Slashdot Mirror
Microsoft Details Windows 8 for ARM
MrSeb writes "In an 8,000-word treatise, Steven Sinofksy himself has taken up pen and paper to describe Windows 8 on ARM (WOA) in great detail. There's a lot of good stuff in there, but one point is particularly troubling. Quoting Sinofsky: 'WOA does not support running, emulating, or porting existing x86/64 desktop apps. Code that uses only system or OS services from WinRT can be used within an app and distributed through the Windows Store for both WOA and x86/64. Consumers obtain all [WOA] software, including device drivers, through the Windows Store and Microsoft Update or Windows Update.' It's hard to under-emphasize just how huge a change that is. It's one thing to say that ARM CPUs won't support x86 emulation; something else entirely to split software delivery and installation. Up until now, one of the biggest differences between desktop and mobile operating systems has been the ability to install software. It's true that Microsoft's decision to wall off unapproved software installation is similar to the approach of Android and iOS — but iOS isn't the same thing as OS X. Combining both of these decisions under the 'Windows' brand could be disastrous, not because Microsoft is evil, but because it creates two entirely different user experiences on the basis of which ISA your CPU supports."
Depends how good their marketing department is. Remember, they can afford to lose a few hundred million dollars a year in that department for however long it takes to turn it into something profitable, and they have a history of using their successful products as tools to drive users to their unsuccessful products.
Y'know, for all I hear about Linux being so fractured, I'd expect to see more coherence from Microsoft and Apple.
The vast majority of distros differ in small ways, but they all work with mostly the same paradigms. To install software, you usually install a package from a repository. To add something not in the repositories, it's usually "./configure; make; make install".
Looking at the Windows world, there's worse fracturing, but because it's all under one brand, it's somehow okay. Inter-process communication is done with DDE - no, wait - OLE. I mean sockets. Really .NET has its own IPC and you should use that now. On one system, you install with an executable file obtained from the vendor. On another system, you install through a storefront.
At least Linux accepts that it's fractured, and each distro often learns from the others.
You do not have a moral or legal right to do absolutely anything you want.
The value of Windows is the huge legacy software collection that it runs.
.NET, of course. But apps written in .NET would be fairly new apps. What about apps written in older languages? Some apps may be trivially recompiled. Or recompiled after significant effort. But some apps may be slightly or even deeply wedded to the x86 and maybe x64 architectures.
A "windows" platform that cannot run that software base is basically a new platform. Starting from scratch. Sort of like a new version of Linux, or Hurd, or something new.
Microsoft may port their own apps. Great. But what about third parties?
There is
Other apps may be wedded to legacy languages that may not get ported. Will Microsoft be porting Visual Basic 6? Visual FoxPro? What about Delphi? Etc.
Even if a developer can fairly trivially build their app for WOA, why would they if there is not a large user base on that platform ready to fork over money? The developer has to expend effort (eg cost) today on a platform where customer demand may not materialize. If WOA doesn't run an end user's favorite legacy applications, then why would the user want to migrate to WOA? It's cheap and easy to buy a desktop or laptop running legacy Windows that is familiar and runs your legacy apps.
WOA has a chicken and egg problem. Which came first? The large number of third party apps / developers supporting WOA? Or the large end user base running WOA?
Furthermore, a developer who expends effort to port their product to WOA, even if "porting" is little more than a trivial recompile and repackaging, and tracking new SKU's, that developer will want to be compensated for that additional effort. You can bet that developer will want top dollar (eg price gouging) for that new WOA edition of their product. Do you really think you'll see a $99 Photoshop on WOA? Also don't expect a free upgrade to the WOA edition of your current application.
WOA may be Microsoft's counterpart of the PS/2 moment. The market may "just say no" (as they say in the '80's). The problem with PS/2 was that it was nothing more than an attempt to recapture IBM's monopoly using a new platform. That is what WOA is. Microsoft wants their legacy monopoly on these new mobile ARM platforms, just as IBM wanted their legacy mainframe monopoly in the PC market.
Another problem is that these new platforms are fundamentally different. They bring things that legacy PC's don't have deeply baked into the system and applications. Android for example can support both the legacy keyboard / mouse setup as well as touchscreen and voice commands. Those pesky new PC's offered a lot more than a mainframe terminal had, such as mouse and GUI. Oh, and cheap software, just as the new platform app stores offer pretty cheap apps.
Want to see WOA go exactly nowhere? There's an app for that!
I'll see your senator, and I'll raise you two judges.
You are probably right.
That is why I believe WOA will fail.
Leaving the legacy applications behind means WOA is a new platform. Starting basically from scratch. Competing with already entrenched players (iOS and Android). It starts off with little third party software where iOS and Android already have a huge base of developers.
I'll see your senator, and I'll raise you two judges.
It's possible that this is the first step toward leaving 'Windows legacy applications' behind. They are going to have to do it sometime, and ARM isn't similar to x86 like Atom is, so this may be a logical starting point.
Exactly. It's legacy software that's tied mainstream computing to the x86 architecture for two decades. Tthe x86 is a power-hungry architecture best suited for desktop and rack computing, but recent trends are towards mobile computing. Laptops started to outsell desktops. Netbooks hit the market. Smartphones reached a level where they could be used as complete computers. And the Tablet PC concept, which had been bubbling under for over a decade, finally found it's niche as a "maxi-smartphone". x86 is dead, and MS need to encourage people to produce standardised, architecture-neutral code if they're going to migrate to another architecture.
If they don't migrate to another architecture, what happens? Smartphones with HDMI out (and a built-in focus-free laset picoprojector) and Bluetooth or USB for keyboards and mice displace the traditional computer. If Windows is still split between phone and desktop, Windows dies.
So why the separation between desktop and phone OS? Why WOA an not just maintain their ARM version as Windows Phone? Because right now, phone apps are phone apps, and desktop apps are desktop apps. An OS lives and dies on its third party software, and this move is calculated to ensure that there is a back catalogue of desktop software available when the two paradigms merge.
It's a smart move, and shows a lot of foresight. Google should take note, and start working on standardised compatibility layers that encourage Linux app developers to produce software that can be easily migrated to Android.
Got them moderator blues I blieve I walk out the do', With these mod-points I been gettin', I 'most never post no mo'
Exactly. It's legacy software that's tied mainstream computing to the x86 architecture for two decades.
It's legacy software that's tied people to Windows for two decades. Break the compatibility and no-one needs Windows any more.
But if you develop your apps in WinRT, the same code will be able to compile into ARM or x86. I don't see a big deal, honestly. It's not like it will take developers extra work, since .NET and the JIT compiler should handle that workload. In fact, it makes Windows a more appealing development environment because you're hitting multiple platforms, form factors, etc... all with a single set of code.
But correct me if I misread.
The price is always right if someone else is paying.
Why? So far t his is just the ARM version. It sounds more like they just are going the cheap route, and not fulling integrating the ARM version with their Intel version.
Basically to my thinking: A) Other Win8 versions have these features, then this is laziness. No borg icon warrented. B) All Win8 versions lack these features (then why the big deal about the ARM version?), then this is a closing of the walls intentionally for a purpose. Borg icon deserved.
There is another possibility: Microsoft has massive legacy commitments. Practically all the world's boring corporate stuff that isn't old, specialized, or hip, enough to be running on some sort of intimidating big iron or linux/web/cloud/thingamawhatsit. Most of that software is absolute dreck, and rather boring, but much of it is also quite critical to a variety of high value operations and impossible or uneconomic to port or even modify very much. For this reason, Microsoft's walled-garden options on x86/64 are pretty minimal. Architecturally they could roll it out tomorrow(Software Restriction Policies are basically that, but under the control of your domain admin); but the customers that matter would scream like nobody's business.
However, since there isn't any legacy Windows software or legacy Windows device drivers, on ARM, since it has never run on ARM before, there is no legacy market to worry about. Microsoft has a free hand, more or less. As with the xbox, the other recent situation where MS started clean, without legacy impediments, they apparently see a walled garden as their best option.
It remains to be seen how long the momentum of more-or-less-open x86 IBM compatibles will carry them into the future; but so long as the legacy/in-house/custom demand is there, they'll be hard to kill entirely. However, I'd say that it is "outlook not so good" for open platforms any time somebody starts a new one from scratch...
It's API compatible minus features not physically available on a tablet. A quick recompile and voila, instant port.
You're not a software developer, are you? Because I wouldn't want to work with someone who thinks that they can just do a 'quick recompile' and ship a product out the door.
And, in any case, that wouldn't help the bazillions of old, unsupported Windows programs that keep people tied to Windows.
The problem with Microsoft's new walled garden is that they have an already entrenched walled garden competitor: iOS.
They also have an entrenched open competitor: Android.
Windows on ARM is just Microsoft's PS/2 like attempt to recreate its monopoly on the new platform where all the excitement and momentum already is. It's not that Microsoft's existing Windows platform doesn't also have momentum, but that is already in a very long slow decline and Microsoft knows it. (I hope they know it, since the decline of PC sales and thus Windows sales was in their quarterly report.)
I'll see your senator, and I'll raise you two judges.