KEY POINTS:
Facts are stubborn things. Get them wrong and you can end up in all sorts of trouble. I have an appointment to meet Jimmy Wales, the founder of Wikipedia, the internet's fount of all knowledge, on East 59th St, New York, but the address I have been given seems to be a branch of the Mexican Embassy dealing in visas and asylum. Families are queuing out on the street.
With a sinking feeling I argue my way inside, the only gringo in sight, and jump a queue.
"Is this, um, where Wikipedia's offices are?" I ask, brightly. No face has ever looked more blank than that of the Mexican immigration officer. I show him my piece of paper with the scrawled fact of the address; he turns down his mouth.
"Wi-ki-pe-di-a?" I say slowly."Ji-m-my Wa-le-s."
The man looks at me for a second and makes a curt slicing motion with his hand in front of my face before turning to more pressing demands on his attention.
Facts can be treacherous. I know Wales is in town for only a few hours.
So I run round Manhattan's streets in search of an internet phone box to access my email to check the address and discover I've written down the address right but the phone number might be missing a nought, call Wales, and explain all this in a rush.
"Ah yeah," he says slowly, "it's WEST 59th. I must remember to make that clear to my PR."
I hurry a few blocks, sweating, and find myself at the door of an office so decrepit that it looks hardly more likely a hub of wisdom than the Mexican embassy. Wales is inside, though, sitting beside a knackered desk, his bearded face illuminated by the screen of his Mac. He is surrounded by four studenty young men with novelty items stuck to their computers. No one looks up.
"Sorry I'm late," I say.
Wales, who has something of the air of that desperate novelist in the film Sideways, grins, takes a couple of calls on two mobile phones he has wedged under his chin, writes some energetic emails, then sits me down and explains, quietly, his personal mission to "bring the sum of human knowledge to every single person on the planet, free, in their own language".
Jimmy Wales is the world's first one-man open university. Or at least he has set in motion a movement of volunteers who believe that by pooling knowledge, refining constantly, checking and rechecking, they can produce an encyclopaedia that might become as authoritative as anything produced by academic experts.
A decade ago this would have sounded impossible; now it just sounds absurd. Still, it is hard to argue that it is not working.
The Wiki community has been nailing facts for seven years now, since Wales and his then technical partner Larry Sanger had the idea of starting an online encyclopaedia called Nupedia.
To begin with they went about it in the traditional way, soliciting articles from experts. By the end of the first year they had collected precisely 22. It was at that point that they came across wikis. A wiki is, to give its Wikipedia definition, "a website that allows visitors to add, remove, and edit content ... a collaborative technology". Wikis would be used, Wales decided, to create the first people's encyclopaedia.
He says he knew after two weeks it would work. By that time they already had many more articles online than they had managed in two years with Nupedia. Another big moment came at 9/11, later the same year, when Wikipedia volunteers "just began jumping in and - writing background articles - from nowhere, it seemed. Very quickly, we had articles on the World Trade Centre, the airlines involved, the terrorist groups mentioned in the TV news that day. It was very compelling".
In many ways, the project was a test of Wales' faith. Having left his job trading options on the Chicago money markets in 1998 he had followed the virtual gold rush to California to start an internet portal. He rather coyly describes this as "a male-oriented site, a little like Maxim magazine", with a small proportion of its revenue coming from soft-core pornography.
Once involved in the internet, however, Wales became interested in the open source movement, which believes in the provision of free software for all. He had, too, read political philosopher Friedrich von Hayek's arguments that all knowledge is partial and the closest you can get to truth comes from the aggregation of as many partial understandings as possible. Wales wondered if these two principles might somehow be combined: the ultimate open source book of reference.
In some ways he was born to this work. Wales grew up in Huntsville, Alabama, where his mother and grandmother ran a one-room schoolhouse, which he attended from the age of three. Huntsville was not an average southern town. It was rocket city. After the war German rocket scientists, led by Wernher von Braun, inventor of the V-2, were lured to Huntsville to lead the American space programme.
This, Wales recalls, gave the town a very unusual culture.
"There was a huge hiring boom and suddenly this little place was full of the families of all these rocket scientists from around the world. It was a place where that American myth of anyone being able to be President, you know, if you just did your math homework, was very strong."
Wales did all his homework. He remembers the magical arrival of one specific aid to learning. One evening after supper there was a knock on the door from an eager salesman. Wales' parents were persuaded and bought young Jimmy his first encyclopaedia, the World Book.
He pored over it as a child, and has never forgotten the thrill.
Thirty-five years on, with his own 6-year-old daughter in mind, you could say Wales has upped the stakes somewhat.
The Wiki project has worked so spectacularly, it seems to me, because it feels like it is exactly what the internet was created for - a community where an aggregate mass of individuals in the virtual world can combine to produce something greater than the sum of its parts.
Just as the internet is not located anywhere, so the Wikipedia Foundation has no real home beyond a rundown office in Florida and single-room outposts in California and Poland. Wales himself has a home in Florida, but he has been there only one day in the past two months; he travels the globe, often with his wife and daughter, spreading the Wiki word. As we talk he spends a lot of time rubbing his eyes.
The New York office where I meet Wales is home to Wikia Inc, a discrete company run on more commercial lines, developing new projects, notably a revolutionary "people's" search engine to rival Google and be much more transparent in the ways it prioritises sites.
"Essentially, if you consider one of the basic tasks of a search engine, it is to make a decision: 'This page is good, this page sucks'," Wales says.
"Computers are notoriously bad at making such judgments, but we have a great method for doing that ourselves. We just look at the page. It usually only takes a second to figure out if the page is good, so the key is building a community of trust that can do that."
I'm interested to know if he views his work as altruism, but he flinches at the thought.
"I'm uncomfortable with that word because it implies self-sacrifice," he says. "I want people to be involved because they enjoy it and get some value out of it. The word I'd use is benevolence."
Why does he think so many people - several hundred thousand have contributed to Wiki pages - want to give up their time to spread knowledge free?
"Well," he says, "I think the big-picture goal is pretty exciting and that helps them through the dark days of grunt work. But that is not enough. If you tell someone you write encyclopaedia articles for a hobby, you may not be invited to too many parties; but if you explain the social side, how fun it is to interact with someone about something you care about, either to build something, or to argue, whatever, then people can see the value of that."
Wales' genius lay in the incorporation of discussion platforms and talk forums for each page. The magic of the site lies not only in the pages themselves but in some of the discussion behind them: the reasoned argument for this phrase or that word, each step preserved in a comprehensive page history.
"The ideal Wikipedian in my mind is someone who is really smart and really kind," Wales says, without irony. "Those are the people who are drawn into the centre of the group. When people get power in these communities, it is not through shouting loudest, it is through diplomacy and conflict resolution."
If this sounds like a kind of Utopia ("Wikitopia" is the word Wikipedians prefer), there are also plenty of Lord of the Flies moments. The benevolent Wiki community is plagued, from time to time, by "Wikitrolls" - vandals who insert slander and nonsense into pages.
A policing system has grown up to root out troll elements; there are now 1000 official "admins", working round the clock, who arbitrate and block unruly users from the site if need be; they are supported in this work by the eyes and ears of the moral majority of "good" Wikipedians.
At first Wales feared that any contentious entries would swing wildly from one point of view to another. In fact, he suggests, in some of the best cases a paragraph about a controversial point in history will be negotiated by divergent viewpoints until every last comma has been argued over, and it becomes a stable thing.
"You could have hired an expert to write an article," he says, "but you could never have had the number of interested parties engaged in this as we have here. The danger is that you get this very choppy writing style because everyone is fighting over every word and the big picture of the paragraph is lost. Someone brave then has to go in and smooth out the sentences a bit."
Wikipedia moves incredibly swiftly, and, because of that fact, and the way it is produced, it is criticised for being unreliable. But the scientific journal Nature tested the comparative accuracy of Wikipedia in 42 randomly selected articles: there were 162 mistakes in Wikipedia versus 123 in Britannica. In a 20-page rebuttal, Encyclopaedia Britannica Inc characterised Nature's study as flawed and misleading and called for a "prompt" retraction. Nature refused.
I wonder if Wales believes in absolute truths. "Writing things that are true is possible, but it is an arduous task requiring great mental focus," he says.
"It's what we aim for, but when people say, 'Should we trust Wikipedia?' my response is, well, we are incredibly upfront about where you should not trust it. There are lots of flags that we as a community put up to say: 'Tread carefully here. Warning, this section does not cite its sources'."
Microsoft Corporation tried to pay an insider to edit some of the sites about their software that they believed had been got at by IBM interlopers. Paul Dacre, editor of the the Daily Mail, apparently has misgivings about his own entry, which suggests a degree of megalomania, and has, according to satirical magazine Private Eye, "tasked a team from the Mail's much-respected library with drafting something more impressive to put in its place".
A recent polemic on the Edge website by the essayist Jaron Lanier caused a storm among the "virtual elite". Lanier likened Wikipedia's collectivism to "Digital Maoism" not a million miles from the worst totalitarian excesses of communism and fascism. (Wikipedians would no doubt be reciting one of their favourite maxims, Godwin's law, which states that: "As an online discussion grows longer, the probability of a comparison involving Nazis or Hitler approaches one.")
Lanier's objection was partly on style but also that "some Wikitopians explicitly hope to see education subsumed by Wikis". This, he argued, could lead to a sudden dangerous empowering of the "hive mind".
Wales feels the attack is a little hysterical. "We don't have any particular faith in collectives or collectivism as a mode of writing," he says. "Authoring at Wikipedia, as everywhere, is done by individuals exercising the judgment of their own minds."
Since I saw Wales I have become a Wikivirgin. For all the futuristic paranoia about hive minds, I have been struck by a kind of village fete atmosphere within the Wikicommunity; I am forever being prompted to clean up pages, articles to "Wikify", tasks to do.
On my first encounter with the site I was invited to join pages seeking contributors, which included "Fungi, Buses, Veterinary Medicine, European Microstates and a Jesus work group".
A good clue to the nature of the site lies on the page "lamest edit wars".
Here there is a history of factual skirmishes from the mists of Wikitime: the battle over Freddie Mercury's ancestry (Iranian), the ongoing dispute over whether apple pie should be considered American. Most disputes on Wikipedia are concerned with POV (point of view); subjectiveness is outlawed. You can vandalise pages, but once you have had some interaction on the site it seems absurd. It would be like turning over a charity cake stall.
A worse crime is to attempt to edit a page about yourself (even Wales at one point succumbed to this temptation over a detail of his fall-out with original collaborator Larry Sanger, who wants to establish a rival, peer-reviewed alternative).
I committed the cardinal sin of creating a modest page about my own towering contribution to the culture. Within five minutes my entry had been picked up by someone called Lisa Small, an attorney at the United States Supreme Court. She queried whether searches for "Timothy" should be routed to my page, or whether that would create confusion with another Timothy Adams, Under-Secretary in the State Finance department. Within the hour my site had been removed by an admin named Wafulz who assumed writing about myself was a newcomer's error. I refrained from repeating the comment of another troll who had described Wafulz as "a big four-eyed moron who wears the same stupid sweater every day ... "
I asked Wafulz, a graduate student in Toronto why he wanted to be a Wiki policeman. He said it was a kind of addiction to disseminate knowledge, not far removed from a journalistic impulse.
When I asked what he got out of it, he replied as all good Wikipedians do that, "I may only be providing a small amount, but I know that every time I edit an article, I've made a difference in someone else's learning experience."
The more time I spent on the site the more I came to think of Wales as some kind of Queen Ant, letting the vast colony go about its work, at the centre of a system where the knowledge of the community is infinitely larger than the sum of experience of all its individuals.
Wales is not sure how big Wikipedia can become, and he has long since ceased to have any say in that fact.
He recently, in what sounded like a slightly controlling way, suggested the need for a blogger's code of conduct across the web.
Before I left his office, I asked him if he lay awake worrying that he had created a monster, and he laughed a bit wildly. If so it is quite a benign one, he suggested - a big friendly giant.
I returned to the site, in search of some sense of scale. The largest-known ant colony, according to Wikipedia, was on the Ishikari coast of Hokkaido, Japan. The colony was estimated to comprise 306 million worker ants and 1 million queen ants living in 45,000 interconnected nests over an area of 2.7km. In 2002, however, a super-colony of connected nests was found to stretch nearly 6000km across Europe. Another, measuring approximately 100km wide, was found beneath Melbourne in 2004.
Wikipedia, I can't help feeling, is only a beginning. The problem with facts is, as someone once said, there are so many of them.
