TORONTO -- On the first day of the 10th Computers, Freedom and Privacy Conference -- the unique annual meeting that brings together an unlikely combination of programmers, activists and government officials -- two very different events took place simultaneously.

One: About 30 participants and 50 observers crowded into a hotel meeting room for a workshop led by Lenny Foner -- computer guy in jeans and long hair, MIT Media Lab Ph.D. Foner was trying to get the group interested in starting up a new domain name system for the Internet. He was probably thinking Linux; he was most likely hoping for a Linus Torvalds sort of role. His idea was to maybe "route around" the current, dispute-prone system of matching Internet addresses to names. Maybe we should make a superset of the DNS, the workshop considered, or an alternative to it, or something -- no one could even agree on the precise nature of the problem, let alone its solution.

At any rate, this didn't stop Foner. He had a programmer's idea of how things get done in the world: Forget about the government; don't form a committee. Just write up a short proposal, give your idea a silly hacker-ish sort of name (even he admitted that the name he chose, "Smoosh," was somewhat unfortunate), talk about it to some very smart people, get a small group of them interested, then just start hacking out some code.

John Gilmore, a founder of the Electronic Frontier Foundation and self-described libertarian, was at the workshop, and with terrible succinctness he laid out the thinking behind Foner's vision of the programmer-created world. Gilmore was opposed to too many people getting involved in whatever Foner is going to do. "Almost everything that works on the Net grew out of tiny groups of people working in isolation," he said.

Meanwhile, as Foner was talking about "how to prototype something new," there was event No. 2: The Canadian Parliament was passing Bill C-6, a data protection act like the European Union's Data Directive -- leaving the United States as the sole highly industrialized nation without legal data-privacy protections.

Evidently, the process leading to the passage of the C-6 was nothing like the "tiny groups working in isolation" that John Gilmore had described just a few minutes before. According to Stephanie Perrin, who worked with the Canadian Department of Commerce and Industry for 20 years and who took part in the drafting of the bill, it had involved hundreds of people. It required concessions on all sides. The resulting law is not perfect. "It was a long and difficult process," she said, "where everyone fought."

These two events -- the programmers workshop and the passing of a federal data-privacy law -- are like the ends of a rope in a heatedly fought game of tug-of-war, a game that has been battled at CPF over the course of the conference's 10-year existence.

On one side are the geeks, nerds, crypto-anarchists, libertarians and cypherpunks -- mistrustful of government, suspicious of all attempts at regulation, believers in the ability of technology, in and of itself, to solve society's ills (maybe with a little marginally legal hacking on the side, just to keep the political pot boiling). Austin Hill, president of Zero-Knowledge, opened the conference like a true techno-believer, quoting John Gilmore as saying, "I want to guarantee [privacy] with physics and mathematics, not with laws."

Opposing the technologists are the believers in law above all else: the think-tank and activist lawyers; the privacy commissioners in their well-cut European suits; the pragmatists advocating commissions and studies and meetings -- participants in the rough-and-tumble of political life, with all its confusions and compromises and imperfect results.

In the past, the techno-believers ruled CFP. The programmers' vision of creation -- the lone geniuses -- prevailed over the data-privacy "bureaucrats" -- so hard to listen to, after all, with their thick foreign accents and their tedious, confusing laws.

But something different happened this year. The flag in the middle of the tug-of-war rope moved. Two well-known technologists, known for their belief in working code and skepticism about the workings of law, stepped across the divide, moving, maybe despite themselves, toward a recognition of social and political realities. Two others, whose views have been more balanced, questioned libertarianism -- the limitations of a technocentric approach to the complicated questions of privacy and freedom. It was as if some tipping point had been reached, in which a critical mass of people involved in technology had suddenly looked up and found themselves to be older, grown-up, and in need of social supports -- grown-up like the Net itself.

The first famous technologist over the line -- albeit tippy-toeing -- is Phil Zimmermann, creator of PGP (Pretty Good Privacy) encryption software, techno-hero, defier of the government when it tried to declare encryption a "weapon" and Zimmermann a felon for "exporting" it.

His moment comes during the discussion following the dinner speech on Wednesday night. Neal Stephenson, a writer with a cultlike following among the technologically minded and author of the classic "Snowcrash," has given an over-long, hugely digressive -- and brilliant -- speech. After many, many turns and a deep stack of points and stories, Stephenson gets around to saying that the best defense for one's privacy and personal integrity turns out to be not cryptography but, what do you know, "social structures." He is not explicit about the exact nature of these structures, but from the slides that follow, we get a sense of every sort of social relationship from neighborly friendliness to political parties. The slides show drawings of small circles representing areas of social trust. The circles widen and merge, to create a field of autonomy, a trusted space.

Stephenson is making a point about code: Without a sociopolitical context, cryptography is not going to protect you. He singles out PGP for criticism, saying that relying on the encryption scheme is like trying to protect your house with a fence consisting of a single, very tall picket. A slide shows the lone picket rising into the sky, a bird considering it with bulging eyes.

After the speech, Zimmermann puts up his hand, and of course Stephenson calls on him. It's clear Zimmermann has "gotten" the speech. He doesn't go so far as to endorse anything like "social structures," communities of trust, neighborhoods of understanding -- no, of course not. Zimmermann has been staunchly against laws, rules, regulations: anything that could be considered a form of social coercion. But he does admit that perhaps code is not enough, that he never intended encryption, by itself, to work. "I never meant PGP to be the defense of a lone libertarian," he says.

It is a huge admission, in its way, from a programmer who has championed code as a way to save us. But if this libertarian is not "lone," he is with some other libertarians, presumably. And what are these more-than-one libertarians doing? Organizing? Petitioning their government? Creating zones of social trust? Zimmermann is a man who defines the word "loner"; he has a tight manner; one doesn't imagine he's spent a lot of time working on his empathy or inner doubts. He probably doesn't even let himself realize the implications of what he's just said.

"Let the record show," Stephenson says carefully in reply, "I never said the word 'libertarian.'"

- - - - - - - - - - - - - - - - - - - - -

Tim Berners-Lee, creator of the World Wide Web and director of the World Wide Web Consortium, will indeed say the word "libertarian." He will say it on Thursday night, when he is the recipient of an EFF Pioneer Award, given every year "to honor significant contributions to the advancement of rights and responsibilities in the Information Society."

Berners-Lee can't be there, but he has sent a videotape with his thanks. He feels honored, is genuinely grateful.

And then he looks less happy. Berners-Lee starts thinking about what has happened to the Web since he dreamed it up: e-commerce, big corporations, money. "Libertarians are used to fighting the government," he says, "and not corporations ..."

This must be very difficult for him to say. For the libertarians in the audience to hear that business and free markets may not be the bringers of unalloyed good ... To imagine that a business is something to be fought, not respected ... No. Better to go off, leave the thought, don't say anything more.

But he can't somehow. Another unhappy thought comes: "I know we don't like regulation where we can avoid it, but ..."

And there he surely must stop. Bad enough to imagine fighting a corporation, but to do it with regulations? Regulations, meaning laws, meaning government? He has crossed into libertarian anathema.

Why has this techno-hero raised the specter of libertarianism? Theoretically, Berners-Lee personifies the "lone genius" technology ideal: While working as a consultant at CERN, he went off by himself, just for his own amusement, and coded up what we now call hypertext. Theoretically, he has every right to believe that somebody else will go off alone, just for his or her own amusement, and solve the problem of corporate control of the Web.

But it seems he has recognized a changed world, where neither he nor some other programmer can do it alone. "We have to make sure that when people go to the Internet, they get the Internet," he says, meaning the real Net, the true one, the original -- whatever that might mean to him, or us. Somehow, even if it means laws and rules and governments, we must find our way back to this idyll. We must route around the new bad corporate Net, or create a superset of it, or an alternative. Or something.

Berners-Lee was speaking unpopular truths to the CFP crowd, but his outspokenness is nothing compared to what is about to happen. The next Pioneer award is to go not to an old programmer or to a lawyer at a think tank, but to ... "librarians everywhere." Can we be hearing correctly? Did they say librarians and not libertarians? But it's true: librarians. It is an unprecedented award, the first to a group that can't be associated with at least a few specific individuals. And in the face of this -- this amazement, this recognition of the great unseen and unsung core of mostly women -- the fourth of our techno-heroes will find himself to be, in his startled heart, a lover of civil servants.

Whitfield Diffie bounds to the platform. Diffie is a crypto-king, the discoverer, with Martin Hellman, of public-key encryption, cornerstone of the libertarian worldview in which technology protects the individual from the reach of goverment. He stands now before the audience with his neat gray beard, shoulder-length blond hair and sudden uncontained enthusiasm. "Librarians!" he exclaims. "I'm thrilled with this award."

He goes on to say he was not involved in the judging; this is the first moment he has learned of it. And now that he thinks of it, those wonderful librarians of his childhood, the ones who helped him when he was working on his dissertation -- yes! Librarians!

"I wouldn't have thought to give this award," he declaims in the full throes of the convert's confession. "Therefore it comes as a revelation."

All those invisible, dedicated civil servants. Mostly working for government. In public libraries. Paid for by taxes. Diffie stands there with arms out. He is truly, naively, nakedly, unabashedly amazed to consider it. The whole libertarian edifice crumbles as he looks at it. Revelation.

- - - - - - - - - - - - - - - - - - - - -

But this is more than a startled, unguarded moment. The next day, in a speech he gives at lunch, Diffie reveals the depth of his conversion.

"Everyone stop eating," he begins, and indeed we should, for what we are about to hear is akin to the more common story of a middle-aged person, communist in his youth, who grows more conservative as he grows older, renounces his youthful beliefs -- except we will hear it in reverse, right to left.

He signals it all right away. "Crypto was a security technique that didn't require trusting anyone else," Diffie says. "Now it turns out you have to trust other people." He was younger, he seems to say, he had ideas, he was wrong. "I had a very mathematical and very inapplicable idea about authentication." And there it is: an implicit rejection of the Gilmore-ian ideal of trust in physics and mathematics. Like Stephenson, like the reluctant Zimmermann, like the unhappy Berners-Lee, the father of public key encryption has come to the conclusion that software may reduce the amount of trust you need in human beings, but as one moves about in the world, the sense of security, privacy and autonomy turns out to be "a function of social structures," as Diffie says.

So far, Diffie has gone from being a techno-libertarian to a standard-issue social democrat -- a remarkable move, if not a remarkable place to wind up. But he is not done.

What has sparked his conversion, it seems, is the recentralization of computing: how we have moved from the centrally controlled timesharing system, to the autonomous powerful desktop PC, to the networked computer, and thence -- sidetracked through the network computer and the "thin client" -- somehow back to the dumb terminal. He foresees how knowledge workers will lose their autonomy by being forced to use such slavish machines; how they can become mere objects of surveillance by the companies they work for, as a result of "corporate imperialism over its workers."

Is there something wrong with the microphone? Is he talking about imperialism? Yes, and on he goes, ever leftward. He can foresee a day when workers, doing their jobs from the "convenience" of their homes, are forced to be subject to "spot inspections" by their employers, a time when the home is effectively turned into an occupied zone where corporations exercise power over their property.

What shall we desperate knowledge workers do? Organize! We need "the rise of labor again," says Diffie, former crypto-believer. "We need to tighten up the relationships among knowledge workers," he says, "and bargain as a whole."

I can't believe what I am hearing.

- - - - - - - - - - - - - - - - - - - - -

The conference ends with a session on "diversity" -- and again, one is startled to find this at CFP, former home ground of crypto-anarchists and techno-libertarians. On the podium is Greg Bishop of TheStreet.com, one of only two African-Americans on the entire conference program, perhaps the only one attending CFP. He is telling us how the black people he knows are amazed he uses a computer -- they believe that once you plug it in, the government knows everything about you. The audience goes on to question the homogeneity of the conference itself -- why there are so few young people, blacks and women in attendance, and indeed in the leadership of the technological world. The culture wars have come to CFP.

And why not? The Internet, with its vast public acceptance, letting people who have never even seen a piece of code do everything from buy a car to search for lovers, can hardly be considered a purely technological system anymore. The Net has become a social space, and it is perhaps right that the practices of programmers -- the small group in isolation -- no longer pertain. We've come to the messy part that very senior programmers get to avoid: the part where the system has moved beyond the "new" and "dreamed-up" stage. Where it is successful -- that is, it has users, millions of them, with all their conflicting needs and desires, and only the messy, horrid, compromised, wonderful, exhausting processes of democratic social discourse can sort them all out.

After the conference is all over, it's fun to sit with Bruce Umbaugh, philosopher and member of the CFP organizing committee, and imagine the sort of happy chaos that can happen at the event next year. We'll invite the UAW! We'll invite the Boeing engineers, knowledge workers who have organized themselves for the first time -- and won! There'll be online dykes and gangsta Napster rappers. There'll be kids and students and mothers and just about anything else you can think of. And why not? When we said the Internet represented a "revolution," we meant it -- didn't we?