Wiki's world

On the day that Wales made his plea, at a symposium of ‘Wikipedians’ at the Harvard Law School, a further 4,000 articles were added to the site’s burgeoning archive, bringing the total number of entries – from Arab language to transvestism, from Georg Jan (1791-1866), Italian taxonomist of snakes, to the precise make-up of the United States Senate Agriculture Subcommittee on Production and Price Competitiveness – to more than 1.3 million.

(In the constantly shifting sands of information that is Wikipedia, about half of those new entries were deleted within 24 hours, decreed by the Wikipedia community as trivial or irrelevant.)

As I write this, two months after Wales’s address, the number of entries included on Wikipedia – registered on the site’s home page, beside its distinctive logo of a jigsaw of the globe – stands at 1,440,433… 4, 5, 6.

The contents list increases with a remorseless and unstoppable logic – a countdown to Jimmy Wales’s vision of a world in which ‘every single person on the planet is given free access to the sum of all human knowledge’. That day is not yet nigh; but even now Wikipedia is the internet phenomenon that nobody, least of all its creators, can quite get to grips with.

Launched in 2001, it is the 13th most visited website in the world (ninth in Britain and the US, where it ranks above CNN, AOL and the New York Times). Eighteen thousand people request information on Wikipedia each second.

Based on a free software program that anyone can use, Wikipedia has spawned identical sites in more than 200 languages, including Basque, Kurdish, West Flemish and Mongolian. (The Cornish-language Wikipedia boasts 1,200 entries, the Manx 148.)

In line with Wales’s vision of a universe of free information, there is also a dictionary (or Wiktionary), a directory of species and a free library of textbooks and manuals, all published online under the umbrella of the non-profit Wikimedia Foundation.

What makes this more extraordinary still is that Wikipedia is not simply an encyclopaedia: it is an exercise in the democratisation of knowledge. Anybody is free to write an entry on the site, or to alter one that is already there.

Each entry has forks to a ‘history’ page, where previous versions are stored, and a ‘discussion’ page, where contributors can haggle over changes and improvements. The site is governed by three cardinal principles: entries should not include any original research; facts must be verifiable via another published and reputable source (blogs not permitted); and all entries should conform to a neutral point of view (NPOV).

Its ease of access, and the sheer number of contributors, means that key entries can be updated with bewildering speed, incorporating events as they unfold. Wikipedia had added an article on the North Korean nuclear test within one and a half hours of the explosion taking place.

Wikipedia’s offices are located in St Petersburg, Florida, a town that seems the antithesis of the cyber age. It is popular with tourists and retirees – according to Wikipedia, 17.4 per cent of the population is over 65. Wales, 41, who is married with a five-year-old daughter, says he moved here because property was more affordable than California. Even within this necropolis, Wikipedia makes determined attempts to keep itself hidden.

My meeting with Wales had been arranged through the Wikipedia press officer, who lives in Vancouver. He directed me to an address that turned out to be a box number in a branch of UPS. Wikipedia’s offices are half a mile away, two rooms in a small office building, wedged between an attorney and an educational trust.

The website has only five fulltime employees. The walls are decorated with an art poster for La Cupola di Brunelleschi in Florence (no dedicated Wikipedia entry, but the entry page referred me to an article on Italian mannerist architecture) and a placard from the Harvard Law School symposium, Wikimania 2006.

Jimmy Wales – whom Time magazine last year named as one of the world’s 100 most influential people – is a short, bearded, mild-mannered man, dressed in the standard geek uniform of jeans, T-shirt and deck shoes. Wales was born in Huntsville, Alabama, the son of a grocery-store manager.

According to his own Wikipedia entry, ‘his mother, Doris, and grandmother, Erma, ran a small private school, in the tradition of the oneroom schoolhouse, where Wales was educated.’ He went on to study finance at nearby Auburn University.

An early enthusiast of web culture, Wales was a frequent contributor to the philosophical discussion groups that proliferated in the 1980s, and initiated and moderated a discussion list on the subject of the author Ayn Rand.

Rand was a titanic presence in American letters in the 1940s and 1950s whose books The Fountainhead and Atlas Shrugged espoused a philosophy she called objectivism – emphasising reason, a Nietzschean belief in the inviolable rights of the individual and a laissez-faire capitalism.

Wales says that Rand’s philosophy has partly informed the principles underpinning Wikipedia: ‘a deep respect for the individual; the ideas of freedom of speech and freedom of conscience. And the idea that decentralisation works better than centralised control.’

Wales experimented with laissez-faire capitalism first, working for six years as an options trader in Chicago. In 1996, with two partners, he set up an internet search portal called Bomis, mostly hosting sites offering ‘guy stuff’ – computers, sports, cars and girls. (Wales would later face allegations that Bomis hosted pornography; he insists it was nothing more than ‘glamour’ sites.)

As a child he had been an enthusiastic reader of encyclopaedias, and in 1999 had the idea to set up a ‘volunteer-built’ online encyclopaedia, called Nupedia, that would be free to use. To edit the project, he hired an old friend, Larry Sanger, who had been a participant in Wales’s online Ayn Rand forum.

Nupedia was conceived along fairly conventional lines: contributions would come from scholars and experts and be subject to a seven-step review process (a process similar to that employed by the bastion of encyclopaedias, Encylopaedia Britannica). But the arrival of a computer software device called a wiki changed that.

Invented by a software designer named Ward Cunningham, a wiki is a type of website that facilitates collaborative participation, allowing multiple users to create and edit content.

It can keep track of all changes, storing different versions and allowing users to revert back to earlier states. When Wales was first introduced to the wiki he was immediately smitten with the potential of the device for democratising the accumulation and spread of knowledge, and with Sanger introduced the wiki concept to Nupedia.

The idea met with resistance from some Nupedia contributors, wary of making the pages open to interference by unqualified outsiders, and in 2001 Wales established Wikipedia on its own site on the principle that it should be free not only to use, but also free in the sense of licensing – that is, anybody in other countries could set up their own versions of Wikipedia.

Sanger famously sent a blanket e-mail to all his Nupedia contacts: ‘Humour me. Go there and add a little article. It will take all of five or 10 minutes.’ Within a month there were 600 contributions; after a year, 20,000. In 2002 Sanger quit, in a flurry of disagreement about who had actually founded the project and how it should develop.

He subsequently wrote an essay, Why Wikipedia Must Jettison its Anti-Elitism, in which he claimed that Wales’s ‘radical openness’ was actually a form of ‘anti-elitism’, and argued that Wikipedia’s lack of a ‘publicly credible review process’ means that the site is regarded as untrustworthy among librarians, teachers and academics – precisely the people who most want to use a credible encyclopaedia – even when it is factually accurate. Wales responded that Sangers’s comments ‘betray a complete ignorance of the project’.

Wales refers to Wikipedia variously as ‘a mission’, ‘a movement’ and ‘a community’. The site carries no advertising and is supported entirely by voluntary donations from contributors and users, the majority of them $50 or less.

Wales himself draws no salary from the project. As well as being chairman of the Wikimedia Foundation, he also heads a for-profit company, Wikia.com, that hosts and manages websites on a variety of topics from Star Trek to pet diabetes, and which occupies a room next door to the Wikipedia offices (same coffee-percolator, better chairs).

He was once quoted as saying that during his days as an options trader he had made enough money for his family to live on for the rest of their lives. He now says the quote has been ‘distorted into a fantasy that I am fabulously wealthy. I’m not.’

Indeed, some internet reports claim he has a net worth of less than $1 million, although he says opaquely that ‘by living wisely and within your means, being financially independent is a lot easier than most people realise. I travel a lot all over the world, and one of the things that world travel helps one to realise is that having enough money to live and support yourself for the rest of your life is easy enough. People live on a fraction of a typical American income.’

Jimmy Wales has described Wikipedia as an embodiment of the spirit of idealism that first accompanied the development of the internet – as a place where ‘people can communicate from all over the world and build knowledge and share information. We went through the whole dot-com boom and bust, and the internet seemed to be about pop-up ads and spam and porn and selling dog food.

And now Wikipedia kind of harks back to the original vision.’ In this respect the site is in the vanguard of what has been called ‘Web 2.0’, the ‘second generation’ of websites such as YouTube and Flickr that let users collaborate and create new material. (However, unlike YouTube, as a charitable corporation Wikipedia cannot be sold. It would, as Wikipedia’s lawyer Brad Patrick told me, ‘be like selling Greenpeace’.)

A key text in the development of Wikipedia was an essay (later a book), The Cathedral and the Bazaar, published in 2000 by Eric S Raymond, an internet developer and computer hacker. Raymond proposed two models of software development: one is the cathedral, ‘carefully crafted,’ as Raymond put it, ‘by individual wizards or small bands of mages working in splendid isolation’; the other is the bazaar, where software code is released on the internet for anyone to make a contribution. Raymond argued that the bazaar model actually produced better results, and the more people who contributed to a complex project, the quicker it would be completed.

This idea has been further developed by the book The Wisdom of Crowds by James Surowiecki, which argues that the aggregation of information from a given group of people often results in decisions that are better than any made by any single individual within that group.

Surowiecki cites the example of a crowd at a county fair accurately guessing the weight of an ox when their individual guesses were averaged, the average being closer to the ox’s true weight than the estimates of most crowd members, and also closer than any of the separate estimates made by cattle experts. (The ‘Ask the audience’ option on Who Wants To Be a Millionaire? might be another example.)

Applied to the model of Wikipedia, this is a potentially revolutionary idea. For ‘crowds’ read ‘the wisdom of the global community’. By dint of collective participation, not only will the range and depth of information increase, but consensus will also bring with it its own checks and balances, and factual errors will also become less likely.

That’s the theory, anyway. But it quickly became apparent that what one regular Wikipedia contributor has described as a ‘socially Darwinian evolutionary process’ was destined to generate its own particular problems.

Within months of coming online, Wikipedia had been infiltrated by hordes of ‘vandals’, inserting mischievous or bogus entries. In 2001 Wales was obliged to appoint a small group of administrators, ‘admins’, to police the site. Admins can delete articles or protect them from further changes, and also block users from editing.

(There are now nearly 1,000 admins, selected by the community at large. A ‘robot’ software program regularly trawls the site, striking out obscenities.)

Differences of opinion developed into protracted ‘edit wars’, where antagonists simply deleted each other’s entries and re-instated, or ‘reverted’, their own. In 2004 Wales introduced the ‘3R rule’, whereby any user who reverts the same text more than three times in a 24-hour period is banned from editing for a day.

Tampering with biographies of living people has proved particularly problematic. The entry for President Bush, which was subject to more or less continual vandalism, is now protected and can be changed only by administrators. (The first entry ran to some 1,200 words: the present one is closer to 25,000, replete with sub-sections on all aspects of Bush’s policy, legislation and cabinet changes.)

Perhaps the most infamous example of tampering involved a retired American journalist named John Seigenthaler, who in the 1960s worked as the assistant to the US attorney general Robert Kennedy. Seigenthaler was surprised on browsing his entry to learn that he had been thought to be ‘directly involved’ in the assassinations of both Robert and President John F Kennedy, although ‘nothing was ever proven’.

The libel, which had been planted by someone as ‘a joke’, had remained on the site for almost four months without anyone noticing. Seigenthaler, not surprisingly perhaps, has since described Wikipedia as a ‘flawed and irresponsible research tool’. But as a result of his case, the community has now inaugurated ‘living biography patrols’ to guard against abuses.

To spend any time on Wikipedia is to be confirmed in the truth that any large group of enthusiasts, left to their own devices, will immediately begin to organise themselves into a hermetic society with its own byzantine hierarchies, secret language and arcane practices. In addition to admins there are also ‘bureaucrats’, who have the power to create or dismiss admins.

There is an arbitration committee to formally resolve disputes between users that has the power to block persistent troublemakers; and an informal ‘mediation cabal’ – a sort of counsellor/therapy group where people can rehearse grievances and ‘nip arguments in the bud’. (Current causes of anguish include atheism and the Montenegro national football team.)

As well as ‘vandals’, who wilfully wreck pages, there are ‘trolls’, who play practical jokes to provoke other users. A ‘sock puppet’ is someone who creates a variety of pseudonymous personalities in an attempt to influence an argument. ‘Wikignomes’ spend all their time browsing the pages in search of something to edit, even if it is only changing a semicolon to a full stop.

Contributors award each other ‘barnstars’ for ‘tireless contribution’ and ‘random acts of kindness’. The ‘Village Pump’ contains interminable discussions about such topics as ‘Anglocentric language domination’, online behaviour (or ‘Wikiquette’), governance and protocol. This is Wikipedia’s Whitehall, apparently inhabited by people of a civil service turn of mind who seem hardly to venture on to the editorial pages at all.

Within this hive of egalitarianism, Wales (or ‘Jimbo’, as he is universally known) holds a sort of first-among-equals position that, he says, ‘English people might understand a little better than Americans do. In the sense that…’

He is the king? ‘Yes.’ He pauses. ‘But with limited and mostly theoretical powers that are diminishing over time. In the early days the only way that anybody could be banned from editing would be for me to do it personally. That became unwieldy very quickly, simply because there was a lot of pressure on me.

'I felt a great emotional burden because I have the sense that it’s really important that we do not block people because we disagree with them. It has to be a really serious behavioural issue. And we have a long tradition of good success of trying to help people to understand a better way of interacting.

'A lot of people come in from other internet environments that are quite hostile and they start ranting, and after a couple of days people say, hey, we don’t really do that here…

‘One of our oldest rules is “no rules”, so if the process appears to be failing in some obviously miserable way, then I will just change it and block those people from editing, or change the rules or whatever we need to do. But if I do take action on anything, it’s usually late in the day and when all of the top users have pressed me to do something.’

In fact, Wales’s personal involvement in the English language site has lessened in recent times. Much of his time nowadays is spent on ‘missionary’ work, travelling the world helping to nurture Wikipedia sites in other languages.

Our meeting was abruptly curtailed when he realised he was late for a flight to Poland. While he may be Wikipedia’s founder and its nominal figurehead, he would be the first to admit that the true custodians of the project are actually ‘the community’: the thousands of people who contribute to it every second of the day.

There are more than 200,000 registered users on the English site of whom, Wales estimates, about 30,000 contribute on a regular basis. Because the vast majority of Wikipedians contribute to the pages under pseudonymous user-names, it is impossible to create an accurate picture of who they are. Wales estimates that 80 per cent are male and a large proportion are students.

The common denominator seems to be an eclectic range of enthusiasms, and an obsessive turn of mind. Danny Wool, who works as Wikipedia’s grants coordinator is, in many ways, archetypal. Fortythree years old, an amiable man who resembles a more rumpled version of Michael Moore, Wool previously worked as an educator at the Holocaust Museum in New York.

A book collector (‘I have 6,000 stacked on my floor at home’) and movie buff, he stumbled on Wikipedia while surfing the net and quickly became – his word – ‘addicted’.

He has originated entries on such diverse topics as Billie Holiday, African politics, the films of Pasolini, and Margaret, Queen of Scotland, and says that he likes ‘nothing better’ than poring over Wikipedia correcting people who don’t know how to spell ‘its’. ‘Big pet peeve of mine. My last edit today before I came to work was to change a period to a semicolon. That may seem really minor, but if 1,000 people are doing that every day, we’ll have something really good come of it in the end.’

Certain users hold a particularly exalted position within the community. The most ubiquitous is Simon Pulsifer, a college graduate from Toronto, who has contributed more than 72,000 edits and written on such topics as the history of Central Asia, the Italian Renaissance and the Marshall Plan.

Another user, ‘Lord Emsworth’, is a 17-year-old student from Virginia whose extensive entries on the British monarchy, the Orders of the Garter, St Patrick and the Thistle, Papal Conclave and Tynwald Day have earned him the largest number of featured articles on the site. ‘He could be at home with his mom screaming at him, “Get off the computer and go to bed!”,’ Wool says with a chuckle.

When I click on Lord Emsworth’s user page a note announces that he is on a break from Wikipedia activities. When I try to contact him through Danny Wool, his e-mail is disabled. Lord Emsworth is apparently doing his homework.

The first question that anyone who has ever used Wikipedia asks is, how much can it be trusted? The best-known test of the site’s accuracy is a survey conducted by Nature magazine in 2005, which subjected 42 entries on scientific topics from both Wikipedia and Encyclopaedia Britannica to ‘blind’ peer review.

The results concluded that the average science entry in Wikipedia contained about four inaccuracies, compared with about three in Britannica. Almost close enough for a cigar.

Britannica responded with a 20-page document attacking the survey as ‘fatally flawed’, claiming ‘dozens’ of inaccuracies attributed to Britannica were not inaccuracies at all, and that a number of articles examined by Nature were not even in Britannica, but in supplementary publications.

The rebuttal went on to add that Britannica had never claimed to be error-free, but that the encyclopaedia had a reputation based on ‘strong scholarship, sound judgement and disciplined editorial review.’ Facts are one thing, of course, but how they are presented is quite another. Can the truth of anything be arrived at by consensus?

On Wikipedia, the struggle for the elusive grail of NPOV can be long, hard and extremely fractious, as conflicting ‘facts’ and interpretations are trapped in butterfly nets, wrestled to the ground and hammered in with a sledgehammer.

Certain pages have been virtual war-zones since the day they opened – the Israel-Palestine question, abortion, the Unification Church of the Rev Sun Moon (or is it a cult?). A particularly heated debate between a Left-wing university professor and a Right-wing historian has marked the progress of the entry on Robert Mugabe (despot or champion of anti-colonialism?).

‘They were biting each other’s heads off,’ Wool says. ‘But we were eventually able to synthesise some sort of modus vivendi that was acceptable to both sides.’

Even more abstruse was the argument over whether the city on the Baltic Sea should be entered under ‘Gdansk’ or ‘Danzig’. Which led to further discussion about whether it should be simply ‘Gdansk’, or ‘Gdansk, Poland’. Which, in turn, led to a dispute over whether American cities should be separated from their state by a comma or brackets. ‘Huge fight,’ Wool says. ‘And do we say Chicago, Illinois, or just the abbreviation Ill?’ (It’s Chicago, Illinois.) Wikipedians are nothing if not pernickety.

Many of the entries are amateurish, poorly written, or little more than bullet-points; the process of endless addition and correction can often make even the most comprehensive articles seem stylistically clumsy. In an attempt to raise standards, Wikipedia has introduced a ‘featured article’ ranking for its best entries.

There are some 1,200 designated thus – about 0.1 per cent of the total. The site is also an open door for self-publicists. Just as it has long been common practice for the professionally insecure and the self-obsessed to punch their name into Google, so a Wikipedia entry is rapidly becoming a symbol of being somebody – if only in the ether of cyber-space.

It is considered bad form to submit an entry about yourself, but acceptable to correct factual errors.

Wales acknowledges that politicians doctoring their entries to present themselves in a more favourable light has been a recurring problem. According to his entry, Wales himself has been criticised for editing his own biography in a way some characterised as ‘revisionist history’, including removing references to Larry Sanger as the co-founder of Wikipedia and downplaying the sexual nature of some of the sites to be found on his former company Bomis.

In both cases Wales argued that his modifi cations were solely intended to improve the accuracy of the content. He now says that while he did edit his own entry ‘on some occasions’, he no longer does so. ‘I edited it publicly under my own name; it caused no stir in the community, but when the media heard about it they thought it sounded scandalous, so I decided it was better not to do that.’

A cursory examination of Wikipedia suggests that it tends to be strong on computer sciences, technology and popular culture, weak on the humanities.

Strong on Britney Spears, Star Wars and Homer Simpson; weak on Elizabeth Schwartzkopf, The Battleship Potemkin and, er, Homer. ‘A lot of Wikipedia is very mediocre,’ admits David Gerrard, a 39-year-old computer systems manager from Australia who lives in London, and a dedicated Wikipedian (his user page notes him as an ‘administrator’, ‘ex-arbitrator’ and ‘CheckUser’ – among other roles). ‘But our best stuff is absolutely fantastic – as good as an article can get.’

Gerard himself takes a particular interest in the entry on Scientology (he runs a separate critics’ web-page on the subject), and says he believes the Wikipedia entry is the most balanced and informative account to be found anywhere on the web.

‘This is a good example of why NPOV is one of the most revolutionary things about Wikipedia. On the web you’ll find a lot of Church of Scientology sites, and a lot of critics’ sites, which are generally very bitter. On Wikipedia you have to be neutral, and you have to be able to reference your facts.’

One of the most frequent criticisms of Wikipedia is that it makes no acknowledgement of expertise; a professor of mathematics or philosophy could find themselves going head to head with an enthusiastic amateur (or somebody whose homework they might have marked), and find themselves being worn down by the sheer persistence of their antagonist.

The case of William Connolley is particularly illuminating in this regard. A 42-year-old climate modeller at the British Antarctic Expedition in Cambridge, Connolley found himself embroiled in a ‘revert’ war with a sceptic over the subject of global warming that lasted the best part of a year.

Connolley told me he has no idea who his antagonist was, but it was clear to him that ‘they didn’t know science’. Eventually, the argument went to arbitration, where a ruling was made not on the basis of the facts of global warming but on etiquette and policy. Connolley’s opponent was put on a six-month ban from editing any page relating to climate change, while Connolley was put on six months’ ‘revert parole’ – forbidding him from reverting any article relating to climate change more than once per 24-hour period.

Connolley says the argument about the article ‘sort of resolved itself’ when it became apparent that the majority of editors supported his view: the entry about climate change is now, he believes, ‘fairly good. Things have calmed down’ – although he expects a resumption of hostilities when the next International Panel of Climate Change report is published in 2007. He takes a sanguine view of the dispute. The fact that he happens to be an expert in the field is ‘neither here nor there’, he says.

‘I’ve met experts in other fields on Wikipedia who are clearly not editing in a helpful or constructive way. The fact that you can’t say “I’m an expert and therefore the text must say this” is good. It shouldn’t be that way. But unfortunately the middle ground of “I’m an expert and therefore my opinion should probably have more weight than yours” doesn’t always carry.’

However, he admits that the free-for-all nature of Wikipedia, and the prospect of a lengthy and tedious ‘revert war’ may discourage experts from contributing to the site. ‘I’ve stuck it out, but it’s true you don’t see a lot of other professionals in their fields editing things; in global warming, there seem to be very few.

'But there is no professional recognition for doing this sort of thing; it won’t bring you promotion or preferment; and to be frank, people have jobs to do. But I think it’s worthwhile doing it, and I don’t feel I’ve wasted my time.’

Connolley continues to contribute to the site, ‘mostly on climate-related things’. Although among his recent contributions are a series of edits on the entry for Llangrannog (‘a small coastal village in Ceredigion, Wales’).

Is he an authority on the subject? ‘It’s where I went on my holidays.’