Last summer, Microsoft said it was 'changing strategy' with its mobile OS; instead of trying to sell phones to everyone, it was planning to concentrate on reaching fans with flagship devices and on reaching businesses with a secure phone that's good for productivity.
It's an open question whether Microsoft can manage that and the significant drop in phone sales revealed in the company's latest financial results has naturally led to questions about the future of Windows 10 Mobile.
It's certainly too early to say whether Windows 10 Mobile is going to appeal to business, with only one handset out that can support Continuum, and the FIDO standard (that will use the iris scanner on the new Lumias to log you in to enterprise applications and consumer services without needing passwords, even apps running on your PC or Mac) isn't finished yet.
In very many ways, Windows 10 is still a work in progress, with an annoying number of regressions (from issues with offline Maps navigation, to not being able to record audio in OneNote, to not being able to open some forwarded messages in Outlook, to problems with the swipe keyboard in third party apps, to not being able to sync open tabs between your phone and PC browser... to mention only the ones that particularly annoy me).
The immaturity of Windows 10 Mobile is a casualty of re-implementing the OS on yet another platform, and of the unpopularity of Windows 8.
When Microsoft moved from the Windows CE underpinnings of Windows Phone 7 to the NT kernel for Windows Phone 8, it kept the same interface, with the signature Windows Phone design, making for an impressively smooth transition (as long as you were prepared to buy a new handset).
Because the Windows 8 interface was perceived as such a failure on PCs (although I'm far from the only Windows 8 fan, I'm not in the majority), Windows 10 Mobile had to change the design as well as the internals and all those extra moving pieces have left their mark.
Personally, I'm waiting the same four months it took Windows 10 to stabilize from when it shipped in July to the far-more-mature November update before I make up my mind -- and until the OS is ready for OEMs like Acer and Alcatel to start shipping their handsets.
But in many ways, the future of Windows Mobile is the future of Windows Store apps, and they're not going away whatever happens to Windows Mobile. The Universal Windows Platform, as Microsoft calls the latest version of what we've previously called WinRT, Metro, modern, and Windows Store apps, means the Netflix app that works on Windows 10 and Windows 10 Mobile will also work on Xbox and even on IoT devices (why can't the screen on your smartfridge show you that there's a new episode of House of Cards to watch?).
The appeal of writing for all those different devices at once (with plenty of encouragement from Microsoft, no doubt) is getting some new partners onto Windows 10, like the Ring video doorbell. If that really succeeds, it will deal with a lot of the 'app gap' -- and Microsoft needs it to succeed, for the future of Windows as well as the future of Windows phones.
Windows needs to be mobile. It needs to be a credible tablet OS as well as a strong desktop and notebook OS (and while a tablet OS and a phone OS aren't exactly the same, we increasingly treat phones as computers that fit in our pockets rather than something that's first and foremost for making calls and sending text messages).
IDC's predictions for the PC market in 2016 show a drop of 3.1 percent -- unless you count 2-in-1 PCs and detachable tablets like the Surface as PCs, in which case it's going to grow 1 percent or 2 percent.
Leaving aside the question of why a tablet with a removable keyboard isn't a PC even when it runs Windows, if you look at the predictions and Apple's latest sales figures, it seems that smartphones, PCs, and tablets have something fundamental in common.
Whether it's a PC or a tablet or a smartphone, we keep buying new devices until we get one that's good enough and then we keep it until it needs replacing -- and the devices that have come out in the last couple of years are essentially good enough that we're not rushing to replace them.
Tablets that have their own keyboard when you need it (which is what I think is important about so-called detachable tablets) are a growth area, probably because they're not yet common -- but also because of the continuing trend to mobile devices. Windows needs to be a mobile OS to handle tablets -- and phones and phablets come along for the ride.
But Windows also needs 'mobile' apps, even if you're using a traditional notebook or a desktop (or a compute stick or any other form of PC), because the attack surface in traditional Windows applications is just too big. Viruses are too easy to install in the familiar software world. And desktop software is harder for Windows to control in ways that, say, extend battery life on mobile devices. Modern, mobile apps can run a service that stays connected the way an app on a smartphone stays connected without using enough CPU to dent your battery life.
There are always going to be big applications like Excel and Photoshop and CAD software, but for a lot of what people do on computers, the sandboxed apps of mobile platforms make life much safer. If the Edge browser can run as a sandboxed UWP app, what we've thought of as mobile apps can be more powerful.
Win32 applications aren't dead or dying, but WinRT and UWP apps are the future of Windows app development. Windows Mobile can be the beneficiary of that, giving Microsoft the option of its own mobile OS alongside the strategy of putting Microsoft apps and services on iOS and Android -- as long as Microsoft can turn it into a platform that's efficient for the Windows team to create without needing it to be a major hit overnight.
That way, there's the chance of it staying the course and -- just maybe -- becoming a sleeper hit once it matures, because there are still plenty of great features in Windows 10 Mobile struggling to be noticed.