'Android on Windows': Microsoft tightens noose around neck, climbs on chair Microsoft’s annual developer conference Build kicks off later today, and rumours are swirling that Redmond has a bold plan to rescue Windows Phone from irrelevance. Aside from a modest 10 to 15 per cent share in some markets, the platform has gone nowhere, despite billions of dollars of support from Nokia and Microsoft. The …
"there was no longer any reason to make an OS2 application when Windows support was included"
Yes, but seriously nobody starts developing any new Windows only applications anyhow. It's a legacy plattform. People run Windows because they have this 20 year old software package which cost lots of money and has some obscure features some of their employees believe to need.
In 5 years will we be reading "After a couple of years, Microsoft Google moved the goalposts again. IBM Microsoft couldn’t keep up and threw in the towel" I wonder?
I don't like Google's behaviour with many things, but it is hard to feel much sympathy here.
The main issue for Microsoft and others implementing Android variants is that Google's is the popular version and developers for that platform use Play Services APIs. But, as Google went to great expense to prove, an API cannot be copyrighted. So, with a bit of lag which we expect of Android OEMs anyway, Microsoft can copy all the functionality Google implements (apparently the cloud infrastructure supporting the APIs is the hard bit, but MS should be well placed to cope with that). In short, if Google moves the goalposts it can cause MS to miss the goal, but only by making Play Services users miss at the same time. And they won't do that.
While G did argue that APIs could not be copyrighted they actually failed to establish that (pending however many appeals it goes through). They only managed a narrow ruling that the Java code they were using wasn't protectable.
MS cloning the services without an expensive trip to court might depend on just how much MS manages to piss off Google before trying it. The current EU shit stirring might already have done the trick.
Microsoft's best business strategy at this point is to just drop native Windows phones entirely and start manufacturing Cyanogenmod phones with Windows services. They've already established the business relationship, and Cyanogenmod already has excellent Android compatibility and a solid fan base. They can probably resurrect the Windows 8-like shell from the Nokia X code base for a Windows 10 complementary look and feel. And think of the free publicity and headlines - Microsoft launching an entire new product line based on Linux! Accept the inevitable - it's about profits, not pride.
"They've already established the business relationship, and Cyanogenmod already has excellent Android compatibility and a solid fan base."
Cyanogenmod has android compatibility and solid fan base yes, they're an open and likable team of people... sadly Cyanogen OS (the business version that now has VC money and got into need with Microsoft recently) is making very anti Google remarks, stabbing the first lot of people who took a chance on a cyanogen os phone (one plus), and burning a lot of bridges thanks to very childish management team.
So yes, I can imagine MS taking them (cyanogen is) over and building windows android devices, but the fan base for cyanogen mod probably won't be buying them.
It comes down to whether MS' value-add warrants using their products. This is different from using the OS as lock-in.
In other words, are Windows libraries that much better than Android ones, that devs will write to Windows APIs which are installed (possibly for free) on Android Phones?
For MS, its a question of, "we'd like to use our own APIs because our apps use our APIs." If Google keep pulling more and more stuff into Play Services, MS can expand the use of its own APIs and give devs the chance to run on Google and non-Google Android.
The multi-million dollar question is whether running MS' APIs on an Android phone will spoil battery life, and things like that. Probably, I would think, but MS have a lot of motivation to try to get it right and they may just focus on tablets, with larger batteries and write pure android apps for phones.
Thing is, compared to Winphone, Android is an awful OS. It is unintuitive, freezes all the time, makes crude use of phone real estate, pushes ads at you (then freezes again), chews though battery with no real explanation as to why etc etc. The only advantages it has over WindowsPhone is a) it was first to market and b) it has more users, hence apps. I can see why MS think they can build a better phone os....
> The only advantages it has over WindowsPhone is a) it was first to market
Microsoft has had phones since 2000 or so. If 'first to market' is an advantage the MS have screwed up big time.
> I can see why MS think they can build a better phone os....
Maybe they can but they haven't so far.
"Thing is, compared to Winphone, Android is an awful OS. "
Thing is, I've had all 3, and not the low end underpowered ones, I've had flagship Nokia, flagship android, and iOS. Android seems quite buggy on Samsung phones my wife uses, I avoid them for that reason, nexus, moto and HTC seem to have stable lines of android phones. WP8.1 was buggy as hell and forced me to get another android after 6 months... That was on a Nokia 1020.
IOS I haven't used for nearly 5 years.. recently had to help a friend with his iPhone 6 because he couldn't use data outside of WiFi signal of his house.. Checked APN settings, all fine... Guess what, reboot needed... Worked fine for a day after that until another reboot was needed. Oddly iOS has barely changed in 5 years, guess that's half the appeal for iPhone users... Unified simple ui.
But... They're all buggy... Android in my experience is the least so when looking at high end phones.. I just don't get how people can brand android as a buggy mess compared to win phone though, that's madness.
My experience has been pretty much the reverse of yours. Driving a Galaxy Note 4 at the moment. Hangs all the time, needs rebooting every couple of days, drops calls whenever you try to answer them. Browser is a jerky, stuttering/freezing abomination. And it gets pretty annoying having to constantly leave one application to get something done in anther. Lollipop update just failed to install. The list goes in. Win phone on the other hand was buttery smooth, getting stuff done across multiple apps was seemless, the browser was lovely and I had to reboot it once every couple of months
I did say Samsung were the buggier android phones... has more to do with Samsung / touchwiz than android though.
I don't think I've ever heard anybody who's used IE on winphone for more than a minute describe it as a "lovely" experience... if anything that's one of the major ball-aches of the using WP. Constantly caching then reloading webpages for no other reason than scrolling down a page... not to mention the pain of websites rarely rendering mobile versions of their sites for the thing.
Each OS has it's own advantages... problem is if you pick the wrong handset it can sour that experience considerably. My Nokia 1020 was great for about a month... cue buggy app updates from MS, then freezing apps, apps not opening, and a hardware issue related to the pureview unit (botched FW upgrade as part of an system update)... that mean't all photos were either under exposed, or overexposed (completely white, or completely dark)... which basically rendered the phone useless to me as that was the main selling point to draw me to WP.
Full disclosure though - I've had crap androids too... My Xperia S had update after update, which seemingly were only released to introduce more bugs to the phone. But that wasn't an android problem, that was a Sony problem... one I won't repeat.
What?
What phones have you used? Mate, I have had Samsung's, Sony's, and LG's all of them running stock, nightlies, and the version that came with the phone, never had the phone freeze or crash more than on a handful of occasions and always my own fault for flashing the wrong stuff to the phone.
As per Android being unintuitive it can be improved that is for sure, but I have never encountered anyone that could not get familiar with it after a short while.
The same can not be said about Winphone can it: http://forums.theregister.co.uk/forum/1/2015/04/20/windows_design_reddit_ama_jon_bell/
This post has been deleted by its author
I actually liked Windows Phone, but Microsoft has done everything possible to torpedo the OS. They've spent millions integrating the Windows Phone and Windows codebases with little benefit, all the while not adding anything useful to Windows Phone and actually making the interface worse.
They also lost all of their biggest supporters when they dropped support of ALL Windows Phone 7 devices, some of which were only a few months old, when they launched Windows Phone 8 and the whole thing has been down hill ever since. Now, in following BlackBerry into the realm of giving up entirely they've essentially admitted that they think that their phones are totally doomed. What happened to hyping universal apps? Might as well just give up at this point.
What this will solve is the "casual" games gap. Apps gap - not so much.
While the emulation layers are generally good enough to run "Where is my Water N" for various values of N, they start falling over once GPS, Bluetooth, etc come to play. So, for example, if Torque is your "must have" "obscure" app, this move will not make you switch to windows phone any time soon.
Similarly, where running "Where is my Water" under emulation may work, Shadowgun or the Sonic equiivalent of Minion Rush will probably have issues with graphics performance. The gap to QNX is much smaller than to Windows by the way. The graphics APIs there are very different.
I agree - that is likely to kill the patient
"The gap to QNX is much smaller than to Windows by the way. The graphics APIs there are very different."
That gap is certainly a yawning chasm for Microsoft. At least both QNX and Android / Linux are basically POSIX, so there's going to be a lot of similarities at the system call level.
I've tried a couple of Android 3D games on BlackBerry, seem to work quite well. I'm not really into games, so I haven't bothered taking it any further than that initial level of sating curiousity.
I'm not sure that in BlackBerry's case it's as strong a suicide pill as all that. They have a very strong business offering, and the Androidness is a nice (and fairly successful) thing on top. And actually there's a lot of good native apps for BlackBerry these days too. Whereas Microsoft have f**cked up the business side of WinPhone, so there's even less reason to have one than a BlackBerry.
On the other hand, if either of them could emulate Google Play Services (a large undertaking one imagines) they'd have the complete Android ecosystem in a Google-free environment. There's quite a lot of people who'd find that attractive all on its own...
I think it would be relatively straightforward to produce a Dalvik / ART runtime for Windows which could cope with pure APKs - i.e. those which were just compiled byte code.
But most games will probably have at least some binary backend code for arm / intel which in turn are compiled against a modified C runtime called BIONIC and a Linux kernel. So any runtime for Windows would have to somehow host those calls in a sane way.
On top of that games will be hitting OpenGL ES 2/3 so that has to be supported too. And lots of software apps rely on services from Google which would also have to work. And other background services and activities might have to be supported too for apps which have things running in the background.
So even Microsoft provided some support, it is likely to be half assed. Of course they might just throw money at Amazon and let Amazon figure out the problem. Amazon and Microsoft might both see it as a way of disrupting Google to support Android and Amazon's store on Windows.
Windows Mobile v6 - Ok, you'll need to rewrite all your apps for Windows Phone 7
Windows Phone 7 - Ok, you'll need to rewrite all your apps for Windows Phone 8
Windows Phone 8 - Ok, you'll need to rewrite all your apps for Windows Phone 10
And they wonder why no developers want to get around the platform?
"Windows Phone 7 - Ok, you'll need to rewrite all your apps for Windows Phone 8"
At the risk of someone branding me a shill again, the games I built in XNA for Windows Phone 7 ran perfectly happily for WP8 with no modification. Just speaking from my experience.
Of course, with minimal modification, they also ran on Android and iOS, thanks to Xamarin and Monogame, which feels a bit ironic: if this does happen, I'll still be writing in C# for Windows Phone, but using the Android build to deploy...
I agree. However, Google aren't much better.
I recently had cause to install an SMS blocking app on my wifes phone. I duly scoured the Play store (which is painful enough) and located several "SMS blockers". However (get this) none of them actually block SMSs. One or two actually tell you this, but most don't (read the reviews).
Turns out Google reconfigured Android between Jellybean and Kitkat and dropped the ability to block SMSs (something my Windows Phone does natively). I have a memory from trying to configure a previous device for bluetooth TTS/STT that similar tricks have been pulled in the past.
And the quality (of lack thereof) of "Apps*" in the Play store ....
*I say "Apps" because in a recent adventure to recover some deleted SMSs, I discovered quite a few "Apps" are just manuals you download telling you how to recover SMSs (which is a lost cause on an unrooted phone, and risky on a rooted one).
"Google has no monopoly on phone OS."
That's true, as Amazon Fire and Cyanogenmod and a long list of unlicensed devices from (mostly) Asian manufacturers, as well as Blackberry and Microsoft's own compatibility layers, demonstrate. And if you purchase one of these devices, you can probably load Google Play on them (manually) and enjoy access to the Play store - though compatibility of apps isn't guaranteed, of course.
The complaint isn't focused solely or even primarily on Android, though, although Android running on 17 of 20 smartphones sold in the world does give Google an unhealthy influence over a critical market.
Rather, companies such as Yelp, Expedia, Foundem, Euro-Cities, ICOMP, Hot-Map, and 1plusV have asserted that Google unfairly directs searchers to their own services rather than more popular competitors on both mobile and desktops, in essence giving a special boost in rankings to properties in which they have financial interest. Google is also famously aggressive at collecting and analyzing personal data to provide exactly what each person wants, not always with the most fully transparent approach.
I've been a huge fan of Android as they battled Windows, iOS, Symbian, Blackberry, and the like, and Google's other services have made my life far more convenient. But their amazing success has made me a bit nervous. Am I certain that the results of my searches are representative of web content? Is their tailoring of search results biasing the information I consume and thus molding my opinions in their preferred image?
I'm ambivalent regarding this litigation in part from these concerns.
C#'s origins were "Visual J++ with enough changed so Sun can't sue", and despite C# moving on dramatically since then, the core of the language remains very similar to Java. This makes Android source-code feasible to cross-compile into C#, although it's going to be a non-trivial task.
In short, if MS offers any kind of "Android support" on Windows (there is no Phone/Desktop split anymore after 10 - it's the same API), I suspect we may be looking at an automated porting toolset, rather than an on-device emulator...
It doesn't work like that. As the article said, hosting some Android apps successfully means handling Linux system calls made by native libraries.
Intel has the same problem. Sure, there's Android for x86, but it's nigh on useless because all the Android apps that come with native libraries attached are supplying ARM native libraries, not x86 native libraries. For these to work on x86 Intel have to emulate the ARM instruction set! Not fast, not very energy efficient...
I'm talking about source code, not executables. Hosting non native executable binaries is a pain, as both you and the article quite rightly point out... And simply hosting the Android binary means you're never going to do better than Android's non-stellar runtime performance.
Porting their source will also allow Android devs to target the much more significant Windows x86 desktop market, as both Windows 10 Phone and Desktop use the same API. (As does 8.1, but the ui controls don't adapt as well between small and big screen, and 8.1 "metro" apps can't run in a window on 8.1 desktop, but can on 10).
"Intel has the same problem. Sure, there's Android for x86, but it's nigh on useless because all the Android apps that come with native libraries attached are supplying ARM native libraries, not x86 native libraries."
I think that's less of an issue these days. Most of the well known backend libs have Intel support and I expect most of the major titles would be motivated to support Intel.
Of course Google could have made this a non issue if the native SDK only targeted to LLVM bitcode. So you compile a .so and it's bitcode. Then you upload your APK to Play and it could have produced ARM, mips, Intel versions automatically and served them out according to the device requesting the app. Or it could have been converted to native on the phone when the person downloaded and installed it.
Slight problem there. Apart from Google Play Services and Google's own proprietary blobs, most of Android is open source. It's not currently within Google's gift to smack down anyone.
Whereas iOS is entirely closed source. Much harder to pull off the same trick. It would be possible to reimplement the iOS APIs using publically available documentation but then you'd have to persuade all the app developers to recompile their code and risk the Wrath of Cook.
Is the iOS app developer's API documentation public? I dunno, but I wouldn't be surprised if access to iOS API documentation came with strings attached...
It's all the same answer set isn't it?
One or more of the following:
Panic,
shuffle CEOs,
layoff workers,
buy a small company for a ridiculously large sum of cash,
sell small company they've already bought for ridiculously small sum of money or shut it down completely,
Pass out large bonuses all around at top levels...
lather, rinse, repeat.
I'm sure there's an executive dartboard for sale somewhere that has these options as targets.
"When the wrap-around OS/2 booted and found DOS and Windows were installed, it virtualised and hosted them. Sometimes the Windows apps ran even better inside their new host – but you could never be sure."
Windows 3 Recompile
When IBM licensed the Windows 3.0 / 3.1 source code, apparently the first thing they did was recompile it just to see if they'd got everything right. When they benchmarked it they were surprised to see that their compiled version ran a lot quicker than Microsoft's retail version.
The reason why? IBM had used the Watcom C compiler (which was pretty good), whereas MS had of course used their own C compiler (and it was pants in comparison)...
32bit Windows and OS/2
What really killed off Windows under OS/2 was that IBM couldn't accomodate 32bit Windows drivers and applications (which were just beginning to creep in at the time). OS/2 was architected around 512MB virtual memory per process, whereas Microsoft had gone for 2GB virtual memory per process. That made it impossible to emulate a Windows environment inside an OS/2 process.
I actually used OS/2 a lot back in the day, and I didn't move off it until Windows 2000 came along. Typical IBM - had a good idea, made it just good enough to be able to use as a tool to access IBM mainframes, otherwise they ignored it. When it was shiny new it was (for the time) fantastic, but they never bothered doing anything to it.
It'll be interesting to see just how "dead" BlackBerry become. Comparing Android on BlackBerry to Windows on OS/2 is worthwhile, but there are some critical differences.
1) IBM basically didn't bother improving OS/2 in any significant way after 2.0. There were major architecural problems with it that they didn't do anything to resolve, and they didn't seek to improve the user experience in any way. It would have died anyway. Crucially, IBM as a whole didn't believe in OS/2; real IBMers did servers and mainframes and big database applications. Meanwhile BlackBerry are actively improving BB10 / QNX, and because it's central to their entire business offering (BES12+BB10+everything else) it gets a lot of attention from the whole company.
2) OS/2 didn't have any unique features at all compared to everything else, and had some real day-to-day disadvantages (e.g. driver support, and it was resource hungry). There were very few reasons to have OS/2 at all. In contrast, there's much functionality in BB10 that all the other mobile platforms just don't have (e.g. Hub). There's many reasons (admittedly quite a lot of them are work related) to have a BlackBerry, and Androidness is just an extra nice thing on top.
"OS/2 didn't have any unique features at all compared to everything else"
True multi-tasking? Large memory model? Object-centric UI paradigm? App persistence across reboots? Did you ever *use* OS/2 2.0 and Windows 3.0 on the same machine?
I was never much of an OS/2 fan, because I didn't have IBM hardware and it had stability issues on my clone - and kudos to IBM tech support for investing a full hour trying to debug those issues. But I spent that time trying to get it to work specifically because it had a lot of advantages to offer!
True multi-tasking? Large memory model? Object-centric UI paradigm? App persistence across reboots? Did you ever *use* OS/2 2.0 and Windows 3.0 on the same machine?
Yes, I most certainly did use it, coded for it, embedded it (which it was quite good for), used it as my desktop too, for the very same reasons you've stated.
You have to recall though that at the time no one generally knew or cared about the difference between MS's cooperative multitasking and IBM's true multitasking. They did care about the price; OS/2 was expensive. And whilst it had a 512MB virtual memory model, larger than Windows 3.0, Microsoft chose 2GB for 32bit Windows. That meant OS/2 was forever stuck emulating Windows 3.0, and couldn't accommodate any of the 32 bit Windows apps that were beginning to emerge. It very quickly stopped being "a better Windows than Windows". And with no killer application to offer (Lotus Improv was as close as they got) its irrelevance was assured. Technology-wise it was a mixture of good and bad; multitasking was good, UI was good (though if you look at it these days it feels terrible; no anti aliasing!), thunking for 16 bit code that seems to make up large chunks of the OS was bad, as was the lack of driver support, and it never quite shook off the consequences of being engineered for the 286 in its early days.
both the cheapest and the best smartphone I've had. Microsoft and Nokia outdid everybody with WinPhone 8.1. 10% share of a very large pie is a lot of phones. And developer support is on the up and up.
Making it support Android out of the box does seem to send a counter-productive message. I miss two things from Android: my savings bank account app and my credit card account app - and I keep my Android phone going without a SIM just to run those apps. My current account has a Winphone app.
Here are the IDC figures for 2014 global phone shipments - 10% looks pretty optimistic:
http://www.idc.com/getdoc.jsp?containerId=prUS25282214
OS Shipments (millions) Revenue (millions)
Android 1,060 (82.3%) 255,102 (60.9%)
iOS 178 (13.8%) 116,540 (33.8%)
Windows phone 35 ( 2.7%) 7,782 ( 4.2%)
Other 14 ( 1.1%) 3,480 ( 1.1%)
Is it possible for MS to build a developer following here? If supporting android apps on Windows means it is easier to develop those apps on Windows, it might create a population of devs who develop on Wandroid first and then test on chocolate-flavour as an afterthought. The stage is then set for MS to create an "extended subset" of the Android API.
Microsoft are going to make their phones compatible with all apps.. so I think they might be ok.
I despise the Windows phone interface, but only a complete idiot would think making a phone compatible with all apps from all operating systems is a bad thing.
Android apps will run without the user having to do a thing. Download the app, run it. That's how it will work on a Windows 10 phone.
Supposedly iOS apps with have to be recompiled in Visual Studio - you load your project into VS then compile it as a Windows phone app. That's it, no changes, no modifications. We'll wait to see if that works (I predict issues).
Most people producing iOS apps aren't doing so on PCs and therefore don't have Visual Studio. Problem? Not really, Microsoft are about to distribute Visual Studio for Mac OS and Linux at a cost of zero dollars and zero cents. Because they want iOS app developers to put their stuff on Windows phones. Don't hold your breath Microsoft, Apple users tend to have a slight bias against your software.
So potentially, if you can get past the awful UI, Microsoft phones will be the most flexible phones on the market.
Here's a fun fact. Connect a Bluetooth mouse and keyboard to a Windows 10 phone and it will ask you if you want to run Windows in desktop mode. Not sure why you'd want to run 6" Windows PC.. but you could.
Sorry, they just made their announcement at Build, and Windows Phone 10 can NOT run Android (or iOS) apps "without the user having to do a thing". Rather, the source must be loaded into a development tool, and then ported to the Windows 10 environment (which has been extended to provide more, but not full, commonality with Android and iOS), and then extended if desired to make it look more like a native Windows application.
The good news is that a developer ends up with an actual Windows app (sort of) that will run on desktops and Xbox as well (though not necessarily *well*, depending on the app itself). The bad news is that it will take some non-trivial work per app to get there.
Also, they are unfortunately NOT providing Visual Studio for OS X and Linux, but Visual Studio *Code* - which is essentially just a text editor with Intellisense(tm). Most of the goodness isn't there (yet). But at least it's free. I may download it and give it a whirl on Ubuntu tonight, just to see how it compares to my favorite editors.
Not as much as we'd hoped, but a step in the right direction - if they choose to continue with Windows Phone long-term. However, they *desperately* need to sell a significant number of phones if they want to stay in the game, and "we can run some of the same apps as Android and iOS!" is a pretty weak marketing pitch, I'm afraid. :-/
MS was second-rate from the start, but they gave a generation of unqualified and clueless IT managers a creed: nobody gets fired for buying Microsoft.
To their credit, though, they did turn it into a great gaming platform (eventually, I'm surely not the only one to recall how lame their early efforts looked in comparison with, say, games on a Commodore 64).
Good dev. tools, I would always want a *nix or *nux on the desktop, but would not do without Windoze too. Too many programs I like that won't run on anything else.
As for phone OS wars, really, who cares? I do like many Android programs, don't mind paying, but am disturbed by the chunk that goes to the Google monster. At least they are not as greedy as Apple (30%, FFS).
I would not even be writing this if I had not broken the USB port of my iTron phone, which I still prefer in many ways, and keep charged for the alarm, camera, and old downloads.
Would be writing something slightly, but not much, different.
They're constantly on sale for $50 or $100 Canadian dollarettes, UNLOCKED, including free shipping and a generous free-shipping return policy if you don't like it.
Geesh, buy one as a spare phone. Or as a travel phone. Or just to try it out.
They're worth that just as a 4-inch tablet on wifi.
As Yogi Berra said, "It's deja vu all over again". MS is acting like a company at a crossroads and will not make a decision until too late. They have basically lost the war in the much of the consumer space will doing very well in the business space still. But instead of conceding the consumer space to Apple and Google they insist on spending money on it. This money could be spent on the business side, probably to much better effect. They are acting much like IBM did in the late 80's and early 90's. IBM is now a much less important player in most markets, not the dominate player they were earlier.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
