META Trend: Through 2009, Windows-based personal computers will remain the dominant computing device for corporate and consumer users (80% market share), particularly for information creation and sophisticated collaboration activities. However, smart phones and information appliances will play an increasingly important role as users exploit multiple devices for information consumption, entertainment, and basic collaboration needs. As the line between corporate and consumer computing continues to blur, IT organizations will face an increasingly demanding and technically savvy user base, mirroring the initial PC revolution.
The market for mobile phone operating systems is changing. Indeed, within the next two to three years, we expect to see a significant number of new cell phones coming to market, running embedded Linux as the OS (e.g., MontaVista), particularly at the low end of the spectrum where cost is paramount.Within five years, Linux-based phones are expected to account for 20%-25% of the mobile phone market. This is a leap from nearly zero share currently.TextThis will be particularly true for phones in emerging markets (e.g., China, India) with greenfield local ODMs/OEMs that are less tied to a specific OSs (from historical use or staff expertise) and more interested in developing and designing the lowest-cost devices (because engineering costs are relatively low in these geographies, the penalty in custom design costs is low).
Indeed, within five years, we expect Linux-based phones to account for 20%-25% of the mobile phone market (increasing from near zero currently). While the initial thrust of Linux will be exhibited in the low end of the market, we expect to see an increasing migration to Linux for feature phones and more capable smart phones, as well as adoption by major vendors (e.g., Motorola, Siemens, Samsung, NTT DoCoMo).
The majority of current feature-based smart phone designs are powered by the Symbian OS, with a smaller number of Palm and Windows Mobile-based devices. We expect fundamental design changes in many devices that will separate the underlying OS from the UI. Indeed, PalmSource has already announced it will essentially exit the embedded OS market for phones and instead build the Palm environment on top of a Linux platform (expertise it has recently acquired through its acquisition of China Mobile Systems).
Furthermore, Nokia is making its Series 60 and Series 90 platforms available on a foundation of Symbian OS, moving them up into the UI environment and isolating them from the embedded OS. Microsoft will be the lone long-term holdout for integrated UI/OS with Windows Mobile (we do not expect to see a splitting strategy from Microsoft anytime soon).
There are some compelling fundamental reasons why PalmSource and Nokia would move in such a direction. The core OS must be capable of supporting all of the hardware features of any given device. This requires significant effort in driver designs for each and every device (e.g., screens, cameras, Bluetooth devices).
Extracting the UI to one level up from the embedded OS enables a vendor’s end-user platform to be available on multiple devices without having to test specifically on each device. This allows the vendor to offer a common application development platform without having to offer a custom version for each device. This offers a significant amount of portability to applications, especially important to building a UI ecosystem of enhanced and feature-rich end-user applications. PalmSource is now migrating to this path. Nokia has also taken this road, allowing Symbian to do the hard work of making sure devices are enabled with the proper drivers, etc. at the OS level, while Nokia concentrates on its UI (Series 60/90).
This is not only a smart move for the vendors (e.g., PalmSource, Nokia), but also a positive development for the enterprise environment. By extracting the OS and its non-common features across the various devices, it is possible to come closer to “write once” for the UI and not have to deal with custom versions for each device – though class differences like I/O navigation and screen size may still pose a challenge. Furthermore, upgraded hardware will have minimal or no impact on the developed applications (e.g., moving to an enhanced version of a smart phone).
It is important to note that Microsoft continues to follow the path of integrated UI/OS with its Windows Mobile offering. While this makes it easy to write apps for the combined UI/OS, which will work on all devices, it will limit Microsoft’s market opportunities, because OEMs that might be inclined to take the UI portion of the Windows Mobile environment and layer it on an embedded OS (e.g., Linux) will be unable to do so.
Instead, they will be forced to design the entire device for Windows Mobile, thus requiring each device to be optimized for the OS, with the subsequent need to create/deliver appropriate drivers, etc. Although we do not expect Microsoft to change strategy anytime soon, we believe it would benefit Microsoft to move in a similar direction to PalmSource and Nokia if it wants to expand its currently limited market presence in the mobile phone market, particularly at the lower end.
We also expect RIM to provide a full Blackberry client for deployment on Linux-based devices within the next one to two years. It has already moved to separate the UI from the OS by licensing clients to PalmSource and Nokia, among others. Providing a Blackberry client/UI for Linux should be a fairly easy undertaking for RIM.
The move to separate the UI from the OS further enforces our position that enterprises should select a particular platform and not a specific device for long-term strategic deployments. By selecting a platform (e.g., Palm, Nokia Series 60/90, Windows Mobile), the organization can concentrate on building apps that work on almost any device supporting that platform, and let the UI interact with the underlying OS to secure operability. Companies should endeavor to set standards on platforms that can be deployed for the long term and that fit within the strategic direction of the organization (e.g., a proclivity to things Windows, a strong preference for Java/J2ME).
Bottom Line: A fundamental shift is taking place within the mobile phone market. Companies will see much more emphasis on UI platforms and less on the underlying OS. This is good news for companies defining an enterprise mobile application strategy.
Business Impact: Mobile smart phone devices are gaining in popularity, but companies often struggle to identify a true ROI for deploying the devices.