Wikipedia and Network Effects – by “Michael H – YLT2012”

Who among us hasn’t observed a teacher sneer at the thought of a student referencing Wikipedia over traditional, non-digital sources? These teachers laud the immutability and consistency of books over time; however, this great strength of books can also be a significant drawback: they lack the generativity and adaptability to incorporate changing historiographical opinions, cutting-edge scientific research, and innovative discoveries that a more flexible medium provides. Indeed, while a biology textbook from the 1940’s or an NBA records almanac from the 1980’s is certainly “consistent,” each fails to incorporate new information as it becomes available.

Generativity and informational accuracy don’t have to be mutually exclusive though. Indeed, a study by Nature in 2005 found that in a representative set of 42 scientific articles Wikipedia contained 162 factual errors or misleading remarks, while Encyclopedia Britannica contained 123. (1) To realize just how remarkable it is that a website that relies on a decentralized, peer-production process could rival an information source with 100 paid, full-time editors and 4,400 contributors, it is necessary to look at the underlying framework of Wikipedia. (2)

Background

Using money earned from his humble beginnings in “erotic photography,” Jimbo Wales sought to create a free, online encyclopedia. In 2000 he conceived of Nupedia, which in the vein of traditional encyclopedias hired experts to write articles. Over the course of 3 years, Nupedia managed to churn out 25 articles. At this juncture, Jimbo Wales sought relief in Postel’s Law (“Be conservative in what you do; be liberal in what you accept from others”) and created a revamped version of Nupedia called Wikipedia, which allowed the general public to create and edit articles using wiki software. The rest is history. Today, Wikipedia contains 23 million articles, spans 285 languages, and appeals to 365 million readers around the globe. Currently, Wikipedia is the most widely used general reference website with 2.7 billion page views monthly. (3) The triumph of Wikipedia over traditional, pay-for-use encyclopedias can be partly attributed to Gresham’s law, which can summarized colloquially as cheap and convenient drives out expensive and high quality.

Wikipedia Model

Encouragement that the Wikipedia model—a model that relies on the collective wisdom of a large number of unpaid volunteers—could be viable was provided by the NASA ClickWorkers experiment, which ran from November 2000 to September 2001. In the experiment by NASA, unpaid volunteers visited NASA’s website to mark and classify craters and “honeycomb” terrain on Mars. (4) The study produced two surprising and interesting results. First, people are willing to engage in an unpaid, novel, and productive experience merely for the fun of it.  Second, an amalgamation of data contributed by many unskilled volunteers can be virtually indistinguishable from the output of a trained worker. Thus, large groups of people are capable of producing high-quality work for free.

A Counterintuitive Proposition

It seems hard to fathom that a website that allows users cloaked in a veil of anonymity to edit the content of articles could rival the quality of Encyclopedia Britannica.  In an attempt to understand the success of Wikipedia, it is interesting to observe a city in the Netherlands, Drachten. The city has chosen to forgo basic traffic regulations in an attempt to increase safety on the roads. The experiment in Drachten initially has shown promise. Some attribute this to the difference between the effects of rules and standards. While a rule is a regulation that stipulates precise boundaries and is either followed or broken, a standard is more ambiguous and up to interpretation, calling for people to exercise sound judgment. While people might try to circumvent rules that they perceive to be imposed by arbitrary, external forces, they can become more considerate of others when their personal judgment is called upon. As a result, relaxing rules can have the paradoxical effect of causing people to adhere to the desired behavior more closely. (5)

Putting It All Together

So what do NASA and traffic regulations in the Netherlands have to do with Wikipedia, you might ask? These two anecdotes lend credence to the basic assumptions of the Wikipedia model—that the general public is capable of yielding nearly scholarly work with minimal regulation. While the notion of many small contributions forming a remarkable finished product seems strange with respect to encyclopedia articles, consider the analogy of evolution: slight genetic mutations over time in individual agents within a population lead to the betterment of the species as a whole.  A similar model is used in scientific research: major breakthroughs rest on the small contributions of many scientists. While this model may seem strange for information compilation, it is certainly not novel.

The Good, the Bad, and the Ugly

It is unsurprising that many of the flaws that arise concerning Wikipedia are quickly ameliorated; indeed, Wikipedia relies on the procrastination principle—rather than trying to forecast potential problems, it waits for a particular problem to arise and then fixes it. For example, immediately following initial reports of Michael Jackson’s death, “edit wars” ensued on Wikipedia regarding the veracity of these claims. In response to such edit wars, Wikipedia adopted the three-revert rule, which stipulates that an editor should not make the same correction to an article more than three times in one day. Another example of Wikipedia’s remarkable ability to adapt lies in its response to criticism by a former editor-in-chief of Encyclopedia Britannica, Robert McHenry. When McHenry pointed out that Wikipedia failed to note the ambiguity associated with Alexander Hamilton’s birth year, a mistake of which Columbia and Encarta were also guilty, users on Wikipedia corrected the error in under a week, a testament to how dynamic the website can be. These are just a couple of the controversies that Wikipedia has responded to effectively and expediently. (For more see Essjay Controversy and Wikipedia Biography Controversy)

My Take

When passing judgment on Wikipedia, I think it is important for us to view it in its proper context. Wikipedia is not meant to be a compilation of flawlessly written, perfectly crafted articles. When such a high threshold for quality is set for content, a bottleneck ensues, leading to an inability to cover certain relevant topics of interest. The three pillars that make Wikipedia so desirable—it’s free, convenient, and unparalleled in the breadth of its information—necessarily lead to a softening of stringent requirements for content quality and review. (You can’t have your cake and eat it too…) As an anecdote in support of the incredible amount of interconnected information on Wikipedia, consider a game that I’m sure most people are familiar with: given topic X and topic Y, start at topic X on Wikipedia and get to a page about topic Y in Z clicks or less. As an example, starting at Harvard Law School I was able to get to Lady Gaga in 4 clicks. (Harvard Law School-> United States->American music->American pop music-> Lady Gaga. Can you beat me?)

I do not understand Wikipedia “hata’s.” I think it is a losing battle to try to argue that due to a small number of factual errors (3.86 per article as determined by Nature), (1) Wikipedia  is completely without redeeming value. At a bare minimum, I think one must concede that Wikipedia is beneficial for obtaining background information on a topic. To return to my initial anecdote, this rationale should at least preclude a teacher from scoffing at a student who includes Wikipedia in his or her works cited page. (Note that I have almost exclusively adhered to citing Wikipedia articles for this blog post.) If you are personally unsatisfied with the content of Wikipedia articles, you can ignore them entirely, contribute towards improving the articles, or pursue litigation against Wikipedia (although you almost certainly will be unsuccessful…).

Personally, one of my favorite qualities of Wikipedia is that it provides a consistent format across articles that are (at least to a degree) targeted towards the general public. As a student interested in technology and the natural sciences, I often have to read about scientific discoveries that occurred in the last couple of years: frequently, I only have two sources to turn to: the original research paper and Wikipedia (a testament to Wikipedia’s generativity). Bearing in mind the complexity of the topics, I seek to wrap my brain around the concepts by skimming Wikipedia before delving into the highly esoteric research papers. I believe that using Wikipedia in this manner is an appropriate use of the website. While many people possess a take it or leave it mentality when it comes to Wikipedia, I believe that it is important to apply basic common sense and reasoning when deciding whether to use the website—if you can tolerate 3.86 errors in your reading on the Heisenberg Uncertainty Principle, then have it; if not, put your laptop up and embark in the direction of the nearest university library.

(1) http://news.cnet.com/2100-1038_3-5997332.html
(2) http://en.wikipedia.org/wiki/Encyclop%C3%A6dia_Britannica
(3) http://en.wikipedia.org/wiki/Wikipedia
(4) http://en.wikipedia.org/wiki/Clickworkers
(5) http://futureoftheinternet.org/static/ZittrainTheFutureoftheInternet.pdf