The form factor of personal computers hasn't changed much in the past 20 years--monitor, mouse, and keyboard, with a box where the magic happens. But makers of PCs and computer chips are betting hundreds of millions of dollars in research, development, and marketing that businesspeople are ready for a change not only in the look of their machines, but in how and where they work.

A glimpse of that future comes from an Intel-built prototype that crams together the functions of a notebook computer, a tablet PC, a PDA, and a cell phone. Dubbed Newport, Intel's device has a screen that can be rotated to lie flat like a slate or tablet PC or swivel into a notebook-computer form with a keyboard. There's a secondary display on the outside cover that's slightly smaller than a PDA screen, so users can access calendars, phone numbers, and E-mail without firing up the entire computer. It supports five radio technologies for connecting to company networks or the Internet from just about anyplace. Newport even has a touch of style: Its detachable screen and keyboard are strapped together with a flap of brown leather that clips just below the secondary display.

All in all, it would be a pretty cool machine to have, if it ever moves out of the concept stage. But cool isn't selling so hot these days, when every technology purchase is under scrutiny. Vendors may be spending a fortune on new ideas like Newport, but they'll need to overcome reluctant and skeptical customers and tight IT budgets if they hope to launch a second revolution in personal computing. Today, it's all about "building the business case and providing a return-on-investment time line," says Vince Morrotti, VP and chief technology officer for automaker DaimlerChrysler AG. "There's got to be a real value-add and hard-dollar savings before we bring any technology into the company."

That kind of intensely practical approach makes it difficult for PC vendors to stir up excitement for new products. With more than 250 million PCs being used by businesses worldwide, according to Microsoft, most workers who need a PC, at least in developed economies, already have one. That means sales of PCs have evolved from a major technology growth market into more of a replacement market.

Businesses still spend a lot of money on PCs. Next to salaries, IT managers will spend the greatest percentage of their budgets on hardware purchases, and PCs are at the top of the procurement list, driven by the desire to upgrade desktops to Windows XP, improve productivity, and optimize business processes. InformationWeek Research's Priorities second-quarter survey of 300 IT and business executives finds that 87% put new PCs on their planned-projects list, and 60% say Windows XP is on their priority list, up seven percentage points from the first quarter.

Researchers predict that PC sales will grow this year: IDC expects worldwide PC shipments to rise 7% to 145.6 million units this year, while Gartner sees unit sales increasing by 8% this year and returning to double-digit rates next year, possibly jumping as high as 12%, depending on the economy. Sales have been relatively modest during the past couple of years.

The bigger question isn't about PC sales; it's whether PC technologies will really let people change how they work. Intel is betting that notebook PCs with built-in wireless communications, combined with the proliferation of high-speed wireless data services such as

Wi-Fi and enhanced cellular data services, will transform the way people use computers in the same way LANs and the Internet changed the use of PCs in the 1990s (see story, "Wireless Access On A Roll"). Microsoft has introduced a new desktop operating system, Windows XP, and will soon roll out a new version of its Office productivity suite (see story, "Office Moves Beyond The PC"). The company also believes its Tablet PC software will expand the places and tasks where the PC is relevant. And Sun Microsystems, ever the believer in the network, thinks tight IT budgets may finally generate more interest in its network computer or thin-client approach, in which the operating system, applications, and computing intelligence are contained in the server, not a desktop machine.