Foley on Microsoft

Don't Write Windows' Eulogy Yet

It took years to build the Windows empire, and Microsoft won't stand idly by watching it be dismantled.

• By Mary Jo Foley
• 02/01/2010

Windows 7 was the last of the big Windows releases, and Microsoft's dominance was done.

I think differently. It took decades to build the Windows empire.

Microsoft's not going to stand idly by watching it be dismantled. There are still big Windows releases ahead. Xbox and Bing aren't going to pay the bills any time soon, and Microsoft isn't going to let Windows fade into the sunset.

So what are the Windows -- and other Microsoft platform teams -- going to do in the next year ... or five or 10? (By the way, when I say "platform," I mean everything OS-related that's in the works. That means Windows CE/Embedded, Windows Client, .NET, Windows Mobile, Windows Server, Windows Azure and various Microsoft Research systems and networking projects in the OS space.)

In the near future, the status quo will continue. Microsoft will roll out a new Windows client release (Windows 8) and another (Windows 9) over the next few years. I'm betting we'll see a new Windows client update every two to two-and-a-half years -- regardless of how much padding Microsoft execs place around their dates so they won't miss an official deadline. That would mean Windows 8 should debut around 2011, if my powers of prediction are strong.

From what I hear, there are distinct factions inside the Windows organization. One camp is ready to attempt another, even more major Windows release affecting the kernel and lower-level pieces of the platform. Others favor regular, minor feature improvements and tweaks around the periphery -- such as an overhaul of Windows Update and Group Policies. In either case, I'm betting Windows is going to make an ever-increasing use of virtualization to maintain backward compatibility of applications regardless of how much of it becomes managed code.

Speaking of managed code, I expect in the next five years another entirely new OS that's more suited to running on multi-core processors and across distributed architectures. Exactly how this will happen is murky, but the company has had an incubation project over the past few years -- the infamous "Midori" -- dedicated to exploring these areas. Meanwhile, Microsoft Research is dabbling with a number of OS-focused skunkworks efforts, such as the Singularity microkernel operating system that's the foundation for Midori; the Barrelfish and Helios projects; and the Gazelle browser and OS project, which could end up being a competitor to Google's OS.

While Windows is -- and will remain -- the starting point for all things platform-related at Microsoft, there will be regular updates to Windows Azure, Microsoft's cloud operating system, after it officially launches in February 2010. Expect Microsoft to keep hammering away at customer choice in the next year or two as it fleshes out its private cloud strategy; for example, the promised secure cloud data center gateway, code-named "Sydney."

On the mobile front, the only thing I feel safe in predicting is that Microsoft isn't going to abandon -- or open source -- Windows Mobile in the coming decade. (Whether or not the 'Softies should do either of these things is less clear.) The oft-promised Windows Mobile 7 should finally ship on new phones by the end of this year. Hopefully Microsoft will give phone makers more detailed chassis specs and a new updating mechanism so new Windows Mobile updates can be released more frequently and efficiently in the coming years. Will these moves stop the hemorrhaging of Windows Mobile market share? I doubt it.

What about you? What do you think Microsoft will do in the Windows space in the coming year ... or in five or 10 years?