It has been Microsoft's goal ever since it dumped Windows Mobile: to deliver an operating system kernel that worked just as well on a handheld device as on a laptop. Back in 2008, that goal seemed rather ambitious. And as we saw with Windows 8, it's been something the company has never completely achieved.
As Windows Phone developers discovered, that goal may finally be happening. On Monday, the Windows Phone Store posted a pre-release operating system downloader app called Phone Insider, evidently for testing Windows 10. The night before, certain select developers who had already been testing pre-release builds of Windows Phone 8.1 learned their test update platform had been replaced. The replacement app directed those developers to join Windows Insider, Microsoft's official Windows 10 testing club.
Microsoft spokespersons contacted yesterday declined to acknowledge the Phone Insider app even exists, despite the fact it can be downloaded from this page.
One company representative went so far as to say the app "hasn't been released to the public," referring us to a tweet by Gabriel Aul, the company's executive vice president for devices. That tweet said in its entirety, "We'll talk more about the Phone Insider app on 1/21."
Perhaps the posting caught company representatives unaware. But even if so, it did get people talking again about the "Windows Everywhere" goal, and whether it still mattered. In the wake of Windows 8's rejection by many consumers, experts are asking whether Windows 10 on every device is something users actually want, or even need.
"Microsoft has achieved a long-sought technical milestone in getting the same operating system running on so diverse a set of hardware," said Ross Rubin, principal analyst with Reticle Research. "Actually, the company had also said Windows 10 would run on IoT-class devices with no displays at all. It should represent a win for developers wanting to target laptops, tablets, and phones."
For developers, sure. But what about consumers, whose attention remains fixed on the wonderful world of handhelds? Rubin concedes that it's the apps that really hold their attention more than the devices themselves -- most importantly the apps with a clearly defined mobile context. "Take an app like Instagram. On a laptop, it would likely be a lot less 'insta,'" he said.
"The move to one [Windows 10] operating system comes with a mix of consumer pull and supplier push," Rubin added. "If one is using a particular user interface or app on their laptop, they benefit from familiarity as they move to other devices. It's a balance between enforcing consistency and optimizing for purpose."
According to Jackdaw Research Chief Analyst Jan Dawson, the entire operating system won't be the same on all devices. "The core of the OS is the same everywhere, but the user interface and specific functionality will vary somewhat from device to device.
"Windows 8 went too far in this direction," Dawson added, "imposing the same sort of user interface on all devices; and Windows 10 will dial that back somewhat while maintaining the benefits of a common architecture and development tools."
Windows 8 and 8.1 tried to present a front end that was similar enough to their Windows Phone counterparts that users would consider them effectively equal. That plan failed, spectacularly. While a respectable plurality of tablet users don't dislike the Start Screen, many PC users have opted to stick with the familiarity of the Windows 7 Start Button instead.
What we don't actually know at this point, and can only speculate upon until a week from tomorrow, is whether the front end of Windows 10 will look the same on every device.
"How they'll make the UI work is going to be interesting," said Al Hilwa, IDC's program director for software development research. "I suspect there will be a UI switch which can give the device different personalities, so that existing users of Windows Phone or Surface can feel at home."
Hilwa said he expects to see a kind of "über-UI" that extends the functionality of the existing desktop, giving "Modern" apps (the class designed for the WinRT runtime) more of a functional context alongside .NET and other Desktop applications. But this UI, he believes, would have to enable Windows 10 to present different faces on different screens, effectively preserving the distinction between mobile and fixed contexts.
"Microsoft not only hopes to keep Windows users comfortable outside traditional PC form factors," said Ross Rubin, "but offer a more unified solution compared to the two OS approaches that Google and Apple are fielding on desktops and laptops versus tablets and phones." He noted how much further along Google has come in extending Android into the realm of TV, than Microsoft ever did with Windows.
Whether users care that the two contexts are being bridged by a single platform, or a single brand, or a single patch of duct tape, remains the pressing question. The fact that iOS is not the MacBook's operating system doesn't appear to have hurt Apple all that much.
"I'm skeptical that the universal approach really solves many problems for Microsoft," stated Dawson, "because most of the apps missing from Windows Phone aren't the sort of thing that's on the PC version of Windows. It does increase scale for the overall platform, but I don't think it'll really help to attract more developers to the mobile side specifically."
In a post to his firm's blog last October, Dawson suggested that Microsoft produce one merged Windows for mobile devices, and eliminate license fees for that market segment. Then it could be bold with Windows 10 for desktop computing devices, if it continues to sell Windows 8.1 as an alternative.
We may not know how Microsoft plans to resolve that issue, even after January 21. "I suspect pricing information and other production details to be internally debated right up to launch," noted Al Hilwa. "The rule in rolling out software is never to decide something like that before it is necessary."
Bridging the two worlds
Yet there is also an emerging consensus among experts that a single Windows 10 platform for consumers should also mean a single platform for applications. Granted, there may always be the need for mobile and office contexts, as Rubin points out. But a strong runtime platform could help applications bridge the gap for themselves, adopting a mobile context when they're used in a mobile setting.
The strongest case study for this theory involves Office. From the day Windows 8 was announced, Microsoft deferred the question of when separate "Metro" or "Modern" versions of Word, Excel, and PowerPoint would emerge. To this day, they have never shown up. The original Microsoft Surface and Surface 2 devices, which ran the Windows RT system, presented desktop editions of these applications instead. Surface Pro 3 comes with the desktop version of Office 365.
Experts believed that the "Modern" environment, with all the square tiles, could only prove its worth once Office applications appeared there. Instead, Microsoft shifted its attention to building Office for iPad. The first editions of those tablet apps actually ended up looking comfortably familiar to veterans of their Windows desktop counterparts. That familiarity became vital to those apps' success on iOS.
Had the same similar-looking apps appeared in Windows 8's "Modern" environment alongside the Desktop, experts and everyday users alike might have questioned the need for both.
For Windows 10 to succeed on all platforms, it may need to provide Office and other applications with the means to scale up or down, depending on where it's running. Dawson believes "that's exactly how it will work: same code base, but with instructions for how to display it on different devices."
This way, there would be one Word, one Excel, and fewer than three browsers. Amid the continued presence of two Internet Explorers in Windows 8 and 8.1, rumors of a third Microsoft browser have actually been met with skepticism in many quarters, some of whom are fine with just one ... and a growing number of whom are fine with it being Chrome.
Windows 10 will need to be more than just a single platform. It will need to provide both developers and users with a gentle and even enjoyable way to move between device contexts. To the extent Microsoft fails to pull this off, its January 21 unveiling event could prove a disappointment.