When Microsoft announced plans to release a version of Windows for ARM processors, it created a lot of work not only for itself, but for all the independent software vendors who sell Windows software as well.
Microsoft will need the support of these software developers to make the ARM version of Windows a success, warned Dan Olds, principal analyst of the Gabriel Consulting Group.
[ Get all the details you need on deploying and using Windows 7 in the InfoWorld editors' 21-page Windows 7 Deep Dive PDF special report. | Stay abreast of key Microsoft technologies in our Technology: Microsoft newsletter. ]
"It's not just Microsoft moving to ARM, but Microsoft also must get all the other [software developers to follow suit] in order to have the ecosystem its wants," Olds said. "It has to have apps from everyone else."
On Wednesday at the Consumer Electronics Show in Las Vegas, Microsoft CEO Steve Ballmer announced that the next, as-yet unnamed, version of Windows will be available for the ARM chip architecture.
Thanks to its low power consumption, the ARM architecture is widely used for battery-driven portable devices. The market for such tablet devices has been exploding, driven by sales of Apple's iPad. Other manufacturers are finalizing or have just released tablets running on Google's Android operating system. Research firm Gartner has estimated that 54.8 million tablets will be purchased in 2011. Yet Microsoft has been noticeably absent from this market.
The next version of Windows, due in 2012, will attempt to rectify this situation.
Microsoft engineers have a lot of work ahead of them, Olds predicts. The ARM instruction set is very different from the x86 instruction set that Windows now runs on. And because ARM processors are not as powerful as x86 ones, the engineers will have to be more careful as to how the operating system consumes resources.
But crafting a version of Windows for ARM is only the first challenge facing Microsoft. Another one is getting software developers to rewrite their Windows applications to run on ARM. "For ISVs, it will not be trivial to port applications to a new platform," Olds said.
Yet software developer support will be essential for Microsoft's success. The success of any operating system depends on the number of applications that have been written for it. The applications were what made Windows a success in the first place, Olds said.
As an example, Olds pointed to Microsoft's failed support of another chip architecture. In the mid-1990s, Microsoft offered a version of Windows NT for Digital Equipment's Alpha chip, which had a 64-bit reduced instruction set computer (RISC) architecture. That OS never gained popularity, due at least in part to the lack of applications that were developed for that platform, Olds said.
Apple itself faced a similar challenge in 2005 when it announced it was switching to the Intel processors for its Macintosh computers. Apple was successful in moving its own software developers over to the new architecture, and it has been one of only a few companies ever to survive a switch of platforms. Olds attributes Apple's success to the fact it kept in close contact with its of its software developer community, which was small to begin with.
Microsoft's task of getting its software developers interested in porting their software to ARM will be an order of magnitude larger than Apple's. There are many more Windows software vendors that could supply software. "How do you get them to switch and get them to do it right?" Olds said.
That's the challenge that awaits Microsoft.