At 15, How Wikipedia Became the Symbol of Knowledge | QDaily Business History

One of the symbols of the internet's spirit ought to be this site.

Ordinary article

Unlike Diderot, Montesquieu, Rousseau, Voltaire and the other compilers of the first modern encyclopedia, most of Wikipedia's editors are destined never to appear in a history textbook.

Most of them are anonymous accounts made of letters and numbers, or just a string of IP addresses.

Yet what these editors have written is the largest encyclopedia in history — and quite possibly the greatest.

Wikipedia turns 15 this year. It has more than 290 language editions and over 36 million entries, enough that you would have to read for 16 years straight, without eating or sleeping, to get through it all. And it keeps updating and growing.

For most people, this "online encyclopedia," compiled by ordinary people just like them, is the symbol of knowledge: "Let's see what Wikipedia says" is usually the most reliable and efficient way to end an argument about a fact in everyday conversation.

Yet in 2001, the idea of letting everyone edit an encyclopedia was so bold that Wikipedia began as nothing more than a stopgap its founders reached for when they had no better option.

Over 15 years, Wikipedia solved one problem after another: it brought in "wiki" technology to speed up writing, set organizational rules for an open community, went non-commercial, and raised its credibility. Only then did it grow into the internet age's "symbol of knowledge."

"Wiki": the technology that made writing an encyclopedia fast

Wikipedia's two founders came from very different academic backgrounds: Jimmy Wales studied business, while Larry Sanger held a doctorate in philosophy, in the field of epistemology.

Amid the American internet boom of the late 1990s, Wales founded a portal site called Bomis; Sanger, meanwhile, grew more and more interested in programming, and in 1998 even ran a personal site discussing the "Y2K problem".

The "online encyclopedia" the two first set out to build did have volunteer experts writing entries — but every entry still had to pass "peer review," like a journal paper, before it could go online.

That was Nupedia, generally regarded as Wikipedia's predecessor. It began operating in December 1999 and formally launched in March 2000. It survived only four years before the Wikipedia that followed replaced it completely.

Made clumsy by its strict review, this encyclopedia could turn out only 42 entries a year; in its first six months online, Nupedia published just two entries long enough to qualify. Worse, the online edition of the Encyclopaedia Britannica had just opened to the public for free. Both Wales and Sanger knew Nupedia was moving far too slowly.

In the end, it was "wiki" technology that rescued their dream of an online encyclopedia.

In January 2001, over dinner with a friend, Sanger heard about the "wiki" (Wiki) concept. This was a system the programmer Cunningham had invented in 1994, originally for collaborative software design, its name taken from the Hawaiian "Wiki Wiki," meaning "quick."

The wiki system's biggest feature was that it let many people edit and revise the same text while keeping every past version, so users could compare them and roll back at any time.

Sanger quickly saw that this new technology was the best way to speed Nupedia up. According to a report in The Atlantic, the moment dinner ended he ran back to his apartment and called Wales.

On January 10 the two of them built the Nupedia Wiki system. They meant to graft it onto the slow-moving Nupedia site, but most of the experts had no interest in the new gadget.

So Sanger and Wales had to release the wiki system as a site of its own. They named it "Wikipedia," a word made out of "wiki" and "encyclopaedia." That day was January 15, exactly 15 years ago. Wikipedia was born.

The new site grew fast. By the end of 2001 it already had 15,000 entries and 350 members, a pace that startled even its founders.

At first, when Sanger posted the launch announcement on Nupedia's discussion board, all he wrote was: "Go there and write something — it'll take you five or ten minutes at most." He and Wales treated Wikipedia as a diversion on the side.

At the time, Cunningham's own experience had already shown that a wiki could work for collaborative software design. But could a wiki with no review mechanism at all produce a high-quality encyclopedia? Nobody was confident, and nobody bothered to design proper rules for it.

So the early Wikipedia had no order at all. Entries were a mess for a while, as people copied in material of every kind: dictionary definitions, research papers, historical documents.

But Wikipedia had one virtue: speed. True to the Hawaiian word, information that surfaced anywhere could be posted to Wikipedia almost at once.

The steady flood of entries and new members forced the founders to take it seriously. They split the work: Wales ran Bomis, Wikipedia's parent company at the time, from behind the scenes, while Sanger took the title of "Chief Organizer" and handled Wikipedia's day-to-day operations.

First, they posted a long thread, "What Wikipedia Is Not", separating Wikipedia from dictionaries, journals and the like, and setting the first rules for editing entries. Over the next 15 years that document was edited thousands of times, and it is still Wikipedia's "basic law."

For the daily work, Sanger picked for himself the somewhat odd title "Chief Organizer" rather than "editor-in-chief," the job he had held at Nupedia — a sign he already sensed Wikipedia would differ from earlier online communities.

Judging by what came next, he underestimated how much.

Ousting the "Big Brother," building an open community

As more and more members joined, voices calling for a still more open Wikipedia began to appear — hard to bear for Sanger, whose focus was on producing an encyclopedia. The entries this untrained volunteer force wrote were sometimes hard for him to accept as well.

The conflict sharpened fast. In mid-October 2001, Wikipedia had its first "edit war": Sanger and the "opposition" deleted, rewrote and reverted each other on a set of entries, over and over. By November 1, Sanger was demanding more power for himself in blunt terms on the discussion board, which provoked an even bigger backlash.

Even Wales could not go along with this. Wikipedia, he believed, was an open community and needed no "Big Brother."

The quarrel ended with Sanger's departure.

This fight to "oust Big Brother" let Wikipedia, besides fulfilling its original aim of becoming an "online encyclopedia," turn into an early version of a "social network." That alone was worth far more than an ordinary encyclopedia.

With no Big Brother at the gate, Wikipedia built a two-tier editing structure of ordinary users and administrators. Administrators were unpaid volunteers too; they simply had more power to change or undo what others had written.

By Business Insider's 2009 data, most Wikipedia entries were written by ordinary users and then edited, cleaned up and reformatted by administrators.

These "Wikipedians," after neither fame nor money, put in an unusual amount of enthusiasm, time and energy, digging up sources and arguing back and forth, all to make an entry's content — or even the accuracy of its citations — a little better.

Such devotion looks hard to understand. Wales himself has said these people are driven by "nonsense."

In fact, inside the Wikipedia community, administrators are promoted from among the ordinary users who contribute most, while users are ranked by how much they contribute and rewarded with virtual honors. This simple-looking mechanism works remarkably well, drawing tens of thousands of internet users to hand over their cognitive surplus.

Wikipedia also keeps entries and talk pages strictly apart: people can chat and argue as they like on the talk pages, but what goes into an entry has to meet certain standards.

Over 15 years they have built an intricate set of rules, the most important being the "neutral point of view" principle, which requires an entry on a contested subject to reflect the full range of views. Otherwise a warning appears at the top of the page: "The neutrality of this article is disputed. Its content or tone may reflect a clear personal viewpoint or local bias."

All of this keeps a vast cognitive surplus, organizing itself, from turning into chaos. It ensures that this "leaderless" community produces neither the meaningless "slices of life" of Facebook nor the wildly uneven discussion threads of the forums popular at the time, but human knowledge, gathered in something close to encyclopedia form.

Under pressure to survive, Wikipedia finally chose not to be a business

But at the time, Wikipedia's crisis was not over.

In fact, the official reasons given for Sanger's departure included one more: "financial pressure requiring layoffs."

That pressure on Wikipedia was real.

Unlike today, Wikipedia at first belonged to Bomis, the commercial portal Wales had founded. Between 2001 and 2002 the American internet bubble burst and dragged Bomis down with it; finding new revenue became a more urgent task than compiling an encyclopedia.

Wales — and Sanger, before he left — had both hinted that Wikipedia might start selling ads.

Nothing was said outright, but it was already enough to frighten some people: how do you resolve the basic conflict between commerce and an encyclopedia's neutrality?

In February 2002, Edgar Enyedy, a senior user of the Spanish-language Wikipedia, led the whole Spanish-language community out, founding a new Enciclopedia Libre and moving entries from the Spanish Wikipedia to the new site.

This walkout also carried the Spanish-language community's wish for more independence, but worry about commercialization was always the main reason.

Enyedy later told the UK edition of Wired, in an interview, that Wikipedia's domain was still a ".com" — that is, commercial — and that Bomis, as a commercial company, ran against Wikipedia's public-service purpose.

He had demanded the domain be changed to ".org," asking: "Why should I volunteer for a '.com' website?"

At first Enciclopedia Libre grew quickly, its entry count far ahead of the Spanish Wikipedia's — at the widest, six or seven times as many.

But the walkout also pushed Wikipedia to start reforming itself.

In June 2003, the nonprofit Wikimedia Foundation was founded.

Wales announced that Wikipedia and a series of other wiki projects would be handed to the foundation to run, and that he himself would be only one of its five board members, with no special privileges.

Wikipedia's domain changed to ".org," making the non-commercial direction explicit.

To this day, Wikipedia lives on the foundation's investments and individual donations. Since 2010, at the end of every year, it has also run an annual "small-donation ad".

Looking back now, 2002 was the most chaotic and fragile year in Wikipedia's history.

But it was in that mess that Wikipedia laid the values it would later grow on: an open community and no commerce.

After that, the community grew fast.

The Spanish Wikipedia permanently overtook Enciclopedia Libre again around 2004, and the English edition passed 2 million entries in 2007, formally becoming the largest encyclopedia in history. That same year Wikipedia's traffic entered the global top ten, and it has not dropped out since.

How far is Wikipedia from the truth?

But keeping information reliable in a community this large, and open to anyone, is a genuinely hard task.

On social sites like Twitter and Weibo, a fair share of false information always mixes in before a story has been passed around and picked over — debunked, and debunked again. Wikipedia is much the same: most entries do settle into a relatively reliable version after repeated editing, but until then, holes are unavoidable.

That was especially true during Wikipedia's fastest-growing years, when the flood of inexperienced new editors left entry quality uneven.

The Colbert Report, a veteran American comedy talk show, seized on the weakness and gave Wikipedia a public thrashing.

In August 2006, Colbert had a group of people add one inconspicuous false line to Wikipedia's "elephant" entry: "The number of elephants has tripled in the last six months."

The figure, impossible to check at a glance, slipped past the editors and became the classic case for a word Colbert coined on the show, "wikiality": if you can persuade enough people to believe what you wrote on Wikipedia, even a falsehood becomes truth.

Carried by The Colbert Report's reach in the United States, the incident quickly became a crisis: is an encyclopedia like this still worth trusting? And if it isn't, is there any point to it?

At that year's Wikimania, founder Wales said that from then on Wikipedia would put the quality of entries ahead of their number. "We're moving from an era of growth to an era of quality." Shortly after the crisis passed, Wales said the same to The New York Times.

Later, Aaron Halfaker, a senior researcher at the Wikimedia Foundation, built the Objective Revision Evaluation Service (ORES), which lets editors look through an entry's revision history and, with an easy click, remove or undo substandard edits.

Halfaker, who oversees this work at Wikipedia, told Wired that editors now put more of their effort into correcting errors in entries, and that volunteers patrol pages far more carefully than before. The "elephant" entry Colbert once vandalized is still locked under "protection."

Since 2010, Wikipedia has also worked with institutions such as the British Museum and the U.S. National Archives to make its entries more accurate.

Wikipedia will grow old, but "wiki" will live on

Today, at 15, Wikipedia still sits seventh in Alexa's global website traffic ranking.

And most free internet programs built for the poor are sure to preload Wikipedia among their services, since it is the most reliable and convenient source of information people can think of. That makes the room for its traffic to keep growing look boundless.

But Wikipedia's problems are surfacing too.

By MIT Technology Review's numbers, on the most important edition, the English one, the number of editors has fallen nearly 40% since 2008, to just over 30,000 today; the rate at which new administrators are promoted has dropped from more than 60 a month at its peak to fewer than one a month.

Wikipedia's past success rested on a volunteer community large, active and open enough that anyone would edit an entry and jump into a heated argument on a talk page.

Today, though, Wikipedia is more and more closed off. This site, perhaps the earliest successful "social network" in internet history, no longer looks "social" to the public at all.

In fact, even as its contributing users decline, Wikipedia's entries keep appearing at the old rate.

Which means editing on Wikipedia has grown more and more concentrated. The whole model has drifted toward a group of professional volunteers writing and "peer-reviewing" each other, while most of the public just watches. In a way, it resembles the old Nupedia.

The reason: editing Wikipedia is complicated work, and with the current system the barrier to revising entries, comparing versions and arguing with others on a phone screen is simply too high.

So in the mobile internet era, the community's expansion naturally stalled, or even declined. According to a 2014 report in The New York Times, only 1% of entry edits on Wikipedia are made entirely on a phone.

At the same time, over the past 15 years Wikipedia has developed a rich community culture, with its own etiquette and conventions. That has standardized entry formatting and strengthened cohesion, but it does not help newcomers fit in. One survey found that 60% of Wikipedia's users created or edited no entries at all when they first registered.

Wikipedia's future may still look unclear, but for now, at least, it keeps up a startling pace: 15,000 edits and 7,000 new entries every hour. Not long ago the Wikimedia Foundation also set itself the goal of raising another US$100 million within ten years.

More important, Wikipedia has carried "wiki" technology so far that it is now a common way to organize and share information across the internet.

Big companies including Adobe, Amazon, Microsoft and Intel have built internal wiki systems where employees upload and look up company documents and internal information.

Wiki communities have also sprung up in one vertical field after another.

As early as 2003, inspired by Wikipedia, Wikitravel launched. Built on wiki technology and open to anyone to edit, this travel guide is something like a specialized Wikipedia for travel (though it has no connection to the Wikimedia Foundation).

Wikia, launched in 2004, does something similar: it sets up wiki sites for internet users free of charge and makes its money from advertising.

Over more than a decade, Wikia has gathered a large number of specialized "Wikipedias," the better known among them the Star Wars Wiki, the World of Warcraft Wiki and the A Song of Ice and Fire Wiki. Their overall influence falls far short of Wikipedia itself, but these narrower wiki communities show stronger cohesion.

Perhaps one day Wikipedia will grow old. What it leaves the world will not: an open, efficient way of organizing information.

Cover image from: The Economist, Wikipedia, Zeenews