Larry Sanger seems to have a thing about free online encyclopedias. Although his main claim to fame is as the co-founder, along with Jimmy Wales, of Wikipedia, that is just one of several projects to produce large-scale, systematic stores of human knowledge he has been involved in.
Sanger's love of philosophy and epistemology - the study of the nature and scope of knowledge - began around the age of 16. His PhD, from the University of Ohio, concerned "epistemic circularity", and it was as this was nearing completion that he started thinking about less abstruse matters - like earning money.
In 2000, he drew up a business proposal and sent it to a few people, one of whom was Wales, whom he had known from philosophy mailing lists in the mid-90s. "He saw that I was essentially looking for employment online and he was looking for someone to lead Nupedia ... Nupedia wasn't there at all, in fact it was just the vaguest of ideas. Jimmy was frustratingly brief in his instructions, he simply wanted an encyclopedia that everyone could contribute to and that would be released under what he called an open content licence."
The basic concept came from the open directory project Dmoz (http://dmoz.org), short for Directory Mozilla. This volunteer effort to create a free version of Yahoo's hierarchical listings began in 1998 under the name Gnuhoo - which was inspired by GNU/Linux - before turning into Newhoo. It was acquired by Netscape and released as open content.
One of Sanger's tasks was to recruit academic volunteers to write, edit and organise entries. All the information was carefully peer-reviewed - creating an entry was a seven-step process. This ensured rigour but also throttled the rate of production. This problem was evident as early as the summer of 2000, a few months after it had been designed.
Community solution
While casting around for a way to speed up article production, Sanger met with Ben Kovitz, an old friend, in January 2001. Kovitz introduced Sanger to the idea of the wiki, invented in 1995 by Ward Cunningham: web pages that anyone could write and edit. "My first reaction was that this really could be what would solve the problem," Sanger explains, "because the software was already written, and this community of people on WikiWikiWeb" - the first wiki - "had created something like 14,000 pages". Nupedia, by contrast, had produced barely two dozen articles.
Sanger took up the idea immediately: "I wrote up a proposal and sent it [to Wales] that evening, and the wiki was then set up for me to work on." But this was not Wikipedia as we know it. "Originally it was the Nupedia Wiki - our idea was to use it as an article incubator for Nupedia. Articles could begin life on this wiki, be developed collaboratively and, when they got to a certain stage of development, be put it into the Nupedia system."
Things didn't quite work out that way. "The editors and peer reviewers of Nupedia, mostly professors and other professionals, looked at the wiki tool and didn't want anything to do with it." The idea was too revolutionary. As Sanger points out: "It actually took the success of Wikipedia" - as Sanger later baptised it - to make the idea plausible to a lot of people."
Wikipedia grew quickly - perhaps too quickly. There was an influx of visitors driven by two postings on the Slashdot news site who were not clued in to the Wikipedia culture, recalls Sanger. "They tended to be quite a bit more anarchist and egalitarian" than the original Wikipedians. This led to vandalism of pages, and increased bickering about issues such as control and quality. Then, in 2002, Sanger left Wikipedia. In the wake of the dotcom crash, the company that owned and ran Nupedia, Bomis, was suffering too, and Wales had to lay off most of the staff, including Sanger. Wikipedia was henceforth run by Wales and the Wikipedia community.
Sanger began teaching philosophy at Ohio State University. But he kept a close eye on Wikipedia. He saw the same problems he had observed in the early years, but now writ large. Although an outsider, he felt compelled to offer his thoughts in an essay, Why Wikipedia Must Jettison its Anti-elitism. As well as provoking a rebuke from Wales - who wrote "Larry's comments betray a complete ignorance of the project" - it had another, unexpected consequence: a job for Sanger on a new online encyclopedia.
Expert guide
Two people who read Sanger's article were Joe Firmage and Bernard Haisch, respectively founder and president of the non-profit Digital Universe Foundation. Firmage made a fortune during the dotcom boom from a web consultancy, before becoming notorious for his self-published book The Truth, in which he described his encounter with aliens. Haisch is an eminent astrophysicist.
As Sanger explains, the Digital Universe project, funded in part by Firmage, hopes to become "a non-commercial competitor of the big web portals such as Google, MSN and Yahoo. It would essentially be a [free] expert-selected guide to the best of the web in every subject. Unlike those other portals it would focus on the reliability of the information it was linked to." In a way, it is a throwback to the Dmoz directory that inspired Nupedia.
The main Digital Universe portal (www.digitaluniverse.net) is made up of many subsidiary portals. Each portal categorises weblinks by type, and there will be related content such as video and audio streams. Potentially, each portal will exist in multiple versions for different languages.
There will be an encyclopedia for each portal, and that's where Sanger comes in. The encyclopedias will be built collaboratively using MediaWiki, the software that runs Wikipedia, although the presentation will be different. The wiki will be run by acknowledged experts - the "elite" that Sanger feels Wikipedia has turned its back on.
Sanger recognises that non-specialists can also make useful contributions, perhaps through parallel wikis. "The public can work in a workspace and the results ... are made available" to the main encyclopedia - similar to the approach used in the days of the Nupedia wiki.
Sanger is also starting a collaborative project called Textop (Text Outline Project), which aims to create "a central outline of human knowledge based on specific details of definition, explanation and argument contained within scholarly texts". He believes this "will facilitate research and learning in a way never before possible", because it can be used "to organise all sorts of textual information, such as a dictionary, a guide to debates and summaries of recent events".
Textop is not part of Digital Universe and is running on a server space that Sanger pays for. This highlights perhaps Digital Universe's biggest challenge: finding the money to sustain its vision.
Curriculum vitae
Age 37
Education
1986-1991 BA philosophy, Reed College; 1992-1995 MA philosophy, The Ohio State University; 1995-2000 PhD, OSU.
Career
1992-1996, 1997-1998 Graduate teaching associate, OSU; 2000-2002 Editor-in-chief, Nupedia; Co-founder and "chief organiser", Wikipedia; 2002-2004 Instructor, Columbus State Community College; 2004-2005 Instructor, OSU; 2005-January 2006 Director of distributed content programmes, Digital Universe; 2006- Director of collaborative projects, Digital Universe.
Hobbies Traditional Irish fiddle music
· If you'd like to comment on any aspect of Technology Guardian, send your emails to tech@theguardian.com