Foley on Microsoft
The Many Faces of Microsoft...
Could there be fewer versions of Windows? More importantly, does it matter?
At Microsoft, there's one Windows uber-brand, but there's not just one Windows. The question is: Will there ever be just one? And does it really matter to anyone who's not channeling Bob Marley's classic tune "One Love"?
Today, Microsoft offers Windows client, Windows Server, Windows Azure, Windows Mobile, Windows Phone 7, Windows Embedded Compact, Windows Embedded Standard and Windows Embedded Handheld, along with other rogue Windows-trademarked OSes. No two of these are the same. They're developed by different teams at Microsoft, sport different UIs, run on different processors and are compatible with different sets of apps.
I'm sure the 'Softies would love to consolidate at least some of those platforms. Doing so would save the company money, simplify Redmond's developer story and make the whole "three screens and a cloud" idea a lot simpler.
Look at what Apple's doing: It's making iOS its OS for smartphones, iPods, iPads, its next-generation AppleTV (as many Apple watchers are guessing), and maybe even for new Macs. I'd think -- politics aside -- that the 'Softies would love to do something similar. But because Microsoft has a far larger installed user base, not to mention tons of hardware and software dependencies, simply going "Windows everywhere" isn't as easy as it sounds.
According to my sources, Microsoft has been working for years to get Windows ported to the ARM processor (that's what the project code-named "LongARM" was all about). More recently, Microsoft researchers were assigned to the case. The "Menlo" project -- and the related "Experiment 19" team -- is, from what I hear, looking at creating some kind of mobile OS or layer that will replace the Embedded Compact core that's currently at the heart of the Windows Mobile,
Windows Phone and Zune OS. The replacement would be the same NT-based kernel that's currently in the main Windows OS. If this can be done, then Microsoft will be a giant step closer to realizing the "One Windows" vision. Given that Menlo is a research project and not even in incubation, however, it may take a long time for any of the fruits of that effort to make their way into commercial Microsoft wares.
A nearer-term step Microsoft could take toward creating a "One Windows" world is an organizational one. Currently, the Windows Embedded and Windows Mobile and Phone teams are not part of the Windows client or server business units. Until very recently, those teams were part of the Microsoft Entertainment and Devices (E&D) unit. In May, concurrent with the resignation of E&D chief Robbie Bach, CEO Steve Ballmer moved the mobile teams to report directly to him as a stopgap measure. There have been rumors since the start of this year that Windows client President Steven Sinofsky is making a play to get the mobile teams moved under his oversight. If that happens, Sinofsky would end up owning Windows, Windows Live and the Windows Mobile businesses.
If either or both of these scenarios come to pass and Windows and Windows Mobile are united, there will still be other Windows OSes out there that aren't part of the "One World."
Windows Azure and Windows Server, which are becoming more and more closely intertwined -- both organizationally and feature-wise -- are built on the same NT kernel as the Windows client is today. But the server and cloud offerings seem to be moving on a separate trajectory. Microsoft has finally gotten Windows client and server in lockstep, releasing new versions of the OSes and service packs for them at the same time. These days, though, it's the Windows Azure team and the Windows Server team, more so than the Windows client and Windows Server teams, which are sharing code and features (with new features debuting first in Windows Azure and later appearing in Windows Server).
So, maybe the idea of "One Windows" for one and all isn't really useful or feasible. But it sure seems like two versions could do the trick. Do you think Windows consolidation is coming and needed? Or is more better, even when it comes to OSes?
Mary Jo Foley is editor of the ZDNet "All About Microsoft" blog and has been covering Microsoft for about two decades. She has a new book out, Microsoft 2.0 (John Wiley & Sons, May 2008), about what's next for Microsoft in the post-Gates era.