The Windows Mobile-Windows continuum: The plot thickens

The Windows Mobile-Windows continuum: The plot thickens

Summary: One of the most interesting question-and-answer exchanges between Wall Street analysts attending Microsoft's Financial Analyst Meeting (FAM) and company execs has been around the convergence/divergence between mobile devices and PCs.


One of the most interesting question-and-answer exchanges between Wall Street analysts attending Microsoft's Financial Analyst Meeting (FAM) and company execs has been around the convergence/divergence between mobile devices and PCs.

Wall Street analysts attending the July 30 confab had a number of questions about how Microsoft intends to address the new CPU and GPU architectures that are emerging in all kinds of mobile devices, ranging from phones to netbooks.

Up until now, Microsoft has maintained a clear division. Mobile phones run Windows CE/Windows Compact at the base level (with the Windows Mobile environment layered on top). PCs run Windows.

But the Windows and CE lines have been starting to  blur -- and the effects are being felt not just by Microsoft, but by its competitors, as well. (Gizmodo reported recently that there has been talk that Google's Chrome OS may end up on mobile phones and not just netbooks. And Google's Android is being ported to both phones and PCs, as well.)

As one Wall Street analyst attending Microsoft's FAM noted, Intel is working on new lower-power x86 processors that could find their way into phones and other consumer-electronics devices.

Chief Research and Strategy Officer Craig Mundie had a ready response. He noted that Microsoft has been supporting for years a variety of processors with Windows CE and Windows Mobile, "so the technical transfer is not a big leap for us."

"For us, the X86 environment is one that we know well and have an easy time supporting. Clearly, adapting that to the environment that we have in the embedded and the phone business would take some work. But it is not a monumental amount of work as it would be if it was an architecture that the company didn't already have a heavy investment on. So I think our view and (Entertainment and Devices President) Robbie and the business team will have to make the call as to how we see that evolving."

Bach, for his part, hedged a bit more.

"If they (new x86 processors) go into phones, there is probably work for us to do. but we have a pretty good handle on the X86 environment and we can make a business decision based on volume and our operator customers are looking for."

Interestingly, no one asked Microsoft during the FAM Q&A about how, when and whether it might make a version of Windows available on ARM processors. Microsoft has said that Windows 7 has not (yet) been ported to ARM processors, which some company watchers have viewed as leaving the door wide open for Linux on ARM netbooks. Windows Compact, however, does work on ARM processors.

In response to another analyst question about Microsoft's mobile-OS plans, Bach did note that the Windows and Windows Mobile team are working more closely together.

"Today we actually already share components between what we do in the Windows space and the Windows Mobile space. You are going to see more of that share continue over time and you will see us accelerate it and do more," Bach told analysts. "Understand historically that was tough because the underlying architectures were actually quite different. But to Craig (Mundie)'s point about GPUs and CPUs and the underlying chip architecture, as those get more similar, it absolutely makes it easier to share more. You will see that whether it is work in the browser, on development tools, a number of other places where we will be able to do more of that sharing. That's baked into our plan."

Reading between the Q&A lines, do you foresee Microsoft releasing Windows (and not just new versions of CE/Windows Mobile) for future smart phones?

More from the analyst meeting:

Topics: Hardware, Microsoft, Mobile OS, Mobility, Operating Systems, Processors, Software, Telcos, Windows


Mary Jo has covered the tech industry for 30 years for a variety of publications and Web sites, and is a frequent guest on radio, TV and podcasts, speaking about all things Microsoft-related. She is the author of Microsoft 2.0: How Microsoft plans to stay relevant in the post-Gates era (John Wiley & Sons, 2008).

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • "3 screens and a cloud"

    This is a phrase that we will hear more and more from Microsoft. From a developers perspective the goal is one code base that compiles for Mobile, Web and desktop. The basis of that is Xaml/silverlight/wpf. We are not far from that now with projects like prism (on codeplex) and adoption of M-V-VM patterns. Windows Phone is the last part of the puzzle and v7 will have the Xaml/Silverlight/WPF foundation - that is what Bach is talking about here I believe. Not quite Windows 7 on the device exactly but I can see it being possible for a single app to run on all three screens (most likely with different Views/ViewModels and sharing data through the cloud) This is a winning combination IMO and I hope (and expect) we will see it earlier than most people think...
    • you mean like gmail web client, gmail cellphone client

      windows technlogies isn't the only way to do this, and is pretty much the clunkiest and most expensive way.
      Some will go this route, but I predict it won't prosper long term.
      • Yes there are other ways to do it of course...

        But there are 8 - 9 million .net developers who will most likely embrace the Microsoft way for mobile. Not sure why you say it is the most expensive though - unless you are speaking of the cost of the tools?

        I speak to many MS developers and most (if not all) are very happy or keen to start Silverlight development if they can; once Mobile supports the same model, it will be very compelling.
      • Actually, I see this as quite the opposite,

        that this will prosper long term.
        One program for all, not separate versions with conversions.

        As for clunkiest, I would disagree, as there have been years of quality apps for both Windows and WM/PPC, and I have had no issues with the mobile OS itself.
      • Expensive?

        You can develop on the .NET platform for free. It's not difficult to get the tools free from Microsoft.
      • No completely the opposite.

        The gmail way is the clunky way to do it, with different version for each, and some if not all being low grade clunky html/ajax.

        The W/SL/RTM way will be a single RIA across all 3 that adapts to resolution from VGAish phone to netbooks to laptops to large screen tvs.
        Johnny Vegas
  • RE: The Windows Mobile-Windows continuum: The plot thickens

    I don't think the OS *really* matters. It's the API that counts.
    I think you'll see something around using Silverlight on the mobile platform. This means across WinPhone, Nokia and others. The same apps will be able to run on the web and OOB (out of browser). I could also imagine seeing some integration with XBOX and the Mesh.

    A single platform enables a single appstore. With one appstore feeding the three screens, plus non-Microsoft hardware properties (other phones, Macs etc). Maybe this is what the OneApp moniker trademark will be used for?
  • RE: The Windows Mobile-Windows continuum: The plot thickens

    I think they will rather make Windows mobile(phone?) more advanced so that it could be easily used as a netbook/MID OS(32bit screen, OS GPU acceleration, up to 1280x720 screen support etc).
    Something like lite version of desktop Windows(different looks but almost identical possibilities). Kind of complementary system(which would fit well in their strategy).

    Making ARM version of desktop Windows 7 would kill WM.
  • I know you're a fan of the concept of Windows on ARM

    But I've yet to see a compelling argument for it.

    Window CE already provides a lot of core services and DX/CE handles the high end graphics. Since Google and Android cannot run any Windows apps either, there's no obvious competitive advantage here to port Windows to ARM.

    Add to that the fact that ARM based netbooks are still a theory, not a real device, and given the almost universal rejection of Linux on EXISTING netbooks in favour of Windows, we're left with the question of what marketing advantage ARM netbooks have.

    The usual argument is 'lower power CPU so longer battery life' - but that's a weak and suspect argument. First, the CPU isn't really the big power hog in most devices anymore. Remember, the Atom only draws around 12W full on and can drop to 5W. The ARM processor is better, but not THAT much better. Either are swamped by display, memory, storage and other power expenses. There's a reason PDAs and phones have small screens (in terms of resolution and size), limited memory and small storage using Flash/NAND memory- that's where the lion's share of power savings come in.

    On top of that, the use pattern for ARM devices tends to be rather different from Windows devices. ARM devices tend to be in sleep most of the time, and are intermittantly wakened to do a simple task, then go right back to sleep. This is why most iPhone apps are 'one-bangers'.

    Windows devices tend to be awake longer and doing more complex tasks.

    [Sidenote - this is why people who keep trying to argue that a netbook is not a laptop don't get it - they, like Ballmer, see a netbook as a glorified PDA - but the consumers of netbooks see them as cheap, compact *laptops* running Windows and their familar apps. The reality is that netbooks *aren't* underpowered - it runs most people's real world apps just fine as it is.. it's people who have surreal demands and define them as 'minimal' who have driven this argument.]

    The short version is - an ARM based netbook doesn't solve any problem that the general public perceives exists and causes all sorts of new ones that makes that platform undesirable by most people.

    Microsoft generally doesn't get involved with niche products - it's not cost effective for them to.
    • The trend is towards cellphones for email and the web

      I think netbooks are the last vestiges of the laptop paradigm, which is dying.

      As people use paper far far less, wordprocessors and DTPs are becoming irrelevant.

      We still write manuals as documents, but this is plainly doomed long term in favour of wikis which can be far more corroboratively worked on by their nature.

      Top end cellphones can already be used to effectively video-edit. I can read long documents on my regular cellphone using gmail.

      Apart from tasks like CAD, to me, the writing appears to be on the wall for the desktop/laptop/netbook.
      The cellphone market has never been bigger. The internet is now used hugely by cellphones.

      Windows mobile has been in this space for ever, but has both failed and lost market share. I think it's unlikely that a company which has failed over a long time with multiple initiatives can turn around this situation. There are simply too many players, and all the players are excellent.

      There is already a programming platform common to 4 billion cellphones (MIDP2.0) over which a mass of software is already being sold (including many great 3d games). It's hard to see how Microsoft's strategy of trying to add value by adding proprietary APIs is going to cut it in this market.
      • word-processors irrelevant.

        They are still 100% relevant in my software industry and will be 50 years
        from now.

        Wikis will never replace requirements and design.
      • "hard to see how Microsoft's strategy ..."

        We looked at MIDP programming, it was much more effort (and thus cost) than the current WinMo CF platform for our app.

        The Xaml/Silverlight/Wpf model will be easier still.

        Overall surely you use the api that brings the most value for your scenario and runs on a platform you can target. I think that will be a growing win for Microsoft for both consumer and corporate markets
    • I agree with many of your points, but ...

      ... don't discount ARM's power profile.

      ARM chips running at around 1GHz draw AT MOST around 2W - that's 1/6th of the power an Atom currently draws.

      In a small device with a small screen, small amount of DRAM, low-end graphics, WiFi, etc., that actually DOES account for a significant amount of batter power.

      Of course, in a machine with a bigger screen, more RAM, etc., the % of power drawn by the CPU decreases, but it's still not to be ignored.

      In the world of portable devices, every mW matters.

      ARM based devices, on the other hand, do present some challenges, primarily around app compat. Little existing software is going to work unaltered. Most games will not run without modification. Drivers? Ouch!

      The BEAUTY of Windows being ported to ARM is that it opens up the economics - a sufficiently large potential customer audience that many app, tool, hardware, etc., vendors who will be able in most cases to relatively easily recompile their apps for Windows/ARM.

      One of the major things that MS did in Vista and in Win7 was to eliminate and/or segregate the remaining assembly/CPU-specific code. 99% of Windows is now machine agnostic!

      Supporting a whole new CPU and machine architecture has its challenges, but it's not something that MS has never done before - they've ported Windows to Alpha, MIPS, PowerPC, Itanium and x64 and have also ported it to ARM once before (in a POC).

      In short, if MS thinks there's a big enough market for ARM powered nebooks, laptops and PC's they'll be there.
  • What I See Is A Total Mess

    my god, folks, MS is obviously flailing all over the map here.
    emperor ain't got no clothes.
  • What is taken Microsoft so long?

    They took the Mobile sector for granted and Apple came in and beat them to the punch. Now they have to play catch up. I have been waiting for something good by them. How long must I wait. I will never go to At@t, just is not gone to happen. Now, if iPhone comes to Verizon. I just might now wait any longer. Why, should I buy a 6.5 WinMo? When, supposedly there is gone to be a windows phone 7 by next year. Will the WinMo 6.5 phones be upgradable to seven? If not, you lost me.
    • good point

      It's why I am skeptical about 6.5 - you can download the sdk but it's not really an sdk - it's a DTK (Developer Tookkit), just an emulator and some docs on gesture api using native code, not managed code. It seems really strange that the developer info for a new version should be so slim, and that we are pointed to a book for documentation that is over 2 years old (

      Then we have the Mobile Marketplace for Developers that makes a point about "Marketplace will be pre-installed in every single Windows Mobile 6.5 device" but only has links to Windows Mobile 6 documentation.

      What is going on? Could it be that 6.5 is smoke and mirrors and 7 will be the next release? That would certainly make for a news worthy splash of the type Apple are so successful with...
    • They delayed WM7

      I think remember Mary reporting some time last year that there was an internal debate within Microsoft on whether to do a rewrite or start from the WM6 codebase. I may be wrong about the particulars.

      Well WM7 was supposed to release this year but since it's not perhaps they made the decision to do a major overhaul.

      WM 6.5 to me seems like a minor stop-gap release just to stop the market share bleeding so fast.

      Microsoft has the money, research, and smart staff to introduce some innovative technology next year. I hope they do but if they don't they might as well purchase Palm.
  • Midori Play?


    Where is Midori on all of this? Is M$ supposedly
    working on something like Midori that's completely
    managed code based and Windows rebuilt from the ground
    up? Is it possible M$ has moved along with this to the
    point that at least WM or W will be supplanted by this
    new OS in about 2-3 years? I thought the word was that
    WM8 was based on Midori? If so, then how does it
    relate to ARM versus x86? Personally, I have to think
    that Intel's work on Atom's SoC integration suggests
    that we're about 18 months from an x86 chip that comes
    very close to equally the low power consumption of ARM
    • Midori is not about replacing Windows

      Midori is about providing a next-generation subsystem in which applications can run safer and faster.

      Microsoft is not about to rebuild Windows using managed code - that makes no sense.

      But having a second environment in which you can run large parts of complex apps (and even OS features) which are verifiably safe would be a VERY welcome thing.
  • Yeah, I never got that.

    Both Microsoft and Nokia face alarmingly fast eroding marketshare in the mobile space due to outdated mobile operating systems. They both sit on cash hoards, so I don't understand why one of them don't just buy Palm.

    They would own WebOS, blam: problem solved.