Microsoft’s device-side OS strategy may be in chaos at the moment, but there’s a strange symmetry about it too. On one hand, it looks set to collapse its three-tiered Windows platform into just one operating system; on the other, it has launched its Xbox One console with three OSs to replace just one. Ironically enough, it is the three-sided Xbox platform which makes more sense, and reinforces our view that one of the big mistakes of recent mobile and multiscreen efforts has been the failure to put the gaming system at the heart of the whole Microsoft consumer experience.
Clinging too hard to Windows has been a hallmark of the Steve Ballmer years, and a new CEO may be braver about smashing sacred cows, but when Windows Phone made its debut, there were strong signals that Microsoft would increasingly converge the Xbox and mobile OSs and applications platforms, and eventually the gaming console, its remit extended to all kinds of media activity, would be the hub of user experience and branding for all devices.
Instead, Microsoft reined in its ambitions for Xbox, despite the success of the offering, and although it added various media and video capabilities, plus more cloud services, the product remains perceived as a console, not the heart of the digital home. Instead it focused its attentions on penetrating the ARM-based device space, which was encroaching so rapidly on Wintel PCs – but not with a radical new approach and OS, but by adding yet another member to the tired Windows family.
To be fair, Microsoft had devised a highly differentiated and modern user experience for Windows Phone (after years of disaster with Windows Mobile), but this has made limited headway against Android and iOS. It certainly has not replicated the success that Xbox had against Sony and Nintendo in the gaming world. But instead of extending the flourishing platform to new devices, Microsoft chose to extend the struggling Windows Phone, in effect, to large-screened post-PC products like tablets. Not a terrible idea if the company had actually done that, but instead it came up with Windows RT, which shared the user interface and some of the development tools of Windows Phone, but was a separate OS. Since it was also segregated from the other new release, Windows 8 for x86 ‘ and lacked that option’s backwards compatibility with existing Windows apps ‘ confusion ensued, among OEMs and developers, let alone consumers.
Now it seems that Microsoft will back away from Windows RT. This speculation was sparked by Julie Larson-Green, EVP of devices and studios at Microsoft, who told a UBS technology conference in California: “We have the Windows Phone OS. We have Windows RT and we have full Windows. We’re not going to have three.”
She did not, of course, divulge whether Microsoft would revert to just one device-side Windows flavor, or ‘ perhaps more likely ‘ converge WP and WRT into a single ARM-based OS spanning all device types in the same way Android will (though reports that Google is to introduce a Nexus tablet running Chrome OS could spell muddying of those waters too ‘ see separate item). And while most of the headlines focused on the possible demise of Windows RT, her comments also pointed to a new project up ahead, one which might finally cut the ties with Microsoft’s venerable OS.
Of course, Larson-Green will indeed be focused on several timescales ‘ the near future and how to undo the mess of Windows; the medium term alliance, or even convergence, of Xbox and Windows; and longer term, what platform comes next, one which will, perhaps, tear up the rulebook at last and be designed around the big trends of the cloud and the internet of things. The latter is another area of fragmentation in Windows ‘ currently the various embedded and CE flavors are confined to niches and do not need too much common ground with the main operating systems, but that will change as Microsoft looks, like Google, to extend a unified apps platform and connectivity hub beyond consumer devices to cars, embedded gadgets and even wearables.
For starters, Microsoft would do well just to make a clear statement of direction on the first category, PC/post-PC/mobile Windows directions for 2014, at a time when OEMs for all the flavors are restive and uncertain, and when Intel is throwing its weight increasingly behind Android.
Full merger of the two ARM flavors will certainly make sense once Microsoft seals the deal to own the device business of Nokia, the main vendor of Windows Phone products, an acquisition which will end some of the conflicts of interest in the WP platform (how far Nokia and Microsoft’s own Surface would be allowed to cannibalize one another, for instance). There is certainly no indication that Microsoft wants to get rid of a sealed-case mobile OS altogether, but it does not really need two of them. Larson-Green said: ‘We do think there’s a world where there is a more mobile operating system that doesn’t have the risks to battery life, or the risks to security. But, it also comes at the cost of flexibility. So we believe in that vision and that direction and we’re continuing down that path.”
But she was keen to present WRT in the proud tradition of Microsoft making several attempts at an product before getting it right, saying it was the firm’s “first go” at creating a turnkey, closed system tied to a particular piece of hardware, like iOS. She admitted there had been insufficient differentiation between Surface RT and x86-based Surface Pro and that, perhaps, WRT should not have been called Windows, as it was trying to do something different.’
Those with the best Microsoft sources believe the company will create a hybrid between WRT and Windows Phone, rather than just killing one of them off, and that effort could take another year. Mary-Jo Foley of All About Microsoft thinks the WP OS core will be the starting point, not WRT, but this will be more flexible to work across the ‘three of four phone-like devices’ Larson-Green believes users will end up carrying, switching between them according to application. A few years from now, that group of devices will likely include wearables ‘ an area where Microsoft is known to be carrying out significant R&D ‘ and whatever the form factor, their main role will be accessing cloud content and services.
At that point, of course, the OS takes on a whole new role, and Microsoft could start to drive some of its power on the cloud platform down to its devices and client software. “We’ve been working on becoming more modern in our thinking, both in the cloud infrastructure and how you access that to build applications, and run your business, and in the operating system itself,’ Larson-Green went on. ‘And thinking about how Windows can scale from a small device to a large device up to a server, and the power that gives developers and IT professionals to manage those devices, and to give information out to people in their business no matter where they are.”
In the cloud world, the ‘operating system’ becomes a collection of APIs and tools that allow developers to target cloud apps, mobile apps and full-blown desktop software at the same time. Microsoft is already unifying its stores and developer platforms at a brisk pace, and once that happens, the differences between the OSs will become ones of branding and UI. For instance, the latest version of the WRT SDK allows Windows Store apps to access many of the features used by desktop software. Portable class libraries in .NET allow developers to split UI code from core business logic and use the core code on many different devices and spanning cloud services to desktops, and web UIs to phones. Similarly, Windows Phone 8 is based on the NT kernel and bridges various gaps with mainstream Windows, while its WinPRT development model is getting increasingly close to that of WinRT.
These are baby steps towards a fully unified platform, whether that is called Windows (for old times’ sake) or not. The end game must be a common kernel which compiles on x86 and ARM; runs on any kind of device and in the cloud; and has a single development model. That will have to embrace technologies emerging from Microsoft’s non-Windows activities, notably Azure and Xbox, and that will be the job of Terry Myerson, who in the most recent reorganization, took responsibility for all core Windows engineering across the whole product line, including Windows Phone, Azure and Xbox.
Xbox One does give a hint of how this could all evolve, by running three OSs, but united under one interface. Previous versions of the console had just one custom OS optimized for games, but as Microsoft looks to expand the functionality of its product and turn it into a general media hub, it has become more complex. It still runs the Xbox OS, alongside a slimmed-down implementation of Windows 8, which is there to support non-gaming applications like Skype and Hulu in an efficient way. Thirdly, there is a layer to connect the other two, and support fast switching between games and apps.
This is the real test of the ‘One Microsoft’ idea proclaimed by Steve Ballmer shortly before he announced that he would step down when a successor is found. Boyd Multerer, director of development for Xbox, admitted to CNET that the three-OS approach had been a ‘tremendous mindset change’ but was necessary to support a wider range of functions, and also to enable the Xbox console to be updated to support new, perhaps unforeseen, uses, without the hardware being replaced (since most users hang onto a console for 5-10 years, far longer than a PC or phone). The key is virtualization, which of course will help to make individual OSs irrelevant on all kinds of enterprise and personal platforms, allowing several to run together, each optimized for a different activity.
Implementing a hypervisor will enable Microsoft to update apps without impacting on existing games, as well as supporting the creation of Xbox applications using standard tools. The hypervisor was built by a team headed by Dave Cutler, which highlights its strategic importance ‘ Cutler designed Windows NT (and before it, Digital Equipment’s VMS), and is a very high profile figure both at Microsoft and in the industry as a whole.
This is the kind of thinking which suggests there is another important software platform left in Microsoft after all ‘ something the Windows m’nage a trois has tended to speak against. It recognizes that Microsoft may steal a few points away from Android but will never stand a chance of defeating the Google platform on its own mobile ground. It will only seize back real power if it thinks beyond traditional PC and mobile systems and tries to do something different, something that starts with the cloud and media and non-traditional computing devices. This will become even more important as tablets usurp the position of PCs, given that Android has the upper hand in tablets too ‘ and will push further into traditional Wintel turf when it has 64-bit support. It will be getting that initially, ironically enough, from Intel, which is readying its 64-bit Bay Trail Atom-based chipset for the Google OS. This will also support Windows 8.1, but Intel knows it cannot rely on that space entirely any more. It has got ever closer to Google, but is still only knocking at the gate of Qualcomm in mobile processors. However, in the large-screened world, it finally has a something unique to lay at the feet of Google ‘ a 64-bit headstart to counter Apple’s recent announcements and to propel Android into heftier devices.
Hermann Eul, general manager of Intel’s mobile and communications group, told an investor day last week that Bay Trail was ‘not only about Windows 64-bit, we’ve been talking about Android as well.’ He went on to demonstrate ‘the first ever showing of a 64-bit kernel running on Bay Trail with Android.” Full 64-bit Windows 8.1 on Bay Trail will be available next quarter and Android is likely to be close behind, as Intel will be keen to get to the market ahead of Samsung, which is also promising a 64-bit processor for the Google OS.
At the same event, Kirk Skaugen, general manager of the PC client group, was even more aggressive about taking Android into new areas, stealing a march on the mobile chipmakers and making itself strategically indispensable to Google as it looks for post-PC and enterprise power. Skaugen said he was seeing the beginnings of demand, mainly in emerging economies, for Android-based PCs, but many apps were written for small screens and do not even run in landscape mode. So, he said, ‘we’re going to allow it to scale from Atom all the way to the high end of the Core processor family. We’re going…to enable it to deliver a great experience as we go into larger scale screens, allow multi-windowing.’
This raises the specter of a replacement for the Wintel alliance which defined the PC age. So Microsoft will have to make its own new friends, and while Google and Intel address the post-PC generation, its future will rest on looking beyond even that, to the cloud or post-mobile world.