Microsoft: Bring back the H/PC Pro

Microsoft: Bring back the H/PC Pro

Summary: Back to the Future? The HP Jornada 820 was a "netbook' before its time, circa 1999.

SHARE:

Back to the Future? The HP Jornada 820 was a "netbook' before its time, circa 1999. It ran on Windows CE with a low-power StrongARM CPU.

Netbooks and MIDs are gaining more and more popularity as value-oriented customers are searching for low-cost solutions for everyday computing tasks, which include messaging and web application use.

Currently, the most popular netbook/MID platform is a fairly common reference standard based on the Intel Atom processor. As such, it's an x86 system that's no different fundamentally from the computing paradigm we've been using for regular desktops -- it uses Windows XP and it runs the same applications. However, it is subject to the resource constraints of it being a less powerful CPU and having less storage capacity and memory than a full-blown laptop or desktop computer.

Click on the "Read the rest of this entry" link below for more.

I don't have any problem with Windows XP on netbooks, it's a stable OS and has broad support for applications and drivers. When XP is eventually retired, future netbooks will use special OEM builds of Windows 7 to provide similar functionality. But looking even further into the future, maybe Microsoft needs to re-think whether netbooks and MIDs really should be running on the same platform that run on our laptops and desktops. The install footprint for desktop Windows is really too bloated in terms of local storage and RAM requirements for a netbook or a MID, and it's subject to the same viruses and security exploits as a desktop OS. When you are on the road, the last thing you want to worry about is malware-defending your ultraportable device.

Google is already thinking of releasing their own netbooks with OEM partners using the Open Source Linux-based Android operating system. Initially, these netbooks will use similar hardware designs to that their Windows XP and Linux counterparts use now -- the same Intel Atom netbook reference that systems like the Dell Mini and the Asus eeePCs use.

However, there are some manufacturers that are looking into low-power embedded chipsets which will provide superior battery life and much better device integration, such as the ARM-based processors and the that are in smartphones like the iPhone, the BlackBerry and the T-Mobile G1. Because the target platform is going to be Android, there's little concern for backward compatibility with existing applications -- a whole new Android ecosystem is going to be developed for it, and none of the undesirable baggage from PC OSes is going to go along with them either. As we like to say in the systems integration biz, it's a "Green Field".

Also Read: Why Apple and Google Need to Get Into the Netbook Business

Likewise, it is possible that Apple may decide that instead of building a $400.00 Mac OS-based netbook, they will continue with their existing trend towards consumer electronics devices and release a MID that is essentially an over-sized iPod Touch with a keyboard. This would allow them to leverage their iPod and iPhone developer ecosystem and provide a low-power device with high performance that would truly differentiate itself from just any other x86 netbook that everyone else is making, and would probably allow Apple to justify a price tag of around $500.00, twice that of some budget netbooks such as the Dell Mini 9, which can be had in a SSD configuration for about $270.

The inevitability of this return to low-power non-x86 embedded chipsets with netbooks and MIDs is ironic because Microsoft actually introduced devices just like this 10 years ago. Some of you may not remember, but "netbooks" were once called Handheld PC Pros, or H/PC Pro for short. They were a short-lived Windows CE-based device standard produced by a handful of manufacturers (HP, Fujitsu, NEC, Philips, Samsung, Sharp and Vadem) that had a 90 percent sized keyboard, a trackpad, and a VGA resolution screen, and included embedded versions of Outlook, Microsoft Word, Excel and PowerPoint. I reviewed one of these units, the HP Jornada 820, for another publication back in 1999.

Why did the H/PC Pros fail? Well, for starters, they were too expensive for the limited functionality they provided. As I recall the Jornada 820 sold for about $800 in 1999 money. Windows CE also was also not a popular platform to develop for, as the embedded OS of the time which had the most popularity was the Palm OS that ran on the PalmPilot. The Internet had also very few "clouded" applications to make the device useful (Hotmail was the biggie back then) and there was no ubiquitous wireless access to be had whatsoever. Wi-Fi was brand new and not even fully standardized, and the H/PC units had to be equipped with land line modems for dialup Internet access or pricey Sierra CDPD cards with expensive cellular data plans.

Fast forward to 2009. Embedded device chipsets are now dirt cheap and are produced in the tens of millions. Windows CE, now called Windows Mobile, has the benefit of having 10 years of additional maturity. The .NET Framework has also matured as a common API which can be used to develop embedded Windows applications much easier than it was back in 1999.

So why not bring the H/PC Pro back? Throw a Beagleboard-type integrated TI OMAP chipset into a netbook form-factor with a cheap 10" 720p LCD, put on Windows Mobile 6.5/7 with an updated Internet Explorer browser, port a stripped-down version of Microsoft Office to it, add built-in Wi-Fi capabilities, and you've got yourself a pretty killer netbook/MID platform -- one that can vastly outperform Intel Atom designs on battery life (think 12-16 hours of continuous use) and will have powerful integrated multimedia capabilities. Give away the developer toolkit for free and create an iPhone-like App Store for the thing, and sell it for $300 or less. Suddenly, the old H/PC Pro is starting to sound a lot more desireable.

Will we see a resurgence of non-x86 netbooks? Should Microsoft introduce a Windows Mobile-based MID? Talk Back and Let Me Know.

[poll id="7"]

Topics: Microsoft, Hardware, Mobility, Operating Systems, Software, Windows

About

Jason Perlow, Sr. Technology Editor at ZDNet, is a technologist with over two decades of experience integrating large heterogeneous multi-vendor computing environments in Fortune 500 companies. Jason is currently a Partner Technology Strategist with Microsoft Corp. His expressed views do not necessarily represent those of his employer.

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

Talkback

41 comments
Log in or register to join the discussion
  • I had a couple of those 820s ...

    ... still have them in fact. Batteries are a big old though ...

    Great tool, and I used mine a lot, (I augmented the storage by spending a small fortune on 8MB CF cards) but I can't see a Windows Mobile netbook being relevant today, and I can give you one example why ... buy an XP netbook and you have the choice of loading Win 7 or Linux onto it (some deviant is probably working on loading OS X onto them too ...).

    Not convinced by that? OK then, think about Moore's law ... why limit a machine to the embedded OS?
    Adrian Kingsley-Hughes
    • Oh, and as for your poll ...

      ... I'd add an "they were nice in their time" because I wouldn't say they were a failure.
      Adrian Kingsley-Hughes
    • ARM

      ARM netbooks are WinCE only (until/unless 7 is ported, which is unlikely).

      TripleII
      TripleII-21189418044173169409978279405827
      • A big problem for MS. 12 hour battery life would be so compelling, that it

        would far outweigh many other considerations. Top that off with better prices, and you can see that MS will have a big problem here. Do not forget, that Google could offer a WinLib based development environment for vendors to port Win32 apps to Linux and sell them on the Google store.
        DonnieBoy
  • RE: Microsoft: Bring back the H/PC Pro

    Adrian, I see what you are saying, but there is no way I think that an x86 Atom design will ever be as well-integrated or as well-performing as an embedded device chipset. With x86 we have 20 plus years of compatibility baggage. And If we're going to do Linux or something like the Mac OS that's in iPod/iPhone, why do we need x86? Take a look and see what one of those BeagleBoards can do, you'd be very surprised of how little wattage they use and how powerful their multimedia capabilities are.
    jperlow
  • The problem with Windows CE is that it is NOT a full OS, and it is very

    different to program for than Windows 7 (or a full/real version of Windows in general).

    I would expect for Linux based Arm netbooks to have a huge advantage over Windows CE / Windows Mobile based Arm netbooks. Here you have a real OS.
    DonnieBoy
    • Not if they use .NET framework

      Donnie if they use .NET as the programming standard for the device the difference between building for this versus Windows 7 will be minimal. If you can program in C# on Windows you'd be able to write for this thing. It won't have the same limitations as the old Windows CE.

      I agree that Android units will be compelling. But there will always be consumers that want offerings from Microsoft, and this is a way Microsoft can remain relevant and also provide value-add for its Cloud strategy.
      jperlow
      • True, and MS may have a group working behind the scenes to be ready. But,

        they had better be fast, as Arm netbooks are coming, and 12 hour battery life will be a HUGE seller and far outweigh other considerations.
        DonnieBoy
      • Another thing Jason, that sounds all nice and good, but, still, Win32

        applications will not run, and virtually all of the existing Windows applications can not be recompiled for either CE or Mobile, so, MS will NOT have a major advantage over Linux based Arm Netbooks, and in fact might have a little disadvantage given the wealth of Linux applications that WILL run on Arm based netbooks with a simple recompile.

        To top that off, Google could offer a WineLib based development environment to help vendors port existing Win32 applications to Android Arm netbooks and sell them on the Google store.
        DonnieBoy
        • You're not thinking about the infrastructure

          MS has Activesync which will allow all those corporate mobile workers to sync seamlessly with their Exchange back-ends. Also native apps written to talk to Sharepoint can also be written. There are .NET environments that would be able to take better advantage of these devices much more than they would something like an Android.

          Just because you can't run the same exact kind of apps doesn't mean that Microsoft doesn't have a clear advantage. Everyone else still has to integrate with those MS back-end services as well.
          jperlow
          • The inability to do Win32 takes away one of the big reasons for Windows.

            There is just much out there in terms of .Net desktop applications.

            Without a full OS, MS does not have a compelling advantage over Linux on Arm based Netbooks.
            DonnieBoy
      • You mean .NET Compact Framework don't you?

        Windows CE was low powered because there weren't any high-res timers to put unhelpful demands on the cpu. IOW it could deep sleep.

        So the full .NET framework would be unsupportable, as it requires so much of the desktop PC infrastructure to work.

        BTW Windows Mobile is an umbrella term for two (still) separate products.
        [B]Windows CE 6.x[/B]
        [B]Windows Mobile[/B]

        They are both targettable from within the same developer tools
        Custard_over_2x_Pie
        • Initally, yes.

          I've seen .NET Compact apps on smartphones and such. Initially they could put together a platform with .NET Compact but like any other hosted environment with a VM (Java, etc) they could build a fatter subset. I'm not exactly sure how many of the .Net APIs they would need to port over to build a robust device, but I assume it would be more than what they have for PDAs/Smartphones now. Whether this would be a pure .NET device or a mix of .NET and other Win32 APIs is up for speculation. My point is that they can build a robust platform that would be fairly easy to develop for and would not have the limitations of previous incarnations of Windows CE.
          jperlow
          • Actuallly I don't see why they couldn't bundle C# express in Win7

            Partnered with microsoft's online courses, could go some way to stimulating hobbyists, if nothing else. Which is something that has been sadly missing for about ~20 years.

            That's how I got started. Having a universally supported programming language/interpreter, that is guaranteed to be on every machine, makes your target audience huge.

            With XP, .NET it's an optional component, so there's an instant barrier to implementing .NET code, so I've stuck to C/C++.

            Microsoft has the XNA studio for targetting just XBOX360, but to get your code made available to XBOX users, [B]you have to pay microsoft[/B], so as to get it approved.

            Apologies for going slightly off-topic.
            Custard_over_2x_Pie
  • Also Jason, you should mention that the big thing that will drive Arm

    netbooks is that they will use a lot less electricity, and the same time being cheaper and just as powerful.

    We might see netbooks that can go 12 hours very soon. Especially if they would adopt at least the screen technology from OLPC.
    DonnieBoy
  • How about if netbook manufactures cherry pick some of the OLPC tech?

    The screen being the most obvious. In black and white mode using ambient light, it uses about zero energy. Also, by making the memory for the screen separate, and other little tricks, they allow the processor to go into deep sleep for even short periods of inactivity, thus saving a lot of energy.
    DonnieBoy
  • The quickest way to Win32 on ARM is Linux + WineLib!!!

    Of course software vendors need to recompile the applications and do compatibility testing. But, the same would be true if MS ported full Windows to Arm.
    DonnieBoy
  • Can't vote, no option for my thoughts.

    WinCE V6.5 will be a failure. WinCE with a real revamped Windows mobile 7 that is in some way shape or form compatible with Windows 7 (a tall order, but what it will take), then it would be a good idea.

    Look at the alternative. You will have FF no matter the CPU. From a MID to ARM to Netbook to Desktop to Full size laptop, you will have access to the ENTIRE suite of Open Source software. CE never offered that, it was a fork with few applications common across desktops.

    TripleII
    TripleII-21189418044173169409978279405827
    • Yes, good points, MS must offer Win32 compatibility, and be a complete OS.

      The best route for them would be to continue to strip the fat out of Windows 7 and port it to Arm. That would also help for normal desktop / laptop version.
      DonnieBoy
      • Windows Mobile 7

        That's exactly what Windows Mobile 7 is expected to be. That's where all the "minwin" kernel research is going.

        The problem however with running a complete Win32 stack in an embedded processor is that you can't port ALL of the Win32 APIs. You don't need all the legacy cruft in there. Just .NET framework.
        jperlow