Investment in research is crucial to Microsoft's survival and continues to absorb about 17 per cent ofits total revenue, according to company founder and chairman Mr Bill Gates.

"Research has more than paid off for us and we're continuing to increase that investment even as the economy goes through its ups and downs," he said. The company has one of the largest corporate research budgets, at nearly $5 billion annually.

Mr Gates spoke at the Future Forum, an event at company headquarters in Redmond, Washington, marking the tenth anniversary of Microsoft Research (MSR), formed in September 1991. He said all technology companies should make a greater commitment to research and development because the payback is tangible. "The really big advancements come from long-term research investments," said Mr Gates.

The personal computer remains central to Microsoft's vision of the future. However, people will increasingly prefer their PC in a small package, such as the Pocket PC or the portable "tablet" PC, due out next year. All of these will interact more intelligently with users, he said. Computer software "agents" will work for people, sorting their e-mail according to its importance, forwarding only the most urgent mobile phone calls to a meeting, and organising work and home materials.

However, Mr Gates admitted an early attempt at agent software, the "Clippie" animated paperclip in Office applications, had not been one of Microsoft's finer moments.

"The paperclip in Office was actually fairly controversial," he said, to some laughter. "One of the most exciting things we did in Office XP was turn it off by default."

The company demonstrated a 2.5lb tablet PC that users can write directly on with an electromagnetic pen, and which has a screen resolution approaching magazine-quality. Microsoft also has a Pocket PC that turns itself on when picked up, and will scroll up and down through information when tilted.

In contrast to many major research labs, much of MSR's work ends up in Microsoft products, said Mr Gates. "We make sure the boundary (between researchers and product developers) is a very soft boundary. Research workers really want to get that work into our products."

But research does not move quickly into consumer offerings. "The problems we're tackling are not problems that lend themselves to overnight success. Most of the goals we've set out on will probably take most of the next decade to resolve themselves," he said.

A significant proportion of Microsoft's research energies are going into the area of distributed computing, which uses the capabilities of many computers linked across a network to tackle problems or provide new services. That notion is the basis of Microsoft's new - and controversial -- broad-ranging computing strategy, called .NET.

"(Distributed computing) is leading essentially to a new platform," said Mr Gates. "All the different elements of software will be changed by this approach."

But critics worry that Microsoft is attempting to ringfence the Internet, presenting Microsoft's products and services as a persons main link into the Web. Some fear this would enable Microsoft to dominate in the internet arena just as it does in the personal and business computing worlds. As Microsoft faces renewed scrutiny in the American courts and by European regulators, the company is acutely aware that its future lies in cultivating markets beyond its strongholds of operating systems and software.

But Mr Gates, who did not take questions from journalists and analysts, described Microsoft as "an optimistic company, creating tools that would really have no boundaries in terms of what was possible".

A major shift in his predictions computing involves the centrality of high-bandwidth internet connections into the home. Since the launch of Windows 95 Mr Gates has offered a vision of homes permeated with computing devices networked on a broadband internet link. But he says the costs of installing the networks and other issues will delay broadband rollout and stymie some computing advancements.

"The thing I'd really be cautious about is high speed connections out to people's homes," he said. "There's no miracle advance that's making that very inexpensive, and then you have regulatory issues that come into that."