Wikipedia was born in January 2001, at the dawn of a new century -- fitting for a site that would unexpectedly mark a new era in the evolution of human knowledge. In less than a decade, Wikipedia has become the world's most popular encyclopedia, expanding from a lone first article in English (a test post on the site with the text "Hello, World!") to more than 10 million articles in 250 languages. But perhaps even more important than Wikipedia's size is our increasing dependence upon the site. It consistently ranks with Google and Yahoo as one of the top 10 Web destinations. What many of us once viewed as a joke, a place to play pranks on friends (or enemies), has become our first stop for the scoop on everything from the Palestinian-Israeli conflict to Paula Deen to, well, Wikipedia.
In his new book, "The Wikipedia Revolution," Andrew Lih attempts to catch us up to speed on the history behind the virtual encyclopedia. An academic and a media critic who has taught online journalism and new media at Columbia University and Hong Kong University, Lih traces the advancement of all open-source software from the dimly lit halls of university philosophy departments to the cubicles of computer programming companies to the form it takes today.
Wikipedia started as a populist experiment developed by entrepreneur Jimmy Wales and philosopher Larry Sanger. When the site began, anyone could write new articles or edit the existing ones. This openness and inclusivity encouraged the formation of an online community of unpaid volunteers dedicated to expanding the scope and accuracy of Wikipedia's entries. Wales' devotion to Ayn Rand's objectivist philosophy meant that he placed an emphasis on verifiable facts, which can still be seen in Wikipedia's use of footnoted sources in most articles.
Before writing "The Wikipedia Revolution," Lih worked as a Wikipedia editor. And while he makes no attempt to hide his affection for his former hangout, he also recognizes a litany of problems with the site. Sure, Wikipedia is convenient, but is it always right? Since we don't know the identities of the authors of Wikipedia's articles, to what extent can we actually trust the content? The advent of Wikipedia may mean that academia is no longer history's gatekeeper; but are we ready for a world in which anyone and everyone can write history's first draft?
Salon spoke with Lih from his Beijing office about Wikipedia's reliability, its unprecedented growth and whether or not what I ate for breakfast deserves its own place on Wikipedia.
You write, "I have been a Wikipedian for five years ... I can only hope that once in your lifetime you can be part of something this great that advances human achievement." So why do you think Wikipedia is so beneficial for humanity?
Back in 1997, when I was heavily involved in the Internet and teaching at the journalism school at Columbia University, we were contemplating the future of new media. What kind of online content would fill in the blanks between what we call the news and the history books? And I really didn't find anything that could be an answer until I started using Wikipedia in 2003. Suddenly the encyclopedia wasn't just a memorialization of history; it was actually a working draft of history evolving as the news was being produced. That was a complete revelation to me. When I was growing up you could read in the newspaper about the Vietnam War or Watergate but you never really got a deep history. I remember asking my parents, this Vietnam War thing, how did it start and what's going on? And they wouldn't have the time to explain it to me. Today, if a teenager were to say, can you tell me about what's going on, they would immediately go to Wikipedia. It has completely changed the way we consume the news.
But how different is that from blogs or online media now, where everything is instantaneous and we don't seem to ever have to wait for news?
The difference with Wikipedia is that it's actually sifted, edited and made into a coherent article. And that's quite different than if you go to Google News. If you collect all the stories written about the Iraq war for the last eight years, you'd find that an article written today might contradict the facts we knew about a year ago, and you would have to correlate that on your own. Because of the volunteer nature of Wikipedia, people have actually edited and updated the articles. It is actually a distillation of all those facts. So that's why the slogan of Wikipedia is, "It's the Sum of all Human Knowledge."
How reliable is Wikipedia?
That notion that anyone can edit it doesn't mean that anyone does. The people who watch articles for inaccuracies or vandalism or fact-check or look up references, those are the folks who stick around, and in general, the quality of any given article is actually quite high. Now, there are no guarantees. And that's what they're working on right now with a new feature called flag revisions that actually rates or flags articles in terms of their quality so people have some indicators about whether they should trust the article or not.
Wikipedia started as an open source model where freedom was prized. You cite their mantras as "be bold," "ignore all rules" and "anyone can edit." But the site no longer operates without rules and administrative hierarchy. How free or open is Wikipedia now?
Back in 2003 or 2004, when I first started editing, there were not even 100,000 articles. So the task back then was to add things. Now we're at 2.8 million articles on Wikipedia, and that's a very, very different environment. Right now it's in a refinement mode rather than a growth mode. And because of that, you have some restrictions. So anonymous users cannot create articles in Wikipedia as they could in the early days. There are very strict standards about biographies about living people because of the harm that can be done if there's libelous speech on the page.
And Wikipedia now has committees that monitor both the topics and the site's content. So who decides what gets included? Is Britney Spears a valid topic? Is what I had for breakfast a valid topic?
That's a premium debate on Wikipedia. If something is mentioned in the mainstream media in any significant amount -- like it shows up in a Google news search or a Yahoo news search is a pretty good rule of thumb -- then that merits inclusion in Wikipedia. Ninety-nine percent of the problems come from these 1 percent fringe cases that are always up for debate.
What is the harm, though, with me posting about my breakfast if Wikipedia is supposed to incorporate all of human knowledge or as much human knowledge as possible?
There are several issues there. One, is what you had for breakfast verifiable? In most cases, it's hard to say it is. But on another level, it's not significant historically. There are people who say, exactly as you do, what harm is there? The addition of one extra article about something that you may consider fringe does not impact the printing of Wikipedia because it's not printed. It's an electronic resource. And then the other camp argues that the inclusion of topics that are not historically relevant takes away from the credibility of Wikipedia. These two forces, the inclusionists and the exclusionists or deletionists, wage a perennial battle on Wikipedia.
Let's talk about the somewhat notorious case of the journalist John Seigenthaler. Why did that case impact Wikipedia in such a momentous way?
John Seigenthaler is a journalist who was running a journalism institute at Vanderbilt and there was a rather short Wikipedia entry on him. He wrote a commentary for USA Today saying that, to his horror, he discovered his Wikipedia entry falsely claimed he was part of the conspiracy to assassinate John F. Kennedy. And he said this shows you how terrible Wikipedia is. What good is the site if these types of things can crop up?
That was a major crisis for Wikipedia. It was a very public embarrassment that this type of edit could last so long. And it highlighted something else -- that particular entry was written by someone who had not signed up for Wikipedia. It was an anonymous edit. So the only trace you had was this IP address. This led to a lot of soul-searching within the Wikipedia community. And that led to some restrictions on what new or anonymous users could do. This was kind of the beginning of Wikipedia's promotion to adulthood. It was no longer just this tiny, curious project on the side. Mainstream media was noticing it, and its reliability was called into question.
Wikipedia, despite however many people use it and contribute to it, is still a single source. Is there a danger that we're becoming too reliant on a single source?
Yes. Wikipedia is only as good as its sources. So if there's garbage in, there's garbage out. Hopefully enough verifiable sources make up a Wikipedia entry that it's better than reading just one particular newspaper article or one particular TV show. But the founder, Jimmy Wales, has said that Wikipedia should be the start of research and not the ending point of research.
But is that realistic? Wikipedia is so convenient and easy. Won't people just look at the Wikipedia entry and think they have all the information they need?
I think there needs to be more study on how teenage or college-age folks use Wikipedia and what their expectations are. My intuition is that they are pretty media-literate, and they know there's some skepticism that needs to be put into the process.
You make a strong argument that Wikipedia and the whole Wiki model is a community. People aren't going to Wikipedia to write funny posts or do pranks -- this is a committed group of people who genuinely come to know one another. Is that communal aspect part of the reason Wikipedia has worked so well?
Yes. Something that fascinates outsiders is how seriously people consider these policies. The type of conversations you have on Wikipedia are the kinds you'd see in newspaper newsrooms or at an academic faculty meeting. What are the criteria for making sure something is verifiable? What are the policies we should implement to include a lot of people but also ensure quality? There is thoughtful deliberation behind the management of millions of articles. This is not a top-down model but a very Quaker-esque consensus model.
You also mention this theory called the Piranha effect, which, to really roughly summarize, is that if you have a bunch of people nibbling on the same subject, they'll devour it really well. With their new policies, is Wikipedia now moving away from that and becoming more hierarchical, reliant on experts, which is exactly what they were against when they started?
It's not clear that they're relying more on experts. Wikipedia is still based on the idea that your edits, the quality of the work that you do, is what matters and not your pedigree. But that's also one of the funny laments of Wikipedia. That you might have a tenured professor debating a 14-year-old high school student and they are equal -- for better or for worse.
But should they be? Shouldn't experts have a dominant role in interpreting history?
The majority of the articles on Wikipedia aren't on subjects where experts are normally needed. You do get into topics where experts are useful, especially in the scientific articles. This is something that has always been up for debate, whether or not there should be a special place for experts. And that's one reason why Larry Sanger, one of the original founders of Wikipedia, has been kind of shunned by Wikipedia, because he does believe that experts should have a privileged role.
And he's started his own site.
That's right. He started a project called Citizendium. He does still believe that anyone can come on and be a productive editor. But he believes that the people with expertise should have a privileged role.
Wikipedia doesn't have advertisements. In the book, you write that when Larry Sanger merely mentioned considering placing ads on the site, it caused a huge schism in the Wiki community. Wikipedia is now officially a nonprofit organization. So where did the funding to start the site come from and how does the organization generate funding now?
As a normal nonprofit, they needed to raise money for operating servers and some basic technical costs. They were able to raise $250,000 fairly easily in one quarter. And this was simply by putting on every Wikipedia page "We need your support." They got around the same numbers you'd see for a PBS pledge drive: $15, $20, $25 was pretty standard. In the same way that hundreds and thousands of tiny little edits make up the entries on Wikipedia, hundreds and thousands of $20 donations helped to make the money needed for the site. When you use PayPal, you can put in this one-line note, and it was quite fascinating to see a lot of college students saying things like, "I can't afford much this week, but I'm giving you $5 because you saved my rear on this thesis this week."
The Wikipedia organization has since staffed up with about two dozen employees with a $6 million budget. Before 2005, the budget was less that $1 million a year. This is quite fascinating for a site that was rivaled by Google, Microsoft and Yahoo in terms of influence on Internet traffic.
Do you think open sourcing works better in some mediums than others? We've had Wiki novels in recent years, for instance, and to put this bluntly, you know you're not reading Dostoevsky.
Encyclopedia writing was a perfect match for Wiki style. Encyclopedia writing is pretty well understood -- using the inverted pyramid, where the first three sentences give a complete summary of what you're about to read and as you go down the article gives you an elaboration of the basic information. It's very structured data.
When you take the Wiki method and try to apply it to novel writing – one of the more famous examples was "A Million Penguins" -- you need a narrative arc, you need people who are collaborating to agree on this arc, you need to work on the pacing, you need to work on the characters. It's very, very hard to do that with a novel. Wikitorials was a Los Angeles Times experiment where people would write editorials in Wiki style. That turned out to be a mess as well.