How to Make Sense of Microsoft's Multiple Mobile Operating Systems

Let's say you're not convinced that the Apple iPhone or a Google Android OS-based device is appropriate for your business or as an enterprise app platform. You believe Microsoft isn't yet roadkill in the mobile world, so you decide to bet on the new Windows mobile OS. Which one do you choose?

It turns out that Microsoft has four "Windows" mobile options, none of which is compatible with the others. Apple and Google both have or are working on unified mobile operating systems -- iPhone OS (renamed "iOS" yesterday by Apple) and Android OS, respectively -- meant to run on smartphones, slates, and other mobile Internet devices (MIDs). But Microsoft has decided that a one-size-fits-all approach is a bad idea, as different devices are, well, different.

[ Get the best iPhone and iPad apps for pros with InfoWorld's business iPhone apps finder. | Keep up on key mobile developments and insights with the Mobile Edge blog and Mobilize newsletter. ]

Last week, Microsoft announced the beta availability for the fourth of these mobile "Windows" OSes: Windows Embedded Compact 7. Does this four-way mobile strategy make sense? I'm not sure, but it helps to understand what each of the "Windows" mobile OSes actually is. (By the way, I use the quotation marks because only one is derived from the desktop Windows OS; the others use what a Microsoft spokesman calls "broad Windows technology.")

Windows 7 for Slates

Several of the iPad wannabes recently announced, such as the one due from Asustek in early 2011, use the same Windows 7 you'd find on a desktop or laptop PC. The idea is that a Windows 7 slate is simply a PC in the slate form -- no different than the many tablet PCs over the last decade that ran Windows XP or Vista, depending on their era. Such devices run Windows and all its apps, as well as any compatible hardware devices, providing users the maximum capability, says David Wurster, senior product manager for Windows Embedded.

Unlike XP and Vista, Windows 7 doesn't require a special "pen and tablet" edition of Windows; the regular Windows 7 includes the libraries that let both pen devices and touch-based gestures work. Unfortunately, my testing of Windows 7 on touchscreen PCs left me cold. The number of gestures supported is small, and most applications aren't designed for finger-based touch input. Simply put, running Windows and Windows apps via touch is a painful experience, no matter how great the tablet hardware itself may be.

Microsoft had made a lot of noise in 2007 about its Surface technology that was supposed to revolutionize touch input and make slate tablets the next big thing, but Microsoft quietly let the technology drop in late 2008, and Windows 7 ships with the remnants of the Surface dream. Microsoft declined to discuss its touch technology last September when Hewlett-Packard began aggressively marketing its touchscreen PCs, with its PR firm referring me to company blogs that hadn't been updated in nearly a year and that noted further development on the touch UI was unlikely.

Last fall, Microsoft said that just 5,000 Windows 7 touch SDKs had been downloaded in the previous 18 months by 250 companies; by contrast, more than 100,000 iPhone OS SDKs were downloaded in the first four days after its June 2008 launch. It's clear developers don't believe in Windows 7 touch, either.

Every version of a Windows tablet has been a dismal failure, and although companies such as Asustek may be hoping for the iPad mania to transfer to their planned Windows 7 slates, I wouldn't hold my breath. In fact, I doubt they'll actually ship these products -- such vaporware announcements are hardly rare, and OEM companies typically have fast product cycles that make any announcement of a product to ship in six to nine months quite suspect. At best, it means they're starting to work on their development.

Windows Embedded Compact 7 for Custom Devices

Microsoft's announcement last week of Windows Embedded Compact 7 mentioned its potential use in slates, making me wonder whether Microsoft was remedying the Windows 7 touch deficiencies by coming out with a new mobile OS specifically designed for touch-based slates, to better compete with the Apple iPad and the slew of Android-based slates expected this fall.

Not quite -- Microsoft's Wurster explains that WEC7 is essentially the next version of Windows Embedded Compact Edition (Embedded CE), a Windows Mobile-derived OS meant for a variety of devices, such as the signature pads a deliveryperson might have you sign and set-top boxes. It's not a version of Windows 7 but a new version of Embedded CE.

The good news is that it could be used in slates, smartphones, and MIDs. It also has an extensible core gesture library for touchscreen devices and a Silverlight-derived presentation layer for the kind of rich interactivity the iPhone has led us all to expect in mobile devices.

The bad news is that because it is designed for embedded devices, which are usually custom creations, every version could be different, based on what parts of WEC7 a device maker chooses to implement. You won't get a broad platform for which application developers can write to, nor can you expect various WEC7 devices to have the same capabilities or interfaces. It's up the device maker to tailor WEC7 to its needs, Wurster says. That's similar to the freedom that Google's Android OS and Nokia and Intel's MeeGo provide, with the concomitant possibility of intense fragmentation.

However, there's no reason that a hardware maker couldn't create a "universal" version of WEC7 for multiple devices, Wurster says, as Apple has done with the iOS for its iPhone, iPod Touch, and iPad. An Asustek or Acer could use WEC7 as the basis for a family of touch devices that application developers could count on for application interoperability. In addition, HP has decided, through its acquisition of Palm's WebOS, to take the same route as Apple and produce a family of devices built on a common OS.

Thus, WEC7 is probably Microsoft's best hope for a Windows-derived slate that could compete with the iPad. But it would take remarkable discipline from the flavor-of-the-month OEM hardware makers to create an ecosystem that would support application developers sufficiently to create a meaningful alternative to the iPad, Android devices, or even WebOS devices. By not asserting control, Microsoft is likely relegating WEC7 to devices you never realize run Windows technology.

Windows Phone 7 for Smartphones

In March, Microsoft revealed it was jettisoning long-sinking Windows Mobile for a new OS called Windows Phone 7 that, despite its name, is not a version of Windows 7 but instead a new operating system that relies on Silverlight and the Xbox's XNA as its main UI technologies. As with WEC7, it uses "broad Windows technologies" but is not Windows.

With Windows Phone 7, Microsoft is asserting control over the whole platform, including hardware specs and physical user interface. That's meant to correct one problem with Windows Mobile: the fact that every device was different and often had a customized UI. Developers simply couldn't count on their apps working on more than one specific device, and businesses were frustrated that every device was different, which made training and support a nightmare.

Windows Phone 7, based on a glimpse of a prototype, has a radically different UI than Windows Mobile and indeed any smartphone operating system now available (Palm's WebOS is the closest). Microsoft says the device will be very heavily oriented to social networking but will also support corporate security and management needs as Windows Mobile did quite well.

From what little Microsoft has said, Windows Phone 7 will be a consistent platform that users, IT, and developers can count on no matter who makes a specific device. But there are so far no plans to extend Windows Phone 7 beyond the smartphone, Wurst notes; it's WEC7's and Windows 7's role to address other mobile needs.

Windows Phone (Kin) for 20-Somethings' Phones

Last month, Microsoft shipped two Windows Phones (no "7" in the name) under the Kin brand, designed for social networking use and aimed at 20-somethings. The OS is again not the same as any other Windows OS, so it leaves developers with yet another application development choice. Additionally, the OS is not extensible to other devices, so it can't be a platform for slates.

But it doesn't really matter that the Kin OS is different and nonextensible. Although there are some interesting concepts in the UI, the Kin devices and their operating system have been uniformly panned as awkward to use, with inconsistent interfaces and limited capabilities. For example, there's no security or manageability, so Kins can have no place in many businesses. I doubt Kin will survive the year, especially given the forthcoming Windows Phone 7's social networking orientation. It certainly has no reason to exist.

Four Slate Options, Two Iffy Choices

When all is said and done, of the four Windows-derived mobile operating systems, only two might make sense for slates and iPad-wannabe devices: Windows 7 and Windows Embedded Compact 7. Windows Phone 7 applies solely to smartphones, as does Windows Phone (Kin), so they ultimately matter just for smartphone-only applications.

Out of the box, Windows 7 is simply unusable on a slate. Device makers would have to hide most of the operating system, create a good touch UI overlay, and somehow allow only touch-enhanced applications onto it. HP tried that to some extent with its touchscreen PCs and its encouragement of touch app development, but it's gotten little traction beyond kiosk-type uses.

HP has sinced moved onto WebOS, and I can't see OEMs such as Dell, Asus, and Acer that do little original development taking on the challenge of defining and growing an ecosystem -- especially knowing Microsoft could throw a monkey wrench at any time by updating Windows 7 in incompatible or competitive ways. It says a lot to me that these companies are playing with Google's Android operating system, as is the only other OEM to have serious OS-level software effort (HTC, with its Sense UI).

That leaves WEC7. It has the same ecosystem problem as Windows 7, but in WEC7's case, that's fundamental to its purpose for use in custom devices. The same reasons that make a touch-friendly Windows 7 ecosystem unlikely also apply to WEC7 with one slight difference: Because WEC7 is meant to be customized, there's less risk that Microsoft will create compatibility or competitive problems for a WEC7-based mobile ecosystem.

Thus, an HTC, Asustek, Acer, or Dell could theoretically develop a multidevice version of WEC7 that serves as the basis for a Windows-based slate ecosystem. It's even possible that a consortium of companies could do that; an example is the Wi-Fi Alliance, which created a compatibility standard for the IEEE 802.11 protocol. Before that, there was no guarantee the devices would interoperate; today, you can't find non-Wi-Fi 802.11 devices.

I don't think chances are great this will occur, but it is possible. What Microsoft should do, of course, is either extend Windows Phone 7 to multiple devices as Apple did with the iOS or create a "platform" version of WEC7 for devcies such as slates, while keeping the existing customizable WEC7 for those embedded devices such as set-top boxes -- Microsoft could have its cake and eat it too.

This article, "Making sense of Microsoft's mobile OS four-way," was originally published at Read more of Gruman et al.'s Mobile Edge blog and follow the latest developments in mobile computing at

Subscribe to the Best of PCWorld Newsletter