Ballmer: Microsoft Mobile Should Focus On Android Apps Not Universal Apps (theverge.com) 121
UnknowingFool writes: Former CEO Steve Ballmer had some strong opinions about the direction of Microsoft's mobile strategy. As reported last month, Microsoft's Project Astoria has not been received well and is not going well. The strategy is to help build Windows 10 apps by making universal apps via easy porting from Android. Ballmer questions its effectiveness. "That won't work," he said. Instead he suggested that Windows phones should "run Android apps." This is a dramatic departure from the Microsoft-only focus that Ballmer championed during his tenure as CEO.
I agree (Score:5, Interesting)
Re: (Score:2)
But there are technical problems with this right?
What is sounds like Ballmer wants would be programs for Linux to "just work" on Windows. I hope someone can help me out here.
Re: I agree (Score:2, Informative)
Android apps are really Java apps with lots of gui customizations. It's all the gui and OS specific calls that make translation difficult.
Re: (Score:2)
Re: (Score:3)
Do what others have done/started to do/tried to do and build an API/ABI compatible implementation of the "Google Play Services" module. Map the Maps APIs to Bing Maps. Map the Drive APIs to OneDrive.
Some of the Google Play Services dont fit to Microsoft services (e.g. Wallet or Google+) and some cant be made to work cross-device on both Android and Windows (e.g. Drive and Game Services would need different back-ends on Android and Windows and you wouldn't be able to use data stored with an app on one platfo
Re: (Score:2)
Of course the downside to this is that it would show up Microsoft as hypocritical for copying Google APIs whilst at the same time supporting Oracle in its fight against Google for copying Oracle APIs.
If Google is going to undermine Oracle's APIs, Microsoft might as well retaliate and undermine Google. That's not really hypocritical. That's like saying "I want us to go to the circus. But if we decide that we are going to the movies, I'm coming too since I'm not just going to sit at home."
Re: (Score:3)
Clearly you dont understand the meaning of hypocritical.
Arguing in court that something should be illegal while at the same time doing that very thing most definitely IS hypocritical.
Re: (Score:2)
There is an existing compatibility framework for android that can run on other things. Jolla and others have licenced it.
Re: (Score:3, Interesting)
Comment removed (Score:5, Interesting)
Re: I agree (Score:5, Insightful)
and they'll chew CPU because system calls won't be native
I disagree. WINE does basically the same thing: it translates Win32 system and library calls to Linux equivalents, and programs running under WINE, when they work correctly (meaning all the calls they do are fully implemented and done correctly) reportedly run at full speed, sometimes even faster than on Windows itself. Remember, you're not talking about emulation here, you're talking about ABI translation. And since Android apps are (usually) Java-based, they have to be decoded from bytecode anyway, no matter the platform.
because they'll use extra RAM hosting Windows plus an emulated Android environment
Yes, that could be a problem; it's not emulated as I said before, but you do need a translation layer there (since WinPhone apps don't use Java AFAIK) which will take up more RAM. WinPhones I think are usually lacking RAM too, compared to Android phones, so this could be a problem.
and probably the GUI will be slightly off
This will probably be a big problem; they just won't integrate well with WinPhone. People have complained about PC-based Java applications this way for ages, that they look out-of-place no matter what platform they're running on. The same happens with WINE applications: they don't normally look or work like the regular Linux applications. The same will probably happen with any Android apps, unless MS does a really good job figuring out how to mitigate this.
That will make Windows phones appear inferior. Windows phones would have to offer something else that users really wanted (like AD integration) to make people put up with poorer app performance.
Well, they are inferior. Even if you're of the opinion that the OS is better (which I disagree with: I think it's butt-ugly and I absolutely hate the whole live tile thing, though I'll admit that their devices do seem speedier than Android phones which seem to need a lot of horsepower to have a responsive UI, but I'd rather have a laggy UI than suffer with the ugly abomination that is Microsoft's latest UI), no computing platform is of much use unless it has the applications you want or need. No one should know this better than Microsoft themselves: it's the whole reason Windows has been so dominant for so long. It's not because of their crappy OS, it's the availability of applications which keeps people tied to the platform. Well this whole dynamic is what's biting them in the ass now: they missed out on being early enough to the party (unlike with Win95) and building dominance, so other platforms have become the popular ones, and they're left out. If people can't easily install and run their favorite popular apps, the platform is simply a non-starter. Everyone has their own favorite apps; for me, if a phone won't run Tinder, for instance, there's no way I'm going to waste my time with it.
Re: (Score:2)
Regular updates is a good feature, it's true, however you're seriously overestimating the wisdom of the masses. People just aren't that smart about things like that; they'll go for the cheap phone which seems to work well enough, and when the lack of updates shows itself to be a problem, they'll just get a new phone. You can get Android phones for $50 or less these days, and you can even get very nice models a year or two old on the used market for not much money. People don't see these things as an inve
Re: (Score:2)
Re: I agree (Score:5, Insightful)
Android apps are really Java apps
True for most apps, not always true for some of the fanciest games. I actually don't think putting together a compatibility layer that would enable the vast vast majority of Android apps to work would be a terrible technical problem.
No matter what though there will always be the occasional app that does not work and it will probably be a constant source of frustration for some users.
Android also has a lot of well Androidisms, around how notifications are delivered to the user, switching between 'activities' is handled what apps get broadcast messages and so fourth. Unless Microsoft mostly clones the Android UI the workflow will be different inside a lot of apps. Android apps don't operate in a vacuum quite the way typical desktop applications do. For example you encounter a certain media type your android app might send a message to something else to handle it, what that something else is, it does not know and when you are done the Android operating systems UI is depended upon to provide the 'back button' to get the previous activity, or to return you there otherwise.
It would be like running Windows apps in Wine on Linux or OSX, where there is always some oddities around file pickers, windows controls, associations, cut paste interop, etc. It mostly works, and you can figure out what you need to do but it does not always feel right. The only difference is it will be even more pronounce on a phone.
Nobody switches to Linux so they can run Windows apps on WINE. WINE lets them keep that one app they need so they can switch platforms for other reasons. I don't think you attract anyone to your platform by being the 'next best place to run their favorite apps'. So on that score Ballmer is wrong. Microsoft's entire mobile strategy is wrong. They need to admit they made some bad calls on their first gen smart device platform (Windows CE + Windows Phone -lt 6 ) and were simply to late to the party in the 'app' era. If they do anything they should focus on producing Apps for the other platforms. Give business users what they really want, full Exchange/Outlook support with all the manageability, some sort of Sharepoint application viewer, etc. That is where their bread is buttered. Microsoft should focus on being the top shelf app vendor, for their mobile play IMHO.
Re: (Score:2)
Android apps are really Java apps with lots of gui customizations. It's all the gui and OS specific calls that make translation difficult.
Not all Android apps are Java apps. Many are written in a compiled language (usually C++), although they are a minority because they are not platform agnostic, so tend to be ones that require high performance or determinacy.
Re: (Score:3)
Yes. Blackberry 10 can run Android apps. But it doesn't have the Googe Services (Google doesn't allow that). So apps that rely on that have problems or don't work.
I guess that's why Blackberry's latest phone is an actual Android phone.
Re: (Score:2)
I have had a Blackberry for the last couple of years. Yes, Android applications work when they work. But if you want to use an app that depends on Google Services, you have to jump through a million hoops. Sometimes apps just crash. And since it's an emulated environment, the average blackberry hardware is just too weak to maintain acceptable performance.
So yes, it's "just Java", but there's more than that. As much as I hate Android, I had to switch, it just wasn't worth the effort.
Re: (Score:2)
The Start of the End of BB10 was the realization (of the Company) that their App store was never going to compete with Google's, or Apple's app stores.
I really liked my Q30. But in the end, I moved back to Android for a couple of Apps that I couldn't get to run. Now that the Priv is out, I'm really interested in returning to the BB fold.
If MS adopts android apps, it will be a short trip to giving up on Windows Mobile. Microsoft's experiment in trying to use their desktop dominance to muscle in on the mob
My experience w/ Lumia (Score:2)
I had a Lumia 520. Loved it - it had almost all the things I needed. Maps, calculator, email client, messaging, OneNote, et al (OneNote really made it shine). Sure, it didn't have as fantastic games as the iPhone, but then again, it was very useful as a work phone. Any employer who wants to issue office phones to employees and not have them jack it w/ bizarre apps from the stores would do well to consider Windows Phones just b'cos they don't have all that, but do have the things office users generally n
Re: (Score:3)
No, they need to just throw in the towel. They were too late and missed the boat, and nothing's going to change that. The other two ecosystems have all the developer interest and consumer mindshare already, so it's like Linux vs. Windows on the desktop: no one wants to bother changing unless the incumbent gets SO bad that they're basically forced to. "Free stuff"? What are they going to do, give away the phones for free? Android is already dirt cheap and apps are largely free, so there really isn't any
Re: (Score:2)
too late Steveo (Score:4, Insightful)
insensitive (Score:2)
Some of us were hit by the shot of the short chair!
Re:too late Steveo (Score:5, Funny)
I think that your words will be ignored. Sorry Pal, you had a shot at the big chair and threw it. Well, some of us never were given the chance. Enjoy your cash.
Fixed that for you
Re: (Score:1)
We'll keep you posted on this developing developing developing developing story.
Re: (Score:1)
Appers! Appers! Appers!
Ballmers (Score:4, Insightful)
Re: (Score:2)
Exactly. I'm not even a Millenial, but actually *talking* on the phone is one of my lesser-used functions for the device. I spend more time using it for: 1) texting, 2) Tinder, 3) mapping (both Google Maps and OsmAnd), 4) taking photos, 5) internet browsing, 6) games, 7) calculator (RealCalc in RPN mode), 8) random apps (like a Gauss meter, compass, etc.), and 9) visual voicemail (I'm never going back to the shitty old voicemail where I have to press numbers to navigate and listen to audio prompts) than I
Microsoft-only vs. running Android apps (Score:3)
When Ballmer was at MS, he championed Microsoft-only because that was the way to keep people locked into Microsoft. MS had a dominant position already, and keeping things MS-only made sure any competition didn't have enough applications to be attractive. But now MS is going into a market where they don't have a dominant position, so MS-only just insures MS won't have enough to be attractive. The only option will be what MS did with IE back when Netscape ruled the Web: offer compatibility to lure users and especially businesses over, then slowly break compatibility to force a Hobson's choice.
Re: (Score:2, Informative)
The only option will be what MS did with IE back when Netscape ruled the Web: offer compatibility to lure users and especially businesses over, then slowly break compatibility to force a Hobson's choice.
You have no clue. They lured users to IE by having the most functionality. First plugin support (activeX). First AJAX support. First CSS support. People think that Microsoft somehow tricked users to use IE. It's not true. IE was at the time (~2000) the best browser hands down - with superior performance and memory usage.
Re:Microsoft-only vs. running Android apps (Score:5, Insightful)
That's not quite accurate.
ActiveX wasn't the first plugin support, for starters. NPAPI beat them by a year, and Java applets were capable of most of what people wanted ActiveX for.
IE did have the first CSS support, but it didn't mean much. CSS was more or less unknown until the IE4/NS4 era. Yeah, IE3 had some support for it, but CSS wasn't taken seriously back then.
I notice you don't mention IE had the first implementation of the marquee tag. Why the omission?
And if we're talking firsts, I don't think IE is going to beat Netscape here. Netscape's influence on the early web far outweighed IE's. I'll just drop the word "Javascript" here, as a conversation starter.
Microsoft did lure businesses and customers over by being installed by default on desktops. Microsoft did break compatibility with web standards - that CSS you mention was a half-assed implementation at best. Remember how long it took them to fix their messed up box model calculations? Microsoft did target businesses with ActiveX (because only the insane would allow its use outside the intranet), a feature that no other browser wanted to support even if they could, for the horrible security problems. And I'm assuming you never had to deal with ActiveX applications that only worked on IE6, even years after IE6 was out of date, because of incompatibility problems.
Superior performance and memory usage? Have you met users? They don't care about that. They use what's on the desktop. They click that icon that says "Internet" placed oh so handily on the left side of the screen under the recycle bin, or in large print at the top of the start menu. That's what the whole antitrust lawsuit was all about.
The saving grace of it all is that Microsoft is just so very bad at all things internet-related. If they had kept working on IE instead of sitting on their laurels, Mozilla would never have been able to make a comeback, and Microsoft would be dictating the standards.
Re: (Score:2)
They're easy enough to make, with CSS animation or Javascript.
Re: (Score:2)
CSS had nothing to do with IE's adoption. Users didn't (and still don't) know what CSS was. That's like saying Windows XP was adopted by home users because of NTFS.
Hardly anyone was using CSS back then, anyway. I was experimenting with it with NS4 and IE, and while it was nice, it wasn't mature enough to rely on - in either browser.
Besides, IE's CSS implementation was incorrect, and remained that way for ages. I don't know about Edge, but given Microsoft's track record, I'm not holding my breath.
What's
Re: (Score:2)
And job security, let's not forget that, IE meant job security for whole generations of IT security professionals!
Re: (Score:2)
> IE was at the time (~2000) the best browser hands down - with superior performance and memory usage.
LOL, revisionist history much?!
IE was not superior; ActiveX was a security clusterfuck. I was using Netscape back in 2000; While IE had better memory usage because iexplore linked to a ton of system .dll's, Netscape had a far better bookmark system. IE used the stupid file name schema of Windows: You couldn't use / in your bookmarks. In Netscape you could manually control the sort order along with usi
Re: (Score:1)
maybe google should buy up companies trying to make Android apps work on Windows and shut them down?
maybe google should start changing the API's when a working way to get Android apps running on Windows is released on devices?
maybe google should require app developers to throw up a screen warning users when their apps are run on emulators run on Windows saying their application may crash and could be insecure when running on non-google OS's.
shove that up your a55 Steve
Re: (Score:2)
But if Android compatibility isn't done well, then Microsoft could actually hurt what share they do have. Blackberry's Android support was just good enough to be bad, and BB's Android-compatible devices just gained a reputation for shitty native app support and limited Android app support.
I think Microsoft is in the same position Blackberry was three years ago, beginning to realize that it's native apps just aren't large enough in number to make the platform desirable to a wide customer base, but still dith
Re: (Score:3)
Right, Project Islandwood (Microsoft's implementation of the iOS frameworks and Obj-C compiler) is looking promising. Microsoft doesn't really need to get both Android and iOS apps on Windows... it just needs one or the other to provide the easy route for most developers. I tend to see more iOS-only apps than Android-only. iOS apps are generally more polished. And Windows apps are conceptually more akin to iOS than to Android.
Re: (Score:2)
iOS apps are written for a very tiny number of devices. Aren't they even written for specific screen resolutions? They're nothing like Android apps, where they need to be able to run on all kinds of different devices and screen sizes.
And finally, how exactly do you propose for WinPhone devices to connect to the Apple app store to download these apps? Apple sure as hell isn't going to allow that; they're more of a monopolist than Google is by far.
Re: (Score:2)
That's not what Islandwood does. It is not a binary compatibility runtime and it does not allow you to run apps directly from Apple's app store.
Islandwood lets the developer take his iOS app source code and build a 100% native Windows app from it. All you need to do is copy your Xcode project to your PC, run a simple command-line tool that builds a Visual Studio project around it, and then open that project in Visual Studio where it can be debugged, run on the PC, run in a simulator, or run on a Windows p
Re: (Score:2)
Ok, that doesn't help much when developers don't bother going through this process to make WP apps. Just look at the number of iOS apps versus WP apps.
Re: (Score:2)
It helps developers reach a larger market that includes both Windows phones (which is currently a tiny market) and Windows desktop (which is a much larger market) and in the future even Xbox. If they make money on iOS and want to make more money, they'd do well to at least take a few minutes to try this out.
But let's call this what it is: an attempt to make the Windows ecosystem more relevant in the post-PC era (particularly phones). The more decent apps are available, the more legitimate the Windows sto
Re: (Score:2)
I'm sure IBM thought this strategy would work just great with OS/2 as well.
Re: (Score:2)
You don't seem to be paying attention, or you don't understand how OS/2 and Windows application compatibility worked at all.
Re: (Score:2)
I understand perfectly well. I've seen enough history to know that "me too!!" is rarely a successful business strategy.
Re:Dear Andy Rubin (Score:4, Insightful)
Considering the level of spying going on in Windows 10, I don't think Redmond has that credibility.
Re: (Score:2)
I see an opportunity to seize the Android device market here, a much needed Google free, privacy focussed handset.
You can have Google-free Android right now for a spyless experience. It's not even that hard.
Steve Ballmer's ideas are quite valuable. (Score:5, Funny)
Re: (Score:3)
Re: (Score:2, Insightful)
Except he isn't. There's only three reasons to support another platform's apps in that way.
One, the apps are the de facto standard and that support is enough to keep people on your platform (think WINE and OS/2 and see how well those platforms are doing).
Two, the apps are part of your old legacy platform and it's a transitional step (DOS support, Apple's repeatedly migrational steps as it switched CPUs, etc). This is by far the most successful approach because it does nothing to diminish your position and
Re: (Score:2)
Yup. Just as happened with OS/2, once people realize that they can support both platforms by writing the "common" software, but only your platform by writing "your" software, they'll never write another line of "your" software again.
The only way to prevent that is to have some APIs that are only accessible through "your" software. Of course, convincing everyone to write "common" software and then making it suck just means that you've become a terrible Android device.
Not seeing a good path to victory here.
Re: (Score:1)
Maybe the Steve Ballmer idea was the normal Microsoft one:
* Embrace
(* Extend)
* Extinguish
If they don't feel competitive now and think that having the Android apps on their platform would make them that and even be able to take over the control from Google then that could work.
You may not see that as viable and it wasn't for OS/2.
On the other hand I guess JAVA and IE kinda may have showed another story.
Since I was an Amiga fan I've always wanted multi-platform applications because my platform was dying witho
Opinion of a guy who blew it (Score:2)
This guy blew it up at MS and the board scrambled to relieve him out of his chair ASAP. That didn't go well with him so now he is busy giving his "wisdom" retrospectively. Where were your statements when project Astoria started? Remember, MS is your screw-up.
Re:Opinion of a guy who blew it (Score:4, Interesting)
Well from a business perspective, during his tenure he tried to capitalise on MS’ dominant position on desktop and server in order to promote dominance on the emerging mobile-client platform. This he or the company he presided upon failed to do, and he retired after the “screw up”.
But that does not mean that he must continue dumbly advocating that there be strict adherence to what he tried, and failed, to do. After all, consistency is only a value if you are not a screw-up.
He continues to be MS’ largest single investor. That gives him a large vested interest in advocating that it behave rationally, even when rationality in this sub-game is at odds with the strategy he pursued previously.
Ideally MS would have succeeded in perpetuating its dominance to the new platform. It did not, so now it’s the case to do something else which is at cross-purposes with what would have formerly been ideal. This is not evidence of fallacious thinking; much the contrary.
Re: (Score:2)
Well, smart people do change their minds in the face of contrary evidence, but there is no particular reason to believe that his judgement of what is the best strategy for Microsoft is any better than his judgement in years past when he was CEO.
Womdows (Score:1)
Re: (Score:2)
You already can do that [bluestacks.com]. In fact, the app has the seventh largest Android user base [engadget.com], beating companies like Xiaomi and Sony.
I agree with Steve. (Score:2, Interesting)
I spent part of the last 3 years (about 20% of it) developing and maintaining an iOS app (hybrid web app - UIWebView + a GWT/Java server side). It predated Swift. I come from a Pascal/C/C++/Java/C# background, and I hated every day of Objective C. I recently got a taste of Swift, but... why the hell can't Apple use a language with more typical syntaxes?
We did iOS first, because among our particular customers, they had better than 80% market share.
The business case recently came up to need to port it to Wind
Re: (Score:2)
You are correct in many ways, but let's not forgot, a lot of people simply don't comprehend Windows 10 and its depth/reach.
Windows 10 already has 150+ million installs, Windows 10 is the "UWP" - Universal Windows Platform. With a single development methodology to target Phone, Xbox One, Windows 10 on Tablets and Windows 10 on desktop.
I think a lot of people are writing off "mobile" on the concepts set forth by Apple & Android without thinking "Universal" where win10 already has a sizeable market share
Re: (Score:2)
WINE isn't a panacea for the migration of all those Windows gamers to Linux, why would running android apps on a windows phone be any different?
It won't. That's the problem with that approach. OS/2 did something kinda similar and it didn't exactly work out for them either.
The problem is more mindshare. There's just no incentive to go with a Windows phone. Android integrates with windows (kinda), though I don' t think many people do much other than copy pictures, movies or music back and forth. Apple has t
Perhaps. (Score:2, Insightful)
This reminds me of the efforts of the Wine project. As most of you may know, Wine works poorly for a lot of applications. One of the reasons desktop Linux has suffered is not because it's inferior, but because a lot of people rely on various applications only available on Windows. Security issues aside, imagine if Linux users could run any and everything Windows could, seamlessly.
Unless Microsoft does it right, they will fail spectacularly. It will be perceived by the public that Windows mobile is buggier t
Re: (Score:2)
The difference here however is that Microsoft would not need to replicate every last API's from scratch, which is what Wine is trying to accomplish. Further more even if some Android library needs modifying to work with Microsoft's shim layer they have access to the full source code for that library which makes life a whole lot simpler.
So because the amount of services/API's that need replicating are much lower your chances of succeeding are much higher.
Further more in the interim you could simply cut a dea
Re: (Score:2)
Yeah, Blackberry tried all that. Supposedly Android apps actually work quite well on the latest BB phones. It isn't saving them, their company is dying.
It Wont Work! (Score:2)
"Trust me, it wont work, I know from experience!" - Ballmer
Obligatory XKCD reference: https://xkcd.com/323/ [xkcd.com]
Microsoft? Android Apps? (Score:1)
Three words: Embrace, Extend, Extinguish.
Re: (Score:2)
Embrace, Extend, Extinguish, Escape.
Re: (Score:2)
Not any more. "Extinguish" is what happens when this strategy works.
Now, it's "EEF": Embrace, Extend, Fail.
Admitting (Score:1)
This is admitting three things: 1) Microsoft can't gain users to make Windows phone dominant (and profitable); 2) Windows phone is an inadequate selling proposition; 3) An adequate selling proposition requires Android (or iOS) compatibility.
This is about Microsoft refusing to leave a growing market for portable devices. They may even have a plan to increase the appeal of Windows phone at a future time.
When shareholders cheer your exit from a company.. (Score:3)
Oh Microsoft ! (Score:2)
what problem are they trying to solve? (Score:2)
If I'm mostly going to run Android apps, why would I want to do that inside a Windows phone? All I get is extra overhead from emulation and incompatibilities. Before worrying about phones and Android, Microsoft really needs to worry about Windows store, because it looks to me like few of the traditional Windows software vendors are making their software available through the Windows store. I don't know whether that is because Microsoft charges too much money or because the store requirements are too strict,
Balmer has the vision of a blind man in the fog... (Score:2)
Android apps start fast and have a low memory footprint because they use shared memory to share code with the software that runs the phone. You can't do that unless the phone uses java for it's internal functionality, which Windows phones would not, so app's startup time and memory footprint would be much worse than on real Android.
Balmer understands markets, but he doesn't understand tech, or design. His reign at Microsoft showed a complete squandering of technical talent in a series of boondoggles that
CEOs & Group Think (Score:2)
How'd that work for Blackberry? (Score:2)
Re: (Score:2)
I think RIM could have succeeded in becoming a successful Android phone vendor, but they tried it much too late. I mean, it was mostly too late by the time the Playbook came out - and when it didn't have support at launch (which many were expecting it to), that was kind of the end of their chance.
I don't think "going with Android" would have been a bad strategy for either RIM or MS, but neither of them did it at a time when it made sense; instead, in both cases, it comes off as a desperation play that's wa
Re: (Score:2)
Re: (Score:2)
The current Nadella "universal apps" strategy is working. Microsoft Surface has gained a significant amount of both mindshare and market share. the Xbox one is also in that paradigm, and I expect sales this Christmas to be robust. The phone market is a problem, and mostly a problem with the fact they use ARM processors. Microsoft has a very confusing implementation of their "universal apps"--they aren't universal at all. They must be compiled for each hardware platform. Microsoft used to have a universal ap
Re: (Score:2)
Just to followup with a summary... If they want to compete in mobile, they need to make a Surface phone, not this Current Windows RT phone that can't really run much of anything at the moment. A "Surface" phone has a good chance of selling, since the tablets are popular.
Re: (Score:2)