This is the fascinating history of of a most useful website. Historians may consider Wikipedia as significant as Guttenberg printing press. Both contributed immensely to the spread of knowledge. Lih does an excellent job of conveying the history of Wikipedia and drilling down in the technicalities of this phenomenon from a cultural, software, and governance standpoints.
Wikipedia was developed over numerous years through the interactions of maverick programmers. It started with Tim Berners-Lee's creation of the World Wide Web in 1990. Then in 1995 Ward Cunningham creates the WikiWikiWeb software that supports Wikipedia capabilities. This software allows to create, write, and edit webpages and saves every version of a page. Ben Kovitz introduces this wiki software to Jimmy Wales and Larry Sanger, the cocreators of an earlier online encyclopedia: Nupedia. In 2001, Jimmy Wales implements Cunningham's wiki software to create Wikipedia. Within less than a month, Wikipedia achieves more than Nupedia did in a year in terms of number of articles published. Soon after, a German programmer Magnus Manske far improves the wiki software by allowing Wikipedians to maintain a clean page of an article while debating issues freely on a discussion page.
In 2002, Derek Ramsey finds a way to automate the creation of 33,832 articles about small towns in the U.S. by automating the extraction of data from the U.S. Census. Seth Anthony and others add actual maps to those cities. Sunir Shah creates MeatballWiki to discuss online community. It will prove instrumental for Wikipedia's future policies. Ultimately, Sanger adopts three dominant editing principle: neutral point of view (NPOV), verifiability (V), and no original research (NOR).
Larry Sanger, the chief editor of Wikipedia will struggle with his wish to facilitate Wikipedia explosive growth based on its free wheeling nature where anyone can publish article. But, Sanger seeks quality in Wikipedia's articles through formal editorial control. He created a burdensome seven step editing process for Nupedia. This caused the average article to take more than two weeks to get published vs Wikipedia where an author can publish an original article immediately.
At Wikipedia, the editing comes after the fact. But, it has no finish line. Thus even if an article is mediocre at first, it improves rapidly. After a couple of years, Larry Sanger leaves Wikipedia in 2002 as he feels it lacks credibility. In 2006, he develops a competing encyclopedia, Citizendium where a hierarchy of experts dominates the article publishing process. Within its first year it will publish 4,000 articles vs 20,000 for Wikipedia. Also, when comparing a few articles somehow Wikipedia's wild wisdom of crowds approach seems very competitive in terms of quality with the hierarchy of experts at Citizendium. Thus, Citizendium outlook is not that encouraging (as Wikipedia appears to beat it on both productivity and quality).
The chapter describing the different culture of the various language-Wikipedias is very interesting. The Spanish one is the most idealistic. When Larry Sanger mentioned Wikipedia may consider selling ads to generate revenues, the Spanish Wikipedia revolts and copies their entire Wikipedia into a new online encyclopedia: Enciclopedia Libre. The Spanish Wikipedia's growth will never fully recover from this bifurcation. The Japanese Wikipedia culture is unique. It is by far the most polite. Edit wars are unknown. And, all Wikipedians remain anonymous. They don't register usernames. This makes it harder to get mad at "no one" and makes it easier to reach consensus. The German Wikipedia is more rigorous. Articles don't get readily published until they are "Sighted" by senior editors who check spelling, absence of vandalism, and some of the basic facts. As a result, Wikipedia has more credibility in Germany. And, they have formed a close cooperation with several government agencies related to education. Their culture resembles what Larry Sanger had in mind with Citizendium. The Chinese Wikipedia is interesting due to its challenges of having to deal with different icon styles. Zhen Zhu designed a software that translates Chinese articles in six different icon styles. This software will be leveraged by the Serbian and the Kazakh Wikipedias to generate translations in their respective Cyrillic, Latin, and Arabic versions.
The international coverage of Wikipedia is incredible. There are Wikipedias in 255 different languages. And, many are surprisingly prolific. As of March 2008, the Esperanto Wikipedia has over 95,000 articles; the Catalan one over 106,000; Sanskrit over 4,000. But, English dominates with over 2.2 million articles which is three times larger than the German one in second place.
In 2004, Wales creates an arbitration committee. Over a 2 year period, it will handle more than 200 cases. Nevertheless, over the years many editors tired of fighting trolls leave Wikipedia suffering from burn out as Sanger did much earlier.
Near the end, the author raises the issue: is the English Wikipedia actually finished? With over 2 million articles, there is little need for new ones. Now, the mission of the Wikipedians is changing from creation to maintenance. Lih suggests the creative types may find interesting outlets in Wikibooks, Wikiuniversity, and other derivatives of the English Wikipedia.
If you want to study further the implications of online free collaboration you may want to read
Wikinomics: How Mass Collaboration Changes Everything. If you want to study further the mechanical workings of Wikipedia, read
How Wikipedia Works: And How You Can Be a Part of It or
Wikipedia: The Missing Manual.
