For the first time in fifteen or more years, Redmond faces a genuine challenge to its Windows desktop monopoly. The threat isn't coming from Linux or from Mac OS X or from any other operating system. It's coming from a whole new computing concept: the "post-PC." The worry is that upstart tablets threaten to drive the computer out of the home, taking the Windows operating system with it.

It's not just Microsoft that's facing a tumultuous revolution, of course—the PC as a platform, as a concept, is equally under attack. But the biggest loser from this new world order will surely be Microsoft. Hardware makers can just switch to making new hardware, but Microsoft needs that hardware to run Microsoft software, and the company has been consistently unable to crack the tablet market.

Microsoft is no newcomer to the tablet market; in fact, the company has been in the tablet market longer than almost anyone else. But success in this market has been hard to come by. Microsoft's hope, the PC's great hope, is Windows 8. With Windows 8, Microsoft needs to build not just a Windows that PC users want to use; it needs to build a Windows that can succeed in the post-PC world.

Perseverance

One way or another, Windows has run on tablets for most of its existence. Windows for Pen Computing was an extension to Windows 3.x that added support for styluses, an on-screen keyboard, and handwriting recognition—and it appeared all the way back in 1991. A followup, Windows for Pen 2.0, shipped for Windows 95. Neither found much success outside narrow vertical markets, where the ability to write on the device or to use it while standing were particularly important.


The Compaq Concerto was an early tablet PC, running Windows for Pen Computers. It sported a detachable keyboard and a trendy stylus.
Wikipedia
Still, Microsoft persevered. Windows XP Tablet PC Edition appeared in November 2002, and was updated in August 2004 to Windows XP Tablet PC Edition 2005. Though these editions boasted improved handwriting recognition, and though Microsoft shipped its well-regarded tablet-friendly app OneNote in 2003, market impact remained slight.


An HP tablet PC running Windows XP Tablet PC Edition. It too has a stylus.
Wikipedia
The same pattern continued with Windows Vista, and then with Windows 7. Tablet features are no longer branded separately, instead serving as part of the standard operating system. But that doesn't mean they're widely used.

Competing approaches

For two decades now, Microsoft has had a consistent tablet ideology: tablets are PCs. They run PC operating systems, they run PC software, they have the full flexibility that the PC has. This point was explicitly made just last month, when Andy Lees, President of the Windows Phone Division, said:

We view a tablet as a sort of PC. We want people to be able to do the sorts of things that they expect on a PC on a tablet, things like networking to be able to connect to networks, and utilize networking tools, to get USB drives and plot [sic] them into the tablet. To be able to do things like printing, all of the things using Office, using all of the things you would expect from a PC and provide a hybrid about how you can do that with the tablet, as well.
Apple's iPad comes at the tablet question from the other direction. For the iPad, Apple took its successful smartphone operating system and scaled it up for larger screens. All the limitations of the smartphone operating system were retained on the larger device. The result? The iPad isn't a PC. You can't run regular software on it—you can't even run software that hasn't been explicitly blessed by Apple. There's no user-visible filesystem. There's little scope for attaching peripherals. Multitasking is restricted. Even the user interface is severely limited, with its full-screen single-tasking concept.

Yet the iPad has been hugely successful. It's slim, lightweight, goes for many hours between charges, and it's easy to use. It's also built around a touch interface; as good as a mouse and keyboard are (and for serious text entry or complex software, they really do work very well), they're also very limiting, as they're awkward to use when not sitting down with the device on a lap or desk. Touch makes the iPad usable standing up, lying down, or slouching on the sofa. Is the iPad a limited device? Sure. But the limitations don't really matter for the things people use iPads for; millions of people have found that the iPad's upsides more than outweigh its flaws.

Welcome to the post-PC era

So successful has the iPad been—and so unsuccessful were Windows tablets of old—that the iPad's flaws and limitations have been extolled as virtues. Steve Jobs, when announcing the iPad 2, described the new "post-PC" world:

That these are post-PC devices that need to be even easier to use than a PC. That need to be even more intuitive than a PC. And where the software and the hardware and the applications need to intertwine in an even more seamless way than they do on a PC.
The limitations that iOS imposes on users and developers alike are assumed to be essential to the creation of an effective tablet; the restricted software, the limited connectivity, the rigid user interface constraints, all are required to sell tablets. This is the post-PC world: sealed systems that trade flexibility and connectivity for appliance-like simplicity, casual interfaces, and convenience.

The iPad is king of this world, and though PC sales still dominate, they're flat while tablet sales are growing fast—so fast that the post-PC is widely seen as a threat both to Microsoft and to the PC as a whole. Microsoft can survive short-term difficulties in the smartphone market, as smartphones aren't really PC substitutes. A failure to tackle the tablet market, on the other hand, could be disastrous, a point the company has noted in its most recent 10-K filing.

Avoiding the beaten track

The decision to call, or not to call, a tablet a "PC" goes beyond mere branding. It influences the entire approach that Microsoft takes. It colors the user interface design, the sales model, the hardware requirements, the options available to system integrators. Redmond has considered scaling up its smartphone platform the way Apple and Google have both done, but as Lees put it:

Now, a lot of people have asked me, are we going to produce a phone that is a tablet? You know, are we going to use Windows Phone 7 to produce tablets? Well, that is in conflict with this strategy.
This does make some sense. The tablet's size, which has a considerable influence over interface complexity, is clearly more PC-like than it is smartphone-like. The usage model of the tablet is also closer to that of the PC than it is to the phone. Smartphone usage is brief and focused: sending a text to say they you're going to be late, checking a map to see just how lost you are, finding a restaurant. Even smartphone games tend to have lots of short levels so that you can dip into them during an idle moment, then quit without losing progress.

Tablets, due to their size, are never going to be the first choice for that kind of instant gratification. But the tablet remains a much better fit for longer, more considered use cases: settling down with a book or a movie. Writing a letter. Recreational—rather than informational—browsing. This usage style is a lot more PC-like than it is smartphone-like, and an interface designed around smartphone use—quickly jump in, find what you want, and get out—may not be the best option for a tablet.

Still, this tablet-as-a-PC model hasn't worked well despite 20 years of trying. Microsoft's decision to stick with it might look like a mistake—why would this approach start working now when it hasn't before?—but signs suggest it might be more successful this time around.

PCs aren't the problem

Windows didn't suffer on tablets because it was a "PC operating system." Even if Windows had all the same restrictions as iOS—all software has to be authorized by Microsoft, limited multitasking, all programs have to run full-screen, no access to the file system—it would still have been a poor fit. Windows' tablet aspirations have always been grafted on, and the hardware has been, well, wrong.

Whether it was Windows for Pen, Windows XP Tablet PC, or the current Windows 7, tablet accessibility has been treated as an optional extra that you can bolt on to the rest of the operating system. As a result, it's never worked very well. Yes, Windows supports handwriting recognition and an on-screen keyboard, but these aren't well integrated into applications. Windows is stuffed with icons, menus, and toolbars that are too small to be easily prodded with a finger because they've been designed for pixel-perfect mice and physical keyboards. Even simple actions such as toggling the visibility of the docked on-screen keyboard take a second to perform, as every open application has to resize its windows to make room for the keyboard—an unacceptable delay for such a common task. iOS and Windows Phone both pull off this feat instantly.

Windows has demonstrated throughout its life that there's a huge gulf between an operating system that supports stylus and finger input—which Windows does—and an operating system that actually supports tablets—which Windows doesn't. Tablet support isn't about supporting the hardware but about supporting a particular style of human interaction. Tablet users should never have to fight with the interface to perform basic operations like clicking buttons, but that's what Windows, and most Windows applications, makes them do today.


Windows 8 will work well with fingers. No stylus necessary!
Microsoft
Windows 8 will be the company's first genuine tablet operating system. The user interface already demonstrated does something that no current version of Windows has managed: it gives touch users an interface that's actually sympathetic to how they use their tablets. We've only seen a glimpse of what the interface will look like, but that glance shows that Microsoft hopes to do all the right things with easy-to-hit targets, seamless gesture integration, and a streamlined on-screen keyboard.

But the OS will retain its PC roots.

Inspired by, not slavishly replicating, the smartphone

The new interface borrows concepts from smartphones. The look and feel is heavily based on the "Metro" look pioneered by Windows Phone. However, these concepts have not simply been naively duplicated; they're refined to better fit with the way tablets are used. Window management is fiddly without a mouse, so the touch interface uses full-screen applications, just as Windows Phone or iOS do. However, it includes a split-screen multitasking concept not found on the phone. After all, tablets are generally more powerful than smartphones and certainly have larger, higher-resolution screens. Side-by-side multitasking allows users to exploit these hardware advantages.

Similarly, task-switching is simpler—just swipe applications in from the side of the screen—rather than replicating the indirect task switchers found in Windows Phone and iOS. Windows Phone and iOS switching makes sense on platforms where background tasks will generally be frozen. Windows 8 supports proper multitasking and so has an interface that permits and encourages rapid switching between open applications.

Success at matching the iPad's simplicity will also depend on Microsoft being able to provide a viable equivalent to the App Store. A Windows application marketplace is almost certain to be a part of Windows 8, but it has to be populated with high quality applications to be influential. The iOS App Store gives easy access to a vast number of, for the most part, small, simple, cheap applications. These aren't, in general, versatile enough and featured enough to serve as a genuine equivalent to or replacement for PC applications. But for casual gaming, catching up on streaming TV, or reading the news—and these are the kinds of applications that dominate the iPad charts—the App Store is unparalleled.

Hardware matters

But Windows can't do it alone. In the past, tablet hardware it has been a let-down. Traditional PC tablets have been about as big as laptops, about as heavy as laptops, with battery life as short as laptops, and as often as not, with a price higher than laptops. This was the case regardless of whether they were convertible PCs or genuine keyboardless touch-dedicated devices.

The march of progress has enabled slimmer, lighter, longer-lasting hardware to be brought to market. Last year's HP Slate 500 was a 1.5 pound x86 PC with a claimed 5 hour battery life, giving a hint of the kind of thing that's now possible. The Slate 500 was truly tablet hardware in need of a tablet operating system.

Of course, x86 isn't the only processor architecture. Microsoft's decision to support running Windows 8 on ARM processors is clearly a decision made with at least one eye on the tablet market. For all the efficiency gains that Intel has made with its processors, if you want decent performance while only sipping power, ARM processors are today the only realistic choice. Combine ARM processors with the strict hardware control that Microsoft is said to be exercising, and devices that are every bit the equal of the iPad in terms of convenience, portability, and seamless integration between software and hardware are perfectly possible.

Those things that Steve Jobs describes as essential to the "post-PC" are well within Windows 8's grasp.

Windows 8: still a PC

Still, with Windows tablets coming to market a good 2+ years after the iPad launched, "every bit as good" isn't really good enough. It certainly doesn't justify waiting so long to launch a suitable operating system just so that the tablet can "be a PC" when a "non-PC" Windows Phone-based tablet could also be "every bit as good" and shipping right now. The "PC tablet" needs to be better than the "post-PC tablet."

Microsoft appears to have made a key assumption: that the PC can bring useful things to tablet users, who will be better off with a PC-based tablet than a post-PC one. That's coupled with the assumption that this value can be provided without compromising the virtues of either the post-PC or the PC.

The PC's biggest strength is its enormous legacy of software. Bringing that strength to bear on the tablet market would be a great coup for Microsoft. And yet, if it were easy, the company would already be the runaway winner in that market. For 20 years now, the company has tried to square the circle—expose the wealth of PC software to tablet users—and it has met with consistent failure. Most software developers (including Microsoft itself—notice how Office doesn't have a touch-friendly user interface) have ignored the tablet completely; the few that produced software for it tended to produce specialized applications for line-of-business markets.

That's changing. The software industry is realizing that tablets are a market worth tapping and is at last showing some willingness to produce software adapted to the constraints of touch interfaces. That starts to open up new possibilities. Consider the stalwart desktop productivity suite. Apple offers versions of its iWork productivity applications for the iPad—they're some of the few big-selling non-game, non-media applications on the platform—but they're pale imitations of the real desktop applications, much less of Microsoft's Office. Because they're different apps, they have problems like documents not displaying properly or formatting getting broken. Microsoft, in contrast, could offer a touch interface to a proper Office, with the same behind-the-scenes work on page layout, tracked changes, formula handling, macros, and more. The touch user interface would probably still need simplification—but it won't ever break anything, it won't ever look wrong, and you won't ever screw up a document by making a quick edit on your tablet.

The power of the PC—when you need it

And then, what if you need something more? The iPad lets you attach a Bluetooth keyboard for more efficient text entry. But that's all it does. You carry around the bulk of a keyboard, and you're still stuck with the same old software. A PC has the option of noticing that you've attached precise input devices and adapting the software accordingly. So break out your Bluetooth mouse and keyboard and have access to the full Office experience; no need to use cloud or wired sync, no need to interrupt what you're doing. Just enjoy the most appropriate interface to the hardware you have available.

This isn't to say that Windows 8 will necessarily offer this kind of ability—but it's the kind of thing that can be done when you have a PC available. It allows options that don't exist on the iPad or Android. Even without this kind of dual user interface facility, there's always the option of falling back to running plain old PC applications. That's still valuable—but an adaptive, intelligent approach is probably better.

For all the talk of the post-PC, Apple recognizes this, too. Not for nothing does Mac OS X Lion include a host of interface concepts that originated on iOS. The aim is to wed the flexibility and power of the Mac (which is, after all, just another kind of PC) to the simplicity and convenience of the tablet. Apple's decisions haven't been universally popular, and there's plenty of refinement still to do, but the intent is there. If we were truly in a post-PC world, such a move would be pointless. Just slap iOS onto a MacBook Air and call it a day. Lion is an admission that for many of us, the post-PC remains an unacceptable backwards step.

Hardware support is not to be ignored, either. Printing, for example, still matters to many people. So does the ability to plug in USB gadgets like card readers, digital cameras, and hard disks. As our own Ben Kuchera found when reviewing the Toshiba Thrive tablet, this can be much more convenient, familiar, and easy-to-use than Apple's "no filesystem, no USB" approach.

So when Windows 8 is eventually released, the value proposition could: you can get an iPad, which is great for Web browsing, light e-mail, watching movies, and playing Angry Birds. Or you can get a Windows 8 tablet which can do all that, but which you can also use to write your resumé, or crunch those numbers that the office sent for your big presentation tomorrow, or play Flash games on Kongregate. Done well, that's compelling. Why would you go for the lesser device?

Is there a market?

The PC can certainly offer more. Do people want it? There's some evidence—anecdotal, admittedly—that people do want more out of their tablets than iOS or Android can currently offer. Many people appear to be buying Android tablets and then returning them, with numbers ranging from a plausible but high 15 percent to a rather alarming and dubiously sourced 40 percent. Manufacturers continue to dispute the figures, and yet word keeps coming from retail channels that the non-Apple tablets are experiencing large numbers of returns.

One retailer's account also offers an explanation for this: the tablets are essentially being oversold. People are picking them up in the expectation that they can replace a laptop, but they can't.

This explanation has an intuitive appeal. Especially at launch, tablets like the Motorola Xoom have prices that are comparable to those of low-end laptops. Neither class of device is pocketable, either. If you're spending as much as you'd spend on a laptop on something that's got similar portability constraints as a laptop, it's not altogether surprising that you might want as much functionality as a laptop.

There's likely some demand from laptop buyers, too, for some kind of "casual" user interface, application store, and so on—at least some of the time. A traditional interface and a keyboard and TrackPoint/touchpad makes more sense when I'm working, but when I'm casually flicking through Wikipedia and IMDB to see if the person in the film I'm watching really was thingy from that TV show, the tablet model works well.

Redmond can certainly find an audience for tablets that do more than "post-PC" devices. The size of that market is less obvious, however. Not everybody will care about PC features—if you never edit documents on your tablet, the ability to use better software for document editing is irrelevant. What Microsoft appears to be counting on is that most people will want at least some "PC" functionality.

Don't break immersion

Perhaps the biggest risk to Windows 8's tablet ambitions is that "pure" tablet users will get something other than a pure tablet experience. As the Windows 8 demos showed, the traditional, legacy desktop remains a part of Windows 8: just swipe it into view and there it is. To really tackle the iPad, it must be possible to use a Windows 8 tablet without ever seeing this view.

It would be easy for Microsoft to cut corners and say, "Oh, if you want to configure your Wi-Fi/uninstall a program/change your password, just use the legacy desktop." Such a move would be fatal. Windows has become infamous for allowing old bits of user interface to linger longer than they should. The best-known of these was the "Add Font" dialog, which remained a feature of every version of Windows NT, from 3.1 all the way up to Windows Vista. It was never updated because Microsoft thought it was never worth updating, that the testing cost didn't justify a purely aesthetic change. Windows 7 finally solved the problem by scrapping the dialog box altogether.

Until now, Microsoft could just about get away with this approach. The legacy bits made the operating system uglier and more inconsistent than it needed to be, but these leftover remnants of yesteryear did at least work. They won't work in a tablet operating system because the result won't be finger-friendly.

Questions remain about how completely adapted Windows 8 will be for touch users. There are important parts of Windows—Control Panel, for example—that are currently mouse-intensive. Windows 8 leaks show that at last some parts of Control Panel are being made touch-friendly, but just how far this will go is unknown. Updating everything for touch will be a huge undertaking, so there's probably a balance to be struck—but it needs to be considered carefully.

Microsoft is calling the new-style Windows programs "immersive" applications. The name itself is encouraging, as it brings to mind the notion of immersion in fiction and computer games, of constructing an environment that's believable, one in which you can't see the "edges." More importantly, it also brings to mind the problem of breaking immersion, of shattering the world that's been constructed and allowing reality—or the legacy Windows interface—to intrude. The legacy interface isn't bad per se, and it certainly works better for a wide range of programs, but nothing in Windows 8 should break interface immersion unless the user deliberately chooses to do so.

One way of ensuring the appropriate discipline would be for the company to produce a Windows 8 version that simply lacks the legacy desktop. (A good candidate for this version is the same version that's going on the tightly-controlled hardware.) This would have the dual benefit of allowing Microsoft to ship a cheaper Windows 8 license for these cut-price "pure" tablets without jeopardizing Windows license revenue from PCs; such a version might contradict the "tablet is a PC" mantra, however.

Concerns run the other way, too; not everybody wants a tablet interface on their PCs. IT departments, in particular, might not be keen to give users something so new and unfamiliar. The ability to turn the touch interface off, using Group Policies for example, seems essential.

Giving Windows 8 the tools to take on post-PC devices won't be easy. From what little has been shown so far, it looks as if Windows 8 will right the wrongs of 20 years of Microsoft tablet operating systems. But it needs to do more than this to take on post-PC devices.

If Microsoft is right in believing that people don't actually want limitations, and that they've been forced into those compromises because the iPad is the only device giving them the ease of use they're after, Windows 8 will be triumphant. Given the choice between tablets that offer the full power of the PC and tablets that don't, the PC ones could well win out. But that's still a big "if."

Tablets will never replace gaming PCs...so there's always that.

Tablets will never replace gaming PCs...so there's always that.

I thought this generation (or was it the one before) of consoles was supposed to doom PC gaming.

how do you lose mobility if you can snatch it up out of your dock and put it in your pocket?

No one is suggesting you put your 24" monitor and keyboard in your pocket.

The idea is, when it's in your pocket, the OS is optimized for the small screen and limited input options, but when it docks, it functions as a normal computer...

Because gaming on a tiny screen isn't fun?

The fact that the Nintendo DS is one of the best selling consoles ever would seem to disagree very definitively with you. Also, Angry Birds.

Because gaming on a tiny screen isn't fun?