It's a symbol of rigour and authority in a medium notorious for disinformation. But is the fiercely policed Wikipedia now starting wobble? Steve Bowbrick investigates

Wiki means "fast" in Hawaiian. Of course, you don't need me to tell you that. By now you'll have typed "wiki" into a search engine and probably also found out that some people think it's an acronym standing for "What I Know Is" but that they're wrong (that's just a "backronym"). You'll have learnt this, of course, from Wikipedia, the most famous wiki of them all - a vast free online encyclopaedia written and maintained by volunteers.

A wiki is a very simple thing. It's a web page with a prominent button labelled "edit this page". That's it. Whenever you see a button like that you're probably looking at a wiki (go on, click it). That little button seems like a pretty radical addition to a web page when you think about it ("I can write anything I want?"), but back in 1995, when an Oregon techie called Ward Cunningham invented the wiki he was really just returning to the web's original plan.

In the early '90s, when the WWW arrived, Tim Berners-Lee, its inventor, assumed his users would be spending about as much time creating pages as reading them, so his first web browser could edit web pages as well as display them - it was like a word processor for the web. Pretty soon, though, the edit button disappeared as the web evolved into a commercial one-way medium like television, printing and all the others. Ward Cunningham's wiki put the edit button back. Today, there are millions of wikis in use around the world, most of them based on Cunningham's software, including the mother-and-father of all wikis, Wikipedia.

Wikipedia is the fulfilment of a kind of geek fantasy. Geeks - the passionate engineers and programmers who make the net work - can't understand why all information isn't available to everybody right now. They hate the barriers and impediments and bottlenecks that block access to information and consider it their mission in life to eliminate them. They see the messy economic stuff - paying for information, for instance - as just more annoying inefficiency. Information for the geeks is a critical resource, like water or fuel.

Of course, for most of human history there wasn't much that could be done about these technological and economic impediments to the flow of knowledge. The tools didn't exist. Then, at the height of the Cold War, the American military-industrial complex handed a blank cheque to a select group of university engineers. In return, the US government got an indestructible communications network - the internet as we now know it - and the geeks got their tools. The roots of Wikipedia are essentially the roots of the net. The net's genesis as a tool for scientists and engineers embedded a way of thinking about information that's been carried over into Wikipedia's unprecedented openness.

The project's extraordinary cooperative structure - every page has an "edit this page" button and literally anyone can make a contribution, from correcting a grammatical error (my favourite) to writing or amending whole articles (too scary for me) - is both its greatest strength and the source of its problems. It means that without a single paid editor it has grown to be a more comprehensive source than any published reference work. It also means that the risk of deliberately erroneous or malicious content is real. What's remarkable about Wikipedia, though, is how well armoured against vandalism it is.

Fanatical contributors make it their business to protect the integrity of their favourite entries or categories, so dodgy content is usually spotted within minutes, flagged for correction and put right within a few more. Sensitive entries are watched even more closely and a handful (George W Bush, Cuba and Michael Jackson, for instance) are actually off-limits to all but elite Wikipedia editors because they have been vandalised so often. Fifteen hundred dubious entries are deleted from Wikipedia every day, after a surprisingly thorough review process deems them "non-notable". Arguments over the notability or otherwise of such entries rage on the site's talk pages. Of course, the very idea of an editorial elite obviously somewhat undermines Wikipedia's claim to be democratic but the truth is it's not a democracy.

It's more like a vast social experiment or a game played by millions - and it's important because what we're learning from this experiment will shape the way knowledge is shared in the future. The Wikipedia experiment suggests that groups of people - even very big, uncoordinated ones like the Wikipedians - are well-suited to the creation and maintenance of this kind of broadly objective stuff. What emerges from the Wikipedia sausage machine of collective scrutiny and amendment is not perfect (nor could it be) but good enough. Wikipedia entries are right almost all the time, and that turns out to be often enough for a general audience.

Criticism of the project from old-school knowledge merchants like the Encyclopaedia Britannica centres on the potential for inaccuracy in a community-edited work and has shaken users' faith in Wikipedia's usefulness. The logic of the crowd is essentially unarguable, though.

If you want to build the biggest and most up-to-date reference work in history, tens of thousands of unpaid editors working at 95 per cent accuracy beats a couple of hundred working at 99 per cent every time. Sorry old-timers. It's game over. Naturally a story of such vast significance can't have an entirely happy ending. Wikipedia may be a collective effort but it's also the personal passion of a rather proprietorial millionaire: one Jimmy Wales, retired commodities trader and sole founder (if you believe his story) of the whole Wikipedia shebang. It seems wrong to characterise something so anarchic as an empire but it is showing signs of a kind of late-imperial decadence. Wales' recent bust-up with his scary right-wing pundit girlfriend (Rachel Marsden, the "Fox News Fox") turned nasty and broke through to the gossip columns when she sold his clobber on eBay and he used Wikipedia's talk pages to dish the dirt.

Wales has also been rapped over the knuckles by his foundation's board of trustees for being a bit free with the Wikipedia expense account (he denies everything). Wales is hardly Caligula but these embarrassments, alongside the ongoing fuss about the veracity of the encyclopaedia's content, suggest that we might want to at least prepare ourselves for life after Wikipedia. Alternatives already exist - although none comes close to Wikipedia's size and authority yet - and it's a safe bet that the next five years will see at least one competitive punch-up in the universal free access to human knowledge business. Shall we get started?