Pros and cons of the Linux PDA

COMMENTARY--Although the idea that a personal digital assistant running Linux could provide the same functionality as a Windows CE or Pocket PC device at a significantly lower cost is attractive, fact is most costs come from hardware.

Microsoft and Palm both charge only $7 to $10 per operating system license for a PDA, so eliminating that cost by using Linux would affect pricing only at the extreme low end of the market (that is, PDAs priced under $100, which offer only the most basic functions and are unlikely to meet most companies' requirements to support mobile workers).

Linux PDAs also have the disadvantage of having few third-party applications available, while the Palm community offers more than 1,000--and the Pocket PC application list is growing rapidly. On the plus side, Linux offers a stable operating system with many of the features that corporations need for the custom applications they want to develop for mobile or "pervasive computing," including the ability to run Java programs on a Java Virtual Machine (JVM).

But many Linux-powered devices will have only a subset of the typical Linux or Unix features. Any company developing an application for a Linux-powered PDA should carefully examine the device's capabilities.

One of the major issues in PDAs (and one of the largest contributors to their price) is the display, so no matter what operating system is under consideration, corporate customers should carefully evaluate the ability of a given device to meet the various lighting and resolution needs. Older Palm OS models in particular suffer from gray-scale displays with low resolution that fade in certain lighting conditions. Newer, reflective, side-lit and higher-resolution color displays on the latest Pocket PC and Palm systems offer much better displays at higher prices.

We believe that Palm will fade to a minority position in the high-end consumer market and fail to capture a strong share of the corporate market, remaining strong only in the low-end consumer market where personal information management is the primary function. This is mainly because Palm has recently failed to reinvest adequately in upgrading its operating system, giving Pocket PC devices a major opportunity to make market gains.

Palm is now promising to finally introduce its heavily touted OS 5.0, which will move it from the ancient, 8-bit DragonBall processor to an ARM-based processor (which Pocket PC also uses) by the end of 2002. We believe OS 5.0 will include rudimentary multithreading, multitasking and other features that are basic needs for corporate application development and that have always been part of Windows CE.

Further, Palm must include native support for advanced Internet-based services (part of the new Windows CE version to be introduced this month), which will be on Pocket PC PDAs within 18 months. Regardless, Palm will be left at a disadvantage among companies that adopt Microsoft's .Net strategy, since Microsoft will offer a more functional capability for Pocket PC than for Palm.

User organizations must understand that reducing OS license costs for PDAs has minimal impact and that Linux PDAs offer fewer applications than either Palm or Pocket PC. In comparing processor speed, memory, flash and above all screen quality, it makes little difference in overall cost whether the device is powered by Palm OS, Pocket PC or Linux. Applications are the key concern, and businesses should focus on the types of applications (managing personal information, automating a sales force, managing customer relationships and so on) they need and should select and release PDAs accordingly.

On that basis, enterprise users (ranging from large corporations to professional offices) seeking general-purpose PDAs to support pervasive computing will be better served by either Pocket PC or Palm OS choices, depending on their specific needs, with Pocket PC devices offering overall greater expandability and power at a higher price.

Linux PDAs will find a role only in situations where companies need a highly customized or specialized system, or one in which devices will be distributed to extremely large numbers of users, such as fast-food outlets.

Meta Group analysts Jack Gold, David Cearley, Val Sribar, Dale Kutnick, William Zachmann, Peter Firstbrook, David Folger, Timothy Hickernell and Earl Perkins contributed to this article.

Visit Metagroup.com for more analysis of key IT and e-business issues.

Entire contents, Copyright © 2002 Meta Group, Inc. All rights reserved.