original convictions. At the same time, Research in Motion was beating away at Palm's left flank (handheld wireless messaging). Today, taking a long view of the market, Palm looks more like a follower than the leader it once was. This is further evidenced by the preview I received of the Tungsten W, during which Palm officials referred to the thumbboard-sporting handheld as a "BlackBerry-killer." In my estimation, the Tungsten lineup, which includes the recently introduced Voice-over-IP and Wi-Fi-enabled Tungsten C, is doing very little to push the envelope.

On the developer front, things aren't much better. Just about any platform that achieves a notable critical mass must attribute that popularity to the availability of applications and the number of developers working on them. Over the years, Palm has done a great job winning over the development community. But, based on the feedback that I've received from ZDNet's readers during the past year, Palm apparently caused some headaches for its devotees when forward compatibility problems surfaced after version 5 of the Palm OS was rolled out. (Readers have made a similar complaint against Handspring with respect to the Treo's lack of the Springboard slot that was found in the company's Visor offerings.)

However, while Palm and its customers iron out compatibility issues, the maturity of Java in the mobile environment, and its availability on many more devices than the Palm OS will ever be on, begs the question: Why would a developer target anything but a Java or Windows runtime environment? If I'm a corporate developer looking for a platform with industry-wide support from both hardware and software manufacturers, Java and Windows are the hands-down winners. Palm is a bystander.

Almost two years ago, I suggested to PalmSource CEO David Nagel that he resolve this problem by going to a pure Java platform rather than continuing to evolve the old Palm OS. At the very least, Palm should be doing everything possible to make any Palm-based device the best handheld for running mobile Java applications (whether Java is the native OS or just a virtual machine).

But, during a recent interview regarding the Tungsten C introduction, Palm senior product manager Paul Osborne told me that "the Tungsten C doesn't have a Java Runtime Environment (JRE) included with it and neither does the Tungsten W. Several JREs are still being optimized for Palm OS 5. When they are, we might make it a download, but the cost is not clear and, in terms of JRE providers, we can't talk about it at this point."

Meanwhile, Research in Motion recently introduced the tri-band GSM/GPRS radio, color-based 7230 integrated phone, PDA, and messaging device in Europe. The co-CEOs of that company, Mike Lazaridis and Jim Balsillie, have already seen the wisdom in tapping the Java developer community by moving the BlackBerry into a pure Java environment. Today, if you're developing an application for the BlackBerry, you do it in Java. That's the kind of vision that Palm so far has lacked.

Another important benefit of the Java environment is the hardware choice that it creates for buyers. As more handheld devices (especially phones) come out supporting Java--and even Windows--users get more choices. The leverage of that choice is what drives hardware prices down and incents manufacturers to innovate in ways that ultimately benefit buyers. The concept of this leverage brings us full circle to Palm's acquisition of Handspring.

Compared to Java, there are only a handful of meaningful Palm licensees and, consequently, only a handful of meaningful Palm-enabled devices. By acquiring one of the most recognized, and perhaps the most innovative of its licensees (Kyocera is another), Palm is shrinking the ecosystem it so desperately needs to grow. I don't doubt that the dynamic duo of Hawkins and Dubinsky can bring some fresh ideas to the beleaguered Palm. But, between the Treos from Handspring and the Tungstens from Palm, something will have to give. (Something had to give in the PocketPC world after HP merged with Compaq. HP's Jornada gave way to Compaq's iPaq.)

And exactly what does Palm get out of the Handspring acquisition anyway? For starters, it gets a customer list. After that, the Treo. I've been testing the Treo 300 (the CDMA version) for a several months. (The results will be published tomorrow.) Although it's one of the better converged devices out there, it's by no means a slam dunk. It has plenty of shortcomings when compared to other converged devices, and some of these shortcomings are tied to the fact that phone functionality appears tacked onto, rather than deeply integrated within, the operating system. This is not Handspring's fault, since it had no control over the Palm operating system. Unless Handspring had something earth shattering on the drawing board, the addition of the Handspring products to Palm's lineup isn't going change Palm's fortunes.

Perhaps what Palm gets from this deal is distribution. Remember, the wireless carriers control the converged device market. The carriers are the only distribution channel for converged devices because they're the ones who have to provision the devices with voice and data services. Based on what the converged device manufacturers have told me over the last couple of years, getting into that channel is an uphill battle that's won after long sales cycles and multiple rounds of golf. With its wide area, wireless network-enabled Tungsten W, which even Palm suggests you not use as your primary wireless telephone, Palm hasn't exactly wowed the wireless carriers the way RIM and Handspring have with their offerings. So, perhaps Handspring's all-important relationship with the wireless carriers will end up being the most valuable part of this acquisition.

But whether it's enough to instill market confidence that the Palm platform is in it for the long haul --- enough to make enterprises think twice before going to an alternative platform --- remains to be seen. Right now, I'm not seeing it.

What do you hope to see come out of Palm's acquisition of Handspring? If you're developing for a mobile platform, which one are you going with? Palm? Java (J2ME)? PocketPC? Symbian? Share your with your fellow readers using TalkBack. Or write to me at david.berlind@cnet.com. If you're looking for my commentaries on other IT topics, check the archives.