A Berkeley radical with an engineering degree, Lee Felsenstein is notunfamiliar with pursuing altruistic goals in uncommon ways, writes Danny O'Brien.

I'm rushing through a San Jose street market with Lee Felsenstein, looking for a fast food place. I owe him a cheap lunch: years ago, I cornered him in a Pret à Manger in London, interrogated him for hours on Silicon Valley history, then stuck him with the bill.

He knows a great deal about the microcomputer revolution - mainly because he kick-started much of it. There are more expensive lunches owed to Felsenstein in this town.

They called him "the Robespierre of Computing". A Berkeley radical with an engineering degree, he spent the early Seventies looking for a way that he could "use his expertise to help the cause". A local non-profit scored a mainframe; he helped them put terminals in drop-ins across Berkeley by designing one of the first ultra-cheap modems.

Then he grew interested in the idea of building computers cheap enough so that you wouldn't need terminals. The dream was a People's Computer: to have your own machine, at home, that you could use any way you wanted. He and a few others formed a club to discuss the matter: the Homebrew Computer Club.

The first meeting showcased the Altair, the first ever microcomputer, out of New Mexico. Inspired, the amateurs and radicals of the Valley tried a hand at designing their own machines. That's when Felsenstein transformed into a Robespierre.

He led later meetings, finagling video engineers to talk with chip men: coaxing circuit layout experts with programmers; building a social network of contacts among the shyest people in the world.

Most famously, the Homebrew Computer Club encouraged Steve Wozniak to build his own computer. He and his friend, Steve Jobs, were so well received at the club that they quit their jobs, and started Apple Computer. Other amateur hackers from the club followed. Felsenstein himself was hired to design the first portable. And so began Silicon Valley's first personal computer boom.

Felsenstein has recently emerged, blinking, from deep within the last of the cyclical explosions of wealth that have followed ever since in the Valley. He spent the 1990s prototyping for Interval Research, a dotcom R & D lab that was notoriously more R than D.

We talk about the cruelty of the Valley's cycles. "They're getting faster, more intense." He doesn't look worried. In his mid-fifties, Felsenstein has the easy manner of the elder hacker engineer, with hair that won't comb. His face has the pale mottling of someone who shouldn't - and doesn't - spend much time in the sun. We find a table in the shade.

We talk about about how the latest boom attempted to make heroes and business leaders out of engineers - who'd rather just make stuff. "I think that hackers might work better when they're not being put on pedestals," he says, finishing off the fries. "We like something to work against; some private mission to pursue." Some new underground revolution.

The sun's moved around as we've talked: he's back out of the shade again.

The Sixties and early Seventies had a radical effect on how the first microcomputers were engineered, but saw some darker effects too. Lee Thorn was a bomb loader on an aircraft carrier in 1966, participating in the US bombing of Laos (between Thailand and Vietnam). Years later, he formed the Jhaos foundation with a Laotian refugee, Bounthanh Phommasathit, aimed at achieving reconciliation between all groups involved in the war. Last year, he approached Felsenstein with a challenge: could he spread the power of computer technology to the agricultural communities of Laos, as he'd done spreading it to the homes of America?

Felsenstein has just put the finishing touches to his first prototype machine for the project. It doesn't look much like the modern American PC. Powered by bicycle, with ruggedised insides usually found inside industrial factory computers, the Jhaos PC boasts a dot-matrix printer based on a 20-year old design, a screen bought from an ex-surplus reseller, and an aerial the size of a satellite dish hanging from a 20-inch coax lead. Its software is the free Linux operating system, converted into the local languages by volunteers and smooshed into a microprocessor too slow to run the latest Windows.

It looks like one of the old faithful: the machines built in garages by Wozniak and Co. There's a good reason for that: these computers have to be cheap, and repairable, and as open to tinkering and improving by the local community as the computer club's models. The Jhaos PC is tough, but it has to be flexible too.

And it has to be practical. By the end of the year, Felsenstein's Jhaos PCs will be shipped off to five Laos refugee villages, deep in the rice-growing hills of the region. Currently, the villages have no electricity, telephones or good roads between them. The PC's wireless link will connect the villages by WiFi to each other, and the telephone system.

Farmers will be able to monitor the price of crops in the town markets, negotiate group purchases with the other villages, and make business deals without having to spend days travelling away from the farm. And families will be able to make direct contact for the first time with the Laotian Diaspora - relatives who've left the war-torn area to earn money in the capital of the country, and beyond.

Cheap technology like this, dropped into the very poorest of countries, may provide a chance for these nations to leapfrog into the digital revolution.

Of course, there'll always be someone who'll argue that providing this kind of technology to the least developed countries of the world is missing the point: that we should, as Bill Gates said recently, be spending our money instead on medical and food projects. And, of course, everyone involved in the Jhaos project suggests we should do that too. But it's notable that it was the rural villagers themselves who asked for ways to communicate and gain knowledge, not the foundation.

And Felsenstein is not unfamiliar with pursuing altruistic goals in uncommon ways. When he was running around Berkeley in the Seventies, fellow activists must have thought he was mad to skip the protests to build modems, wire up mainframes for community use, and slot together cheap computers "to help the cause".

But "the propaganda of the deed", as Felsenstein calls his work, has proven to have a much greater effect on society than much of the agit-prop of the time. And, with one revolution already under his belt, who's to say he can't start another?