To appreciate the dizzying life cycle of the technology business, consider the case of Marc Andreessen--the man who invented the Internet browser. He has gone from nothing to screaming success, as the co-founder of Netscape, and then to near-disaster when his company was mugged by Microsoft. He has a net worth of $200 million, but he's seen a fortune many times that float away in the vortex of the marketplace. And he's still just 28!

Now Andreessen wants to start all over again. He announced yesterday that he's forming a new start-up called Loudcloud Inc., which will provide advanced software and services to Internet companies. The "cloud" part of the name comes from a telecommunications term for a smart network, he explains, and it's loud "because, if you have a cloud, you don't want it to be wimpy."

Andreessen is the original wonder boy of the technology world. He's tall and baby-faced, and he speaks at such a dauntingly fast rate that you suspect his brain must be running on fiberoptic cable, while yours is mere copper wire. But he has relaxed some in recent years, and as he talked about his plans over lunch here last week, he seemed like the happiest man in Silicon Valley.

Andreessen says his new company will try to gather the best brains in the industry to provide sophisticated services to a market that's increasing by 500 new companies a month. Right now these companies can't afford to develop all the technology they need to compete--there aren't enough smart software engineers to go around, for one thing.

And he is convinced that the Internet revolution is just beginning. The market is at least doubling annually, and at that rate it will grow a thousand-fold over the next 10 years. "That means that 99.9 percent of what will eventually exist hasn't been deployed yet," Andreessen says. "It's hard to conceive of the magnitude."

Loudcloud will jump into this space. It will build large-scale data centers to host operations for smaller companies. It will develop software to help companies operate their sites at higher scale, using the most advanced data-mining and e-commerce technology. And perhaps most important, it will leverage the skills of the very best software engineers, making their scarce talents available to a universe of companies.

Andreessen is cagey about precisely what products his new company will sell. The details seem less important to him at the moment than the pleasure of being back on the high-tech roller coaster.

He explains the excitement of starting a new company the way a composer might describe creating a piece of music. "If you're into technology and business--and I love both--a start-up is an opportunity to draw on a blank sheet of paper the way you think the world should work," he says. "Plus, you don't have to work for someone, and you don't have to follow rules."

Andreessen certainly wasn't following rules when he invented the Internet browser. He was a 21-year-old undergraduate at the University of Illinois at Champaign-Urbana, working on a federally funded computer project. He was bored with his academic work, and he had this idea . . . a vision of an easier way to find information on the infant World Wide Web, which back then had just a few dozen sites.

He wrote code obsessively, working 24 hours and sleeping 12, from Christmas 1992 to March 1993. When he was done, Andreessen had created a program called "Mosaic," which was an easy way to surf the Web from a personal computer. He posted the first "beta" version of his browser in late March. The first day, a dozen people downloaded it; the second day, 100; the day after that, 1,000. By the end of the year, more than a million people were using his technology.

Andreessen was recruited in early 1994 to join a new company that became Netscape, but its first business plan was to write software for what people thought would be the next hot technology--interactive television. They abandoned that idea, and by September 1994 Andreessen and some of his college friends had created the Netscape browser--which was really just an improved version of Mosaic. And the rest, as they say, is history. "You had to be out of the mainstream," he explains. "If you were in the mainstream, the belief system was that the Web wasn't serious." The smart money back then was betting on interactive TV."

Andreessen has lived several lifetimes since then, in Internet time. He's seen his baby, Netscape, become one of the hottest IPOs in history; then watched it get battered by Microsoft; then watched it get swallowed up by America Online; then struggled to find a role for himself as AOL's chief technology officer, and finally given that up (AOL "isn't a technology company," he says) and returned to Silicon Valley. And what about the money? That's what most of us think about, when we hear stories about 28-year-old wonder boys founding new software companies.

"That question makes me deeply uncomfortable," says Andreessen. "The answer is that money is not very important. Nobody wants to hear that. But the excitement, on a day-to-day basis, has nothing to do with the money."

I'm not entirely sure I believe him, but I know what he means.