Looking back at the last 25 years of innovation, it's tempting to think that we're close to the limits of what we can do with computers. That's largely because the progress has been so amazing. We've gone from standalone mainframes to hundreds of millions of incredibly powerful PCs and smart devices. The spirit of Moore's Law has taken processing power from kilohertz to gigahertz, storage from kilobytes to terabytes, and networking speed from mere bits per second to gigabits per second.
Computers have moved out of the IT department into almost every part of our lives. More than 600 million PCs are in use today, a number that will rise to more than a billion in the next five years. Many devices we use every day--from mobile phones to TVs--are becoming like computers, with processing power, storage, and connectivity that meets or even beats the high-end PCs of just a few years ago.
Yet we're only beginning to realize computing's potential. I believe that we're entering an era when software will fundamentally transform almost everything we do. The continued growth of processing power, storage, networking, and graphics is making it possible to create almost any device imaginable. But it's the magic of software that will connect these devices into a seamless whole, making them an indispensable part of our everyday lives.
In the workplace, we're already moving from personally focused software, such as word processors and spreadsheets, to truly collaborative tools that bring teams together and drive a quantum leap in business productivity. Today's productivity software does a good job helping people collaborate, with shared workspaces and management software that helps teams and projects work efficiently. But a coming generation of software will take collaboration a step further, capturing the knowledge and experience of an entire organization, enabling individuals and teams to draw on that information to make better, more strategic decisions.
In the back office, software standards are driving a more model-based approach to developing applications. With the growth of XML and Web services, we're getting closer to being able to visualize any kind of business process and quickly develop software that can adapt to companies' changing needs. For example, today when a firm makes an acquisition or changes a key business process, the IT department often must embark on the time-consuming and expensive task of rewriting and testing the underlying software. But as we move toward a world of rich Web services and development tools that instinctively understand business processes, businesses can simply make the changes they want and the code will take care of itself.
Although it will be some years before the idea of truly self-managing systems is realized, we're already seeing great progress. For distributed systems, management always has been an afterthought, applied after the servers and applications are in place. Going forward, management intelligence will be built in. The service and health modeling capabilities of the recently released Microsoft Operations Manager 2005, for example, already are helping customers significantly cut the costs of supporting existing systems.
We're taking a similar approach to reliability, incorporating best practices throughout the software life cycle, educating our engineers to write more reliable code, and creating innovative development tools and technologies to improve software quality. We're also implementing customer-feedback tools in our products that enable us and our partners to gather reliability data from real-world usage scenarios.
Computing also is extending further into the physical world, with emerging technologies such as RFID tags and a growing number of embedded devices and sensors. This means that software can go places it has never gone before--tracking inventory from the factory floor to the cash register and beyond to watching your home for intruders or keeping tabs on what's in your refrigerator.
Ensuring that all these systems are reliable and secure will be a high priority for many years. Windows XP Service Pack 2 is obviously a big step for us, and we're on track to distribute 100 million copies in the first two months after release. We've trained 500,000 IT professionals worldwide on security technology and best practices. And we're already seeing the benefits of automated testing tools that can verify code and help eliminate common security vulnerabilities, as well as services such as Windows Update that can quickly distribute security patches across vast networks.
For consumers, the biggest impact of tomorrow's technologies will be on entertainment. New graphics hardware, networks that can deliver high-definition video to any screen, and advanced software will redefine TV and video games, delivering a real-time generation of incredibly realistic scenes. TV is becoming more dynamic and interactive, with new kinds of content and advertising made possible by powerful software and rich connectivity. Gaming is rapidly becoming a social activity, as software creates vivid and engaging graphics for players and spectators alike and makes it possible to compete with friends around the world. And at work, these graphical advances are enabling richer visualization of data and innovative new interfaces that can help people become vastly more productive.
At the same time, the growing capacity and tumbling cost of storage is enabling unimaginably large databases so that it soon will be possible for people to store every piece of information they encounter. Gordon Bell, a distinguished Microsoft researcher, is already chronicling his life in a project called MyLifeBits--a realization of Vannevar Bush's 1945 Memex vision that can be found at "MyLifeBits Project: Microsoft Bay Area Research Center Media Presence Group."
Even today, most people track information in their lives with a system of files and folders--the same way they work with paper documents. But software is evolving new ways to acquire, organize, analyze, and understand digital information, with ap- proaches to storing data that go far beyond folder systems and advanced search tools that understand users' needs and help them find answers.
Another field poised for a wave of innovation is natural interfaces. We're making breakthroughs that take us close to the longtime vision of computers that can speak, listen, and learn. Accurate handwriting recognition is a reality with Tablet PCs, but we're continuing to make advances that make digital ink more natural, adaptive, and personalized. Simple speech recognition is commonplace in some specialized applications, such as customer service, but advances are rapidly taking us to a world where any device will be able to understand spoken commands.
As computers become increasingly "aware" of their users, they can adapt their behavior accordingly: for instance, sounding audible alerts rather than displaying dialog boxes when the user isn't looking at the screen. Meanwhile, inexpensive cameras and powerful software are spawning videoconferencing tools that offer users a greater sense of presence and more dynamic collaboration, as well as the ability to review and annotate meetings in real time.
Just as software has driven the past 25 years of innovation, it will be the key to enabling another quarter-century of breakthroughs--and in the process, transforming how people live, work, learn, and are entertained. My optimism for the future of computing has never been greater.
Bill Gates founded Microsoft nearly 30 years ago and is now chairman of its board and chief software architect.
Illustration by Dan Brown