Building Tablet PC Applications (Pro-Developer)

The Microsoft Tablet PC

Although the failure of Pen Windows to revolutionize computing left some at Microsoft disillusioned, there was still a set of core believers within the company who championed the cause of tablet computing. Out of this coterie would eventually arise, nearly a decade later, the Microsoft Tablet PC. The new Tablet PC initiative combines a set of hardware requirements along with a software platform to deliver a rewarding user experience.

The Birth of Microsoft Tablet PC

It was the last straw. Bert Keely, engineering director at SGI (Silicon Graphics, Inc.) in Mountain View, California, had managed the tablet computer project long enough to know that the company s ability to deliver on his tablet computing vision was waning. Although the project started out with a lot of promise, repeated decisions over the years had relegated it to its current status as a research project. And, as with most research projects at large technology corporations, the SGI tablet computer initiative seemed destined never to see the light of day. Cheap and powerful PCs were eroding the bottom end of SGI s market, forcing SGI to focus on higher-end servers. The tablet project would probably not survive. Interview with Bert Keely, Microsoft Corporation, May 2002.

Determined to see tablets rise to the forefront of computing, Keely began to shop the idea around Silicon Valley s elite. Steve Jobs, CEO of Apple Computer, was unwilling to pursue the idea out of what many thought was spite, directed at former Apple Computer CEO John Sculley, who fronted the failed Newton project. Jeff Hawkins at Handspring was uninterested for a different set of reasons he felt that the tablet computer proposed by Keely was halfway between the palm-sized devices that he was creating and the full-blown laptops that dominated the portable market. In separate discussions, Michael Dell, founder of Dell Computer Corporation, told Keely straight out that he would not be interested in such a device from a business point of view until a reasonable market demand was demonstrated. Although he was unwilling to expose his company to the possible market risk, Dell gave Keely a fateful parting thought try Bill Gates s Microsoft Corporation.

Microsoft was the last company that Keely would have considered approaching, especially given his background in the sometimes heavily anti-Microsoft community of Silicon Valley. However, his commitment to tablet computing brought him to an impasse. All other likely suitors were reluctant to carry on Keely s vision. So in early 1998 Keely headed off to Seattle to propose the tablet project to Microsoft.

His proposal was well accepted by the executives in attendance. Although it had been only a few years since the failure of Pen Windows, Bill Gates was still a firm believer in tablet and pen-based computing. It was decided that the time had come for Microsoft to launch a bold new venture into tablet computing. Microsoft senior vice president Dick Brass asked Keely to work in Seattle to begin the development of electronic books, ClearType font technology, and ultimately a new tablet computer.

Independent of Keely s entrance into Microsoft, many others were pulling together towards the nascent Tablet PC project. Chuck Thacker, best known as the chief designer of the Alto computer while at Xerox PARC, and Microsoft Distinguished Engineer Butler Lampson, had both worked on the DEC Lectrice reading device. Charlton Lui and Dan Altman had been foundational at Aha! Software, which produced pen-based applications for many pen computing platforms. Evan Feldman had similar relevant experience, having done usability research while at Compaq on their Concerto device. Alexandra Loeb, currently vice president of the Tablet PC division, pulled together these and other key people from various related projects throughout Microsoft in the spring of 1999. The Tablet PC team would eventually grow to more than 200 members, their efforts culminating in the fall 2002 release of the Microsoft Tablet PC.

Microsoft Tablet PC Hardware Guidelines

In order to be called a Microsoft Tablet PC, an original equipment manufacturer (OEM) needs to meet a set of hardware guidelines. Microsoft crafted these guidelines after conducting an extensive suite of usability studies to determine what sort of features users wanted in Tablet PC hardware. The full set of guidelines is presented in Appendix B. The main guidelines are summarized below:

• Digitizer

  The digitizer must be able to detect a hovering pen and take samples of the pen s location at least 100 times a second. It must also support a resolution of at least 600 points per inch.

• Power states

  The Tablet PC must be able to resume from standby in less than two seconds.

• Viewing mode

  It must support changing between landscape and portrait screen orientations without requiring a reboot.

• CTRL+ALT+DEL equivalent functionality

  A Tablet without an attached keyboard must have a dedicated hardware button that triggers the secure authentication sequence.

Microsoft Tablet PC Software

Meeting the hardware guidelines is the responsibility of each hardware manufacturer. Microsoft s responsibility, on the other hand, is to create and deliver a set of software features to support Tablet PCs and third-party software vendors. In order to support Tablet PCs, Microsoft created a version of Windows XP with Tablet PC features named Windows XP Tablet PC Edition. This special version of Windows XP, sold directly to hardware manufacturers, contains the following set of software features not available in any other version of Windows:

• Digital ink applications

  Several digital ink applications were written especially for Windows XP Tablet PC Edition. The first of these is Windows Journal, a note-taking application that showcases the world of possibilities afforded by digital ink. Sticky Notes is a handy accessory designed to capture random notes and reminders similar to its physical counterpart. On the lighter side, InkBall is a game best played with a pen the objective is to control the motion of colored balls by crafting the right digital ink strokes.

• Tablet Input Panel

  When using a Tablet PC without a keyboard, there needs to be a way to enter text into standard applications. The tablet input panel fills this role on Tablet PCs by providing a combination of on-screen keyboards and alternative text input methodologies. It supports character-by-character text entry in a style similar to that of Pocket PCs, but it also supports free-form cursive handwriting as a way to input text.

• Platform binary executables

  The final (and essential!) element of Windows XP Tablet PC Edition is the set of Tablet PC Platform binary executables. With the release of Tablet PC, Microsoft introduced a set of application programming interfaces (APIs) collectively referred to as the Tablet PC Platform. These APIs are provided for third-party software vendors to exploit the digital ink features made possible by the Tablet PC Platform. By including these binary executables with every copy of Windows XP Tablet PC Edition, third-party software vendors can be assured that their software will be able to leverage special digital ink features when running on Tablet PCs.

Most flavors of Windows (such as Windows 98, Windows Me, and Windows XP) have a set of requirements known as logo requirements. When a third-party software vendor s application meets these requirements, it is allowed to put a special logo on the software s packaging to indicate that it was written especially for Windows X (where X is the particular flavor of Windows whose logo requirements were met). These logos are meant to assure the end user that the software being purchased is indeed fully compatible with a certain version of Windows. In stark contrast, Windows XP Tablet PC Edition does not have its own set of logo requirements. It s quite likely that subsequent versions of Windows XP Tablet PC Edition will have logo requirements, particularly given the specialized nature of Tablet PC software and the many usability-related concerns it raises.

Could This Be the One?

Microsoft s introduction of the Tablet PC follows a long trail of failures by a plethora of companies (including Microsoft itself!) who tried to bring pen-based computing to the masses. Some of the failed attempts were fronted by large companies with good reputations (for instance, Apple s Newton), proving Microsoft s Tablet PC initiative very bold indeed. Is there reason to believe that this attempt will be any different from those in the past? Why would Microsoft continue to invest in something that it, and many other companies, had failed at before?

Although there are no guarantees, there are a few good reasons to believe that Microsoft s latest attempt at tablet computing may succeed where others have failed. There are three major factors that favor Microsoft s Tablet PC over its competitors prior attempts:

• The hardware landscape is vastly different

  Hardware is continuously getting faster, lighter, and more efficient. A huge determinant of the success or failure of a tablet device is the effect of its hardware profile on its usage scenarios. There was a time when portable computers were 20 pounds each, rendering them essentially useless for the majority of consumers portable scenarios. Similarly, when most users think of tablet computers, they envision certain key scenarios, such as note taking in a meeting, annotating documents, or jotting down thoughts. The current state of hardware technology makes these key scenarios more plausible because well-designed Tablet PCs should be about three-quarters-inch thick and weigh about 2.5 pounds. High dots-per-inch (dpi) displays are also on the horizon, further improving annotation scenarios. These and other technological advancements make this a more compelling time than ever to release a tablet device.

• Digital ink recognition technology is much improved

  Closely tied to hardware improvements is the availability of improved handwriting recognition algorithms. These new algorithms make it possible, for the first time in widespread commercial software, to recognize continuous (cursive) English instead of just segmented print characters. Similarly, increased processing power has now made it possible to recognize words faster and more accurately than ever before. Much of the frustration of early tablet computing attempts arose out of poor ink recognition by the underlying software. The situation is now much improved.

• Digital ink is treated as a native data type

  Ideal handwriting recognition, just like ideal voice recognition, is one of those technologies that has been just five years away for the last 20 years. The problem with this widely held expectation is that it is simply unrealistic. The fact of the matter is that sometimes even the very author of a document has trouble recognizing what he wrote how could a computer possibly be expected to do a better job? Similarly, in many conversations we have to ask the speaker to repeat himself surely a computer wouldn t be able to do better, especially without the contextual clues afforded by understanding the conversation s subject matter! We must recognize that there is, and perhaps always will be, a basal amount of error that will not be overcome by handwriting recognition algorithms. That said, the approach taken by Microsoft s Tablet PC differs from many that have gone before it because it acknowledges up-front that at least some amount of error is inevitable. Instead of putting all the effort into trying to eliminate the error, the Tablet PC treats digital ink as a native data type. By keeping ink as ink, as opposed to constantly converting ink into text as soon as it s entered, the human reader of a document is given the opportunity to recognize handwriting that the computer may otherwise have mistaken. Far from shirking responsibility, this philosophy of ink as a native data type, keeping ink as ink, supplements the technology s shortcomings with the human s strengths. The computer tries to be helpful by offering recognized text when asked, but it also preserves the author s original ink because it is equally important as a data type of its own.

There is no way to accurately predict the success of Microsoft s Tablet PC, though it has several notable advantages over those that have gone before it. Only time will bear out whether these advantages are enough to make Microsoft s Tablet PC the one to popularize tablet computing, or whether it will fade into history along with all the previous attempts. In the meantime, there are lots of exciting technologies made possible by the Tablet PC Platform, which we will in turn explore throughout the rest of this book.