The Hive

For those of you who are (as uncharitable Wikipedians sometimes say) “clueless newbies,” Wikipedia is an online encyclopedia. But it is like no encyclopedia Diderot could have imagined. Instead of relying on experts to write articles according to their expertise, Wikipedia lets anyone write about anything. You, I, and any wired-up fool can add entries, change entries, even propose that entries be deleted. For reasons I’d rather not share outside of therapy, I created a one-line biographical entry on “Marshall Poe.” It didn’t take long for my tiny article to come to the attention of Wikipedia’s self-appointed guardians. Within a week, a very active—and by most accounts responsible—Scottish Wikipedian named “Alai” decided that … well, that I wasn’t worth knowing about. Why? “No real evidence of notability,” Alai cruelly but accurately wrote, “beyond the proverbial average college professor.”

Wikipedia has the potential to be the greatest effort in collaborative knowledge gathering the world has ever known, and it may well be the greatest effort in voluntary collaboration of any kind. The English-language version alone has more than a million entries. It is consistently ranked among the most visited Web sites in the world. A quarter century ago it was inconceivable that a legion of unpaid, unorganized amateurs scattered about the globe could create anything of value, let alone what may one day be the most comprehensive repository of knowledge in human history. Back then we knew that people do not work for free; or if they do work for free, they do a poor job; and if they work for free in large numbers, the result is a muddle. Jimmy Wales and Larry Sanger knew all this when they began an online encyclopedia in 1999. Now, just seven years later, everyone knows different.

In 1974, Gary Gygax and Dave Arneson, two gamers who had obviously read The Lord of the Rings, invented the tabletop role-playing game Dungeons & Dragons. The game spread largely through networks of teenage boys, and by 1979, the year the classic Dungeon Master’s Guide was published, it seemed that every youth who couldn’t get a date was rolling the storied twenty-sided die in a shag-carpeted den. Meanwhile, a more electronically inclined crowd at the University of Illinois at Urbana-Champaign was experimenting with moving fantasy play from the basement to a computer network. The fruit of their labors was the unfortunately named MUD (Multi-User Dungeon). Allowing masses of players to create virtual fantasy worlds, MUDs garnered a large audience in the 1980s and 1990s under names like Zork, Myst, and Scepter of Goth. (MUDs came to be known as “Multi-Undergraduate Destroyers” for their tendency to divert college students from their studies.)

Wales began to play MUDs at Alabama in the late 1980s. It was in this context that he first encountered the power of networked computers to facilitate voluntary cooperation on a large scale. He did not, however, set up house in these fantasy worlds, nor did he show any evidence of wanting to begin a career in high tech. He completed a degree in finance at Auburn, received a master’s in finance at the University of Alabama, and then pursued a Ph.D. in finance at Indiana University. He was interested, it would seem, in finance. In 1994, he quit his doctoral program and moved to Chicago to take a job as an options trader. There he made (as he has repeatedly said) “enough.”

Wales is of a thoughtful cast of mind. He was a frequent contributor to the philosophical “discussion lists” (the first popular online discussion forums) that emerged in the late ’80s as e-mail spread through the humanities. His particular passion was objectivism, the philosophical system developed by Ayn Rand. In 1989, he initiated the Ayn Rand Philosophy Discussion List and served as moderator—the person who invites and edits e-mails from subscribers. Though discussion lists were not new among the technorati in the 1980s, they were unfamiliar territory for most academics. In the oak-paneled seminar room, everyone had always been careful to behave properly—the chairman sat at the head of the table, and everyone spoke in turn and stuck to the topic. E-mail lists were something altogether different. Unrestrained by convention and cloaked by anonymity, participants could behave very badly without fear of real consequences. The term for such poor comportment—flaming—became one of the first bits of net jargon to enter common usage.

Wales had a careful moderation style:

Wales was an advocate of what is generically termed “openness” online. An “open” online community is one with few restrictions on membership or posting—everyone is welcome, and anyone can say anything as long as it’s generally on point and doesn’t include gratuitous ad hominem attacks. Openness fit not only Wales’s idea of objectivism, with its emphasis on reason and rejection of force, but also his mild personality. He doesn’t like to fight. He would rather suffer fools in silence, waiting for them to talk themselves out, than confront them. This patience would serve Wales well in the years to come.

Other would-be classifiers approached the problem of Web chaos more democratically. Beginning from the sound premise that it’s good to share, a seventeen-year-old Oregonian named Sage Weil created the first “Web ring” at about the time Yang and Filo were assembling their army of paid Web librarians. A Web ring is nothing more than a set of topically related Web sites that have been linked together for ease of surfing. Rings are easy to find, easy to join, and easy to create; by 1997, they numbered 10,000.

Wales focused on the bottom-up strategy using Web rings, and it worked. Bomis users built hundreds of rings—on cars, computers, sports, and especially “babes” (e.g., the Anna Kournikova Web ring), effectively creating an index of the “laddie” Web. Instead of helping all users find all content, Bomis found itself positioned as the Playboy of the Internet, helping guys find guy stuff. Wales’s experience with Web rings reinforced the lesson he had learned with MUDs: given the right technology, large groups of self-interested individuals will unite to create something they could not produce by themselves, be it a sword-and-sorcery world or an index of Web sites on Pamela Anderson. He saw the power of what we now call “peer-to-peer,” or “distributed,” content production.

Wales was not alone: Rich Skrenta and Bob Truel, two programmers at Sun Microsystems, saw it too. In June 1998, along with three partners, they launched GnuHoo, an all- volunteer alternative to the Yahoo Directory. (GNU, a recursive acronym for “GNUs Not Unix,” is a free operating system created by the über-hacker Richard Stallman.) The project was an immediate success, and it quickly drew the attention of Netscape, which was eager to find a directory capable of competing with Yahoo’s index. In November 1998, Netscape acquired GnuHoo (then called NewHoo), promising to both develop it and release it under an “open content” license, which meant anyone could use it. At the date of Netscape’s acquisition, the directory had indexed some 100,000 URLs; a year later, it included about a million.

Wales clearly had the open-content movement in mind when, in the fall of 1999, he began thinking about a “volunteer-built” online encyclopedia. The idea—explored most prominently in Stallman’s 1999 essay “The Free Universal Encyclopedia and Learning Resource”—had been around for some time. Wales says he had no direct knowledge of Stallman’s essay when he embarked on his encyclopedia project, but two bits of evidence suggest that he was thinking of Stallman’s GNU free documentation license. First, the name Wales adopted for his encyclopedia—Nupedia.org—strongly suggested a Stallman-esque venture. Second, he took the trouble of leasing a related domain name, GNUpedia.org. By January 2000, his encyclopedia project had acquired funding from Bomis and hired its first employee: Larry Sanger.

Sanger fits the profile of almost every Internet early adopter: he’d been a good student, played Dungeons & Dragons, and tinkered with PCs as a youth—going so far as to code a text-based adventure game in BASIC, the first popular programming language. He was drawn into the world of philosophy discussion lists and, in the early 1990s, was an active participant in Wales’s objectivism forum. Sanger also hosted a mailing list as part of his own online philosophy project (eventually named the Association for Systematic Philosophy). The mission and mien of Sanger’s list stood in stark contrast to Wales’s Rand forum. Sanger was far more programmatic. As he wrote in his opening manifesto, dated March 22, 1994:

Wales’s Rand forum was generally serious, but it was also a place for philosophically inclined laypeople to shoot the breeze: Wales permitted discussion of “objectivism in the movies” or “objectivism in Rush lyrics.” Sanger’s list was more disciplined, but he soon began to feel it, too, was of limited philosophical worth. He resigned after little more than a year. “I think that my time could really be better spent in the real world,” Sanger wrote in his resignation letter, “as opposed to cyberspace, and in thinking to myself, rather than out loud to a bunch of other people.” Sanger was seriously considering abandoning his academic career.

As the decade and the century came to a close, another opportunity arose, one that would let Sanger make a living away from academia, using the acumen he had developed on the Internet. In 1998, Sanger created a digest of news reports relating to the “Y2K problem.” Sanger’s Review of Y2K News Reports became a staple of IT managers across the globe. It also set him to thinking about how he might make a living in the new millennium. In January 2000, he sent Wales a business proposal for what was in essence a cultural news blog. Sanger’s timing was excellent.

According to Sanger, Wales was very “hands-off.” He gave Sanger only the loosest sketch of an open encyclopedia. “Open” meant two things: First, anyone, in principle, could contribute. Second, all of the content would be made freely available. Sanger proceeded to create, in effect, an online academic journal. There was simply no question in his mind that Nupedia would be guided by a board of experts, that submissions would be largely written by experts, and that articles would be published only after extensive peer review. Sanger set about recruiting academics to work on Nupedia. In early March 2000, he and Wales deemed the project ready to go public, and the Nupedia Web site was launched with the following words:

The answer, as Wales and Sanger found out, was “not very.” Over the first several months little was actually accomplished in terms of article assignment, writing, and publication. First, there was the competition. Wales and Sanger had the bad luck to launch Nupedia around the same time as Encyclopedia Britannica was made available for free on the Internet. Then there was the real problem: production. Sanger and the Nupedia board had worked out a multistage editorial system that could have been borrowed from any scholarly journal. In a sense, it worked: assignments were made, articles were submitted and evaluated, and copyediting was done. But, to both Wales and Sanger, it was all much too slow. They had built a cathedral.

Apple’s Hypercard offered inspiration. Hypercard was a very flexible database application. It allowed users to create records (“cards”), add data fields to them, and link them in sets. Cunningham created a Hypercard “stack” of software patterns and shared it with colleagues. His stack was well liked but difficult to share, since it existed only on Cunningham’s computer. In the 1990s, Cunningham found himself looking for a problem-solving technique that would allow software developers to fine-tune and accumulate their knowledge collaboratively. A variation on Hypercard seemed like an obvious option.

Cunningham coded and, in the spring of 1995, launched the first “wiki,” calling it the “WikiWikiWeb.” (Wiki is Hawaiian for “quick,” which Cunningham chose to indicate the ease with which a user could edit the pages.) A wiki is a Web site that allows multiple users to create, edit, and hyperlink pages. As users work, a wiki can keep track of all changes; users can compare versions as they edit and, if necessary, revert to earlier states. Nothing is lost, and everything is transparent.

The wiki quickly gained a devoted following within the software community. And there it remained until January 2001, when Sanger had dinner with an old friend named Ben Kovitz. Kovitz was a fan of “extreme programming.” Standard software engineering is very methodical—first you plan, then you plan and plan and plan, then you code. The premise is that you must correctly anticipate what the program will need to do in order to avoid drastic changes late in the coding process. In contrast, extreme programmers advocate going live with the earliest possible version of new software and letting many people work simultaneously to rapidly refine it.

Over tacos that night, Sanger explained his concerns about Nupedia’s lack of progress, the root cause of which was its serial editorial system. As Nupedia was then structured, no stage of the editorial process could proceed before the previous stage was completed. Kovitz brought up the wiki and sketched out “wiki magic,” the mysterious process by which communities with common interests work to improve wiki pages by incremental contributions. If it worked for the rambunctious hacker culture of programming, Kovitz said, it could work for any online collaborative project. The wiki could break the Nupedia bottleneck by permitting volunteers to work simultaneously all over the project. With Kovitz in tow, Sanger rushed back to his apartment and called Wales to share the idea. Over the next few days he wrote a formal proposal for Wales and started a page on Cunningham’s wiki called “WikiPedia.”

Wales and Sanger created the first Nupedia wiki on January 10, 2001. The initial purpose was to get the public to add entries that would then be “fed into the Nupedia process” of authorization. Most of Nupedia’s expert volunteers, however, wanted nothing to do with this, so Sanger decided to launch a separate site called “Wikipedia.” Neither Sanger nor Wales looked on Wikipedia as anything more than a lark. This is evident in Sanger’s flip announcement of Wikipedia to the Nupedia discussion list. “Humor me,” he wrote. “Go there and add a little article. It will take all of five or ten minutes.” And, to Sanger’s surprise, go they did. Within a few days, Wikipedia outstripped Nupedia in terms of quantity, if not quality, and a small community developed. In late January, Sanger created a Wikipedia discussion list (Wikipedia-L) to facilitate discussion of the project. At the end of January, Wikipedia had seventeen “real” articles (entries with more than 200 characters). By the end of February, it had 150; March, 572; April, 835; May, 1,300; June, 1,700; July, 2,400; August, 3,700. At the end of the year, the site boasted approximately 15,000 articles and about 350 “Wikipedians.”

Sanger and Wales knew that online collaborative ventures can easily slide into a morass of unproductive invective. They had already worked out a solution for Nupedia, called the “lack of bias” policy. On Wikipedia it became NPOV, or the “neutral point of view,” and it brilliantly encouraged the work of the community. Under NPOV, authors were enjoined to present the conventionally acknowledged “facts” in an unbiased way, and, where arguments occurred, to accord space to both sides. The concept of neutrality, though philosophically unsatisfying, had a kind of everybody-lay-down-your-arms ring to it. Debates about what to include in the article were encouraged on the “discussion” page that attends every Wikipedia article.

The most important initial question, however, concerned governance. When Wikipedia was created, wikis were synonymous with creative anarchy. Both Wales and Sanger thought that the software might be useful, but that it was no way to build a trusted encyclopedia. Some sort of authority was assumed to be essential. Wales’s part in it was clear: he owned Wikipedia. Sanger’s role was murkier.

Citing the communal nature of the project, Sanger refused the title of “editor in chief,” a position he held at Nupedia, opting instead to be “chief organizer.” He governed the day-to-day operations of the project in close consultation with the “community,” the roughly two dozen committed Wikipedians (most of them Nupedia converts) who were really designing the software and adding content to the site. Though the division of powers between Sanger and the community remained to be worked out, an important precedent had been set: Wikipedia would have an owner, but no leader.

With Cunc clearly in mind, Sanger curtly defended his role before the community on November 1, 2001:

Seen from the trenches, this was a striking statement. Sanger had so far said he was primus inter pares; now he seemed to be saying that he was just primus. Upon reading this post, one Wikipedian wrote: “Am I the only person who detects a change in [Sanger’s] view of his own position? Am I the only person who fears this is a change for the worse?”

On November 4, the Sanger-Cunc contretemps exploded. Simon Kissane, a respected Wikipedian, accused Sanger of capriciously deleting pages, including some of Cunc’s work. Sanger denied the allegation but implied that the excised material was no great loss. He then launched a defense of his position in words that bled resentment:

The blowup disturbed Wales to no end. As a list moderator, he had tried hard to keep his discussants out of flame wars. He weighed in with an unusually forceful posting that warned against a “culture of conflict.” Wikipedia, he implied, was about building an encyclopedia, not about debating how to build or govern an encyclopedia. Echoing Sanger, he argued that the primary duty of community members was to contribute—by writing code, adding content, and editing. Enough talk, he seemed to be saying: we know what to do, now let’s get to work. Yet he also seemed to take a quiet stand against Sanger’s positions on openness and on his own authority:

Wales also made a strong case that anyone deleting pages should record his or her identity, explain his or her reasons, and archive the entire affair.

Within several weeks, Sanger and Cunc were at each other’s throats again. Sanger had proposed creating a “Wikipedia Militia” that would deal with issues arising from sudden massive influxes of new visitors. It was hardly a bad idea: such surges did occur (they’re commonly called “slash-dottings”). But Cunc saw in Sanger’s reasonable proposition a very slippery slope toward “central authority.” “You start deputizing groups of people to do necessary and difficult tasks,” he wrote, “fast-forward two/three years, and you have pernicious cabals.”

Given the structure of Wikipedia there was little Sanger could do to defend himself. The principles of the project denied him real punitive authority: he couldn’t ban “trolls”—users like Cunc who baited others for sport—and deleting posts was evidence of tyranny in the eyes of Sanger’s detractors. A defensive strategy wouldn’t work either, as the skilled moderator’s tactic for fighting bad behavior—ignoring it—was blunted by the wiki. On e-mail lists, unanswered inflammatory posts quickly vanish under layers of new discussion; on a wiki, they remain visible to all, often near the tops of pages. Sanger was trapped by his own creation.

Cunc responded to Sanger’s departure with apparent appreciation:

Characteristically, Sanger took this as nothing more than provocation: “Oh, how nice and gracious this was. Oh, thank you SO much, Cunctator. I’m sure glad I won’t have to deal with you anymore, Cunctator. You’re a friggin’ piece of work.” The next post on the list is from Wales, who showed a business- as-usual sangfroid: “With the resignation of Larry, there is a much less pressing need for funds.”

Sanger made two great contributions to Wikipedia: he built it, and he left it. After forging a revolutionary mode of knowledge building, he came to realize—albeit dimly at first—that it was not to his liking. He found that he was not heading a disciplined crew of qualified writers and editors collaborating on authoritative statements (the Nupedia ideal), but trying to control an ill-disciplined crowd of volunteers fighting over ever-shifting articles. From Sanger’s point of view, both the behavior of the participants and the quality of the scholarship were wanting. Even after seeing Wikipedia’s explosive growth, Sanger continued to argue that Wikipedia should engage experts and that Nupedia should be saved.

Wales, though, was a businessman. He wanted to build a free encyclopedia, and Wikipedia offered a very rapid and economically efficient means to that end. The articles flooded in, many were good, and they cost him almost nothing. Why interfere? Moreover, Wales was not really the meddling kind. Early on, Wikipedians took to calling him the “God-King.” The appellation is purely ironic. Over the past four years, Wales has repeatedly demonstrated an astounding reluctance to use his power, even when the community has begged him to. He wouldn’t exile trolls or erase offensive material, much less settle on rules for how things should or should not be done. In 2003, Wales diminished his own authority by transferring Wikipedia and all of its assets to the nonprofit Wikimedia Foundation, whose sole purpose is to set general policy for Wikipedia and its allied projects. (He is one of five members of the foundation’s board.)

Wales’s benign rule has allowed Wikipedia to do what it does best: grow. The numbers are staggering. The English-language Wikipedia alone has well more than a million articles and expands by about 1,700 a day. (Britannica’s online version, by comparison, has about 100,000 articles.) As of mid-February 2006, more than 65,000 Wikipedians—registered users who have made at least ten edits since joining—had contributed to the English-language Wikipedia. The number of registered contributors is increasing by more than 6,000 a month; the number of unregistered contributors is presumably much larger. Then there are the 200-odd non-English-language Wikipedias. Nine of them already have more than 100,000 entries each, and nearly all of the major-language versions are growing on pace with the English version.

The power of the community to decide, of course, asks us to reexamine what we mean when we say that something is “true.” We tend to think of truth as something that resides in the world. The fact that two plus two equals four is written in the stars—we merely discovered it. But Wikipedia suggests a different theory of truth. Just think about the way we learn what words mean. Generally speaking, we do so by listening to other people (our parents, first). Since we want to communicate with them (after all, they feed us), we use the words in the same way they do. Wikipedia says judgments of truth and falsehood work the same way. The community decides that two plus two equals four the same way it decides what an apple is: by consensus. Yes, that means that if the community changes its mind and decides that two plus two equals five, then two plus two does equal five. The community isn’t likely to do such an absurd or useless thing, but it has the ability.

Early detractors commonly made two criticisms of Wikipedia. First, unless experts were writing and vetting the material, the articles were inevitably going to be inaccurate. Second, since anyone could edit, vandals would have their way with even the best articles, making them suspect. No encyclopedia produced in this way could be trusted. Last year, however, a study in the journal Nature compared Britannica and Wikipedia science articles and suggested that the former are usually only marginally more accurate than the latter. Britannica demonstrated that Nature's analysis was seriously flawed (“Fatally Flawed” was the fair title of the response), and no one has produced a more authoritative study of Wikipedia’s accuracy. Yet it is a widely accepted view that Wikipedia is comparable to Britannica. Vandalism also has proved much less of an issue than originally feared. A study by IBM suggests that although vandalism does occur (particularly on high-profile entries like “George W. Bush”), watchful members of the huge Wikipedia community usually swoop down to stop the malfeasance shortly after it begins.

There are, of course, exceptions, as in the case of the journalist John Seigenthaler, whose Wikipedia biography long contained a libel about his supposed complicity in the assassinations of John F. and Robert Kennedy. But even this example shows that the system is, if not perfect, at least responsive. When Seigenthaler became aware of the error, he contacted Wikipedia. The community (led in this instance by Wales) purged the entry of erroneous material, expanded it, and began to monitor it closely. Even though the Seigenthaler entry is often attacked by vandals, and is occasionally locked to block them, the page is more reliable precisely because it is now under “enough eyeballs.” The same could be said about many controversial entries on Wikipedia: the quality of articles generally increases with the number of eyeballs. Given enough eyeballs, all errors are shallow.

Having seen all of this in prospect, Jasiutowicz asked a logical question: “Can someone please tell me what’s the end point/goal of Wikipedia?” Wales responded, only half jokingly, “The goal of Wikipedia is fun for the contributors.” He had a point. Editing Wikipedia is fun, and even rewarding. The site is huge, so somewhere on it there is probably something you know quite a bit about. Imagine that you happen upon your pet subject, or perhaps even look it up to see how it’s being treated. And what do you find? Well, this date is wrong, that characterization is poor, and a word is mispelled. You click the “edit” tab and make the corrections, and you’ve just contributed to the progress of human knowledge. All in under five minutes, and at no cost.

Yet Wikipedia has a value that goes far beyond the enjoyment of its contributors. For all intents and purposes, the project is laying claim to a vast region of the Internet, a territory we might call “common knowledge.” It is the place where all nominal information about objects of widely shared experience will be negotiated, stored, and renegotiated. When you want to find out what something is, you will go to Wikipedia, for that is where common knowledge will, by convention, be archived and updated and made freely available. And while you are there, you may just add or change a little something, and thereby feel the pride of authorship shared by the tens of thousands of Wikipedians.

Bear in mind that I knew none of these people, and they had, as far as I know, no interest other than truth in doing all of this work. Yet they didn’t stop with verifying my claims and approving my article. They also searched the Web for material they could use to expand my one-line biography. After they were done, the Marshall Poe entry was two paragraphs long and included a good bibliography. Now that’s wiki magic.