Mark Pickavance discusses Microsoft reverse course on allowing Android apps on Windows 10 Mobile
Microsoft has an entire enclosure of elephants to stable, having long exceeded the plausible limits to any room where they might hide in plain sight. And one of the biggest is the total failure of the computing public to engage with what were originally called ‘Metro Apps’ and have now migrated into being ‘Universal Apps’ since the advent of Windows 10.
Conceived as one of the pillars of the Surface/Mobile strategy, it became apparent even before Windows 8.1 arrived that people weren’t embracing them and, as a result, neither did software developers. As a result, it is generally considered that the Microsoft Store is a shadow of that offered by either Google or Apple, as it misses large numbers of the most popular apps and the quality of many it does have isn’t high enough. By way of holding out an olive branch to those that pointed this out, a series of initiatives were created that would enable the easier transition from iOS or Android for developers. These ‘bridges’ would span the app gap, and help both Windows mobile and Windows 10 to establish the universal app market. At least that was officially the plan, until Microsoft changed it. When it became apparent during Windows 8.x that the Microsoft Store wasn’t the rip-roaring success that Microsoft had hoped, it formulated a plan to help populate it with the apps that people actually wanted to use. This by-hook-or-by-crook approach was fermented around four bridging tools that would help developers move from their current environment to one where their apps could run on Windows mobile or Windows 8.x (and Windows 10). The project was subdivided like a garden fork into four prongs:• Project Westminster – for porting web apps.• Project Centennial – for porting Classic Win32 apps.• Project Islandwood – for porting iOS apps.• Project Astoria – for emulating Android apps. When this was announced, the reaction in the developer community was that the information was contentious at best, because those developers who had made a commitment to Microsoft could see what little money they made vaporising when the Store flooded with alternative Android apps. And they also wondered why they’d taken the time and effort to learn the Microsoft API, when they could have used the Android one and ended up with the same conclusion. What really upset many was that word ‘emulation’, because there is some skill needed in porting applications, however good the tools are to help you, but emulation can entirely take away the technical issues, allowing anyone to move existing apps over. In fact, according to those who used Astoria, it would take a standard Android APK file, trap all its external communications and plumb them accordingly, and the app would run on the Android emulator. This isn’t anything new to PC owners, because the Android SDK includes a basic emulator that allows Android apps to run on a PC, in a fashion. What many pointed out at the time was this one utility entirely undermined the effort needed to code native Windows apps if the tool had already been coded for Android. Microsoft’s approach to calm these choppy developer waters was to not generally release Astoria to developers. Instead, it was only provided to a small selected invite-only group, who would discuss their experience on a closed forum. After a reasonable amount of time passed, people quite reasonably wondered when Astoria would be seeing a wider release. And duly, in August, Microsoft released an early preview of Islandwood, its iOS porting tool, and promised that a final version would be included in the upcoming Visual Studio 2015 update, timed for a ‘fall’ release. And that would also be the timescale for Astoria to step out of the shadows, so developers would get an opportunity to either be enraged or delighted with it. But between August and now, something radically changed in Microsoft’s thinking about this product, according to the small number who had access to the limited release. The signals that not all was right with Astoria first started to surface in September, as people prepared themselves for the public beta. The forum created for the project was where those involved usually asked questions, and at some point in that month those questions stopped being answered. Microsoft staff specifically wouldn’t talk about the future of the project or timescales, and eventually they stopped replying to the most straight-forward information requests. Unsurprisingly, this made many people nervous, because up to this point the knowledge exchange had been free flowing, and then without any explanation, abruptly it stopped. Then in October, those brave enough to be taking Windows 10 Mobile Insider builds noticed that the Android subsystem required for Astoria to work was removed. It isn’t unheard of that features are taken out, only to be put back later in the development cycle, but when the RTM build 10586 arrived and it wasn’t in there, many quite reasonably assumed the worst. This prompted some of the few journalists that Microsoft converses with to ask what was up, and those questions weren’t answered even unofficially. Eventually, having numerous people asking it what the situation was, Microsoft issued the following statement:“We’re committed to offering developers many options to bring their apps to the Windows Platform, including bridges available now for Web and iOS, and soon Win32. The Astoria bridge is not ready yet, but other tools offer great options for developers. For example, the iOS bridge enables developers to write a native Windows Universal app, which calls UWP (Universal Windows Platform) APIs directly from Objective-C, and to mix and match UWP and iOS concepts such as XAML and UIKit. Developers can write apps that run on all Windows 10 devices and take advantage of native Windows features easily. We’re grateful to the feedback from the development community and look forward to supporting them as they develop apps for Windows 10.”This was clearly written by someone with the full intention of not actually answering the question of whether Project Astoria is dead, because it neither confirms that nor denies it. That said, from those who used it up until now, it worked in a fashion, and it wasn’t meant to be the finished article but a ‘beta’, so why is it now being classed as ‘not ready yet’?The view of many outside Microsoft is that this project has been shelved indefinitely, and Microsoft hasn’t worked out in what exact order to put those words out so it doesn’t look like yet another promise it failed to deliver on. Like most big companies, Microsoft has a small army of people who will try to interpret the worst possible actions as corporate brilliance on their part. And in covering this story, I’ve read many versions of why Astoria has been shut down, but none of them ring true. It’s been proposed that the angst of developers eventually sunk in, that there was a technical issue that Microsoft couldn’t fix, and that the legal implications were insurmountable. None of these can handle much scrutiny. The latest spin coming from undocumented sources inside the company is that the 80-man team on the project just cost too much, for a company that employs some 118,000 people worldwide. That’s a ridiculous notion given the cost implications of the Microsoft Store failing to get the apps it needs to be a real alternative to Google and Apple, because that will cost it billions ultimately. Microsoft probably spent 50 times as much on rebranding its music service to be called Groove, and lost more than that on the Surface Mini it had made then cancelled two days before launch. I’ve also heard from one source that Microsoft got concerned when some developers trying to find the limits of what the tool could do took applications written by others and converted them. If it worked that well, surely that was the point of the exercise?What none of these creative narratives do is explain how a company like this changes its mind but can’t work out to convey that simple idea. Perhaps it could try using the media of modern dance or flash cards. I’m not sure what to think about this, if I’m honest. How companies throw massive amounts of time, people and money into projects without actually thinking them through is the potent stuff of nightmares. But equally, if you realise you’re doing the wrong thing, then it’s probably the best course of action to stop rather than to shout ‘damn the torpedoes’ and plough right on. Where I have an issue with all of this is that having effectively nixed the Android part of this equation, and with no news on the iOS side of the coin to report either, were does this actually leave those important Universal App plans?On the mobile side, Microsoft has released two new flagship phones that won’t make the slightest impact on its 2% (or less) market share, and the Microsoft phone is as close to being a dead man walking as it’s possible to get. Once that is inevitably dead, and it might as well be, then who are these poorly considered full-screen apps actually for? Windows desktop owners patently don’t want them, and it seems very unlikely to change their minds on that point. In this respect, Microsoft has created a complete ecosystem without any plausible purpose, without developers or customers, and with no overarching plan as to how it can get any of those things other than some almost evangelical belief that the customer will eventually see the world the way it does, given time. By now, it should have delivered the means to package Win32 apps for the store, and that isn’t here either. I’d call that a fail. How it can blindly not see the scope of how it’s got that vision wrong up to this point is beyond me, but everyone else can see it very plainly indeed. Perhaps it needs to ask someone outside Microsoft what it should do, because those inside the Redmond campus must be drinking something mind altering. The Astoria project wouldn’t have helped Microsoft attract developers to its Universal App realm, but it would have made the mobile store seem more like that offered by the successful teams. With that now not happening, the store is exactly where it was when Microsoft cooked up this plan, but without any plan to get out of the hole it’s in. Perhaps if it stops digging, that might be a good start.