Creating the Digital Universe

January 19, 2006

Correcting Some Misconceptions about the Digital Universe

Beginning with this article, $10m for a Wikipedia for grown-ups, from The Register, a number of misconceptions about the Digital Universe have been passed around. Here are the facts:

The Digital Universe will not be an encyclopedia only. It will include a wiki-based encyclopedia project, but it will include much else: an expert-managed Web directory, 3D content, forums, news, books, video and audio archives, and in general will evolve into an authoritative gateway to the best of the Web. If we are successful, the Digital Universe will be a competitor not of Wikipedia, but a nonprofit and expert-driven alternative to Yahoo, Google, AOL, MSN, etc.

I am not founder of the Digital Universe. You must have me confused me with my employer, Joe Firmage, CEO of ManyOne Networks and one of the founders and leading lights of the Digital Universe. The vision of the Digital Universe is brilliant but it is most definitely Joe's, and it's a vision he had long before he hired me to help him carry it out. I started corresponding with the founders of the Digital Universe, Joe and my immediate director, the astrophysicist Dr. Bernard Haisch (President of the Digital Universe Foundation), in January 2005. The idea for the Digital Universe existed long before that. Moreover, ManyOne has dozens of other employees, including some world-class people.

I did not collect $10 million to pay for Stewards (editors). Again, you must have me confused with Joe Firmage, who is among other things an amazingly talented fundraiser. Those millions have mostly already been invested in ManyOne Networks, which has allowed construction of our world-class organization and a brand new, cutting-edge Web platform, as well as the development of content and the launching of the Digital Universe. It is the intention of the Digital Universe Foundation to acquire funding (from ManyOne Networks, philanthropic organizations, government grants, etc.) which will be used to support the stewardship program.

The Digital Universe is a nonprofit, free (open content), and ad-free project. The Digital Universe Foundation has filed for nonprofit status. The nonprofit ManyOne Foundation will become the sole owner of ManyOne Networks as investors are paid off. The Digital Universe will be creating vast amounts of content, including the entire encyclopedia, and a whole lot of other stuff, that will always be free of charge (i.e., the Digital Universe Foundation will recommend that the DU's managing "information coalitions" use a Creative Commons license). Finally, the project will always be ad-free. Our business model does involve subscriptions for some "premium" content (not to say the encyclopedia, e.g., won't also be quite excellent in spite of its being free) and an ISP service, but this too will be ultimately driven by and for the nonprofit, noncommercial, world-benefitting purposes of the Digital Universe.

The Digital Universe is not Nupedia. How will it be different? Let me count the ways. (1) It will be more than an encyclopedia (see above). As to the encyclopedia component: (2) It will use a wiki for article development. (3) It will be quite simple; it will certainly not have seven steps. (4) Ordinary uncredentialed people will be able to start articles in the public area without editor approval. (5) There will be an article rating system for public-area articles. Also: (6) The Digital Universe, or more specifically, ManyOne Networks, has a business model that has an excellent chance of funding the work of experts (Nupedia had no such model)--although we cannot make any commitments yet, as the subscription service is just now getting started. (7) The Digital Universe is led by Joe Firmage and Bernard Haisch, not by Jimmy Wales; unlike Jimmy, Firmage and Haisch have from the start been absolutely committed to a vision of an expert-led Web. They will not abandon expert contributors, as Nupedia's leaders were abandoned. Finally, (8) the Digital Universe is post-Nupedia and post-Wikipedia. I was involved in the founding of both projects, and I understand as well as anyone what the strengths and weaknesses of both projects were and are. I am helping to design the Digital Universe having learned from this experience; so, as far as I am concerned, you can expect the Digital Universe to be better than both projects. It will replicate their strengths and jettison their weaknesses.

One last remark on this point: I am quite sure that the reason that people say that the Digital Universe is "just Nupedia revived" is that they want it to fail, as Nupedia failed--although, as I have argued at length, it is more accurate to say that it was, irresponsibly, allowed to wither untended. Many of those who suggest that the Digital Universe is no more than Nupedia believe that any project in which experts have authority must fail, because, from now on in their eyes, only radically egalitarian projects like Wikipedia can possibly succeed. Of course, this sounds ridiculous, but it is how some people appear to think. And, of course, this viewpoint is wrong. More generally, what very many Web 2.0 thinkers fail to realize is that it is possible to distinguish openness, collaboration, and public participation on the one hand, from radical egalitarianism and near anarchy, on the other. The Digital Universe will demonstrate that it is possible to have the former without the latter.

I (and probably the rest of the leaders of the project) believe the Digital Universe will have the lowest error rate in history. A recent Nature study compared articles from Encyclopedia Britannica and Wikipedia on 42 different scientific subjects, and showed that while the former had on average about 3 errors or omissions, the latter had only 4. For the record, I don't think that one study in one general subject area, measuring just one type of metric, can possibly establish what many people have said it establishes (viz., that Wikipedia is nearly as reliable as the Britannica). Nevertheless, a few people have used the study to argue that there is no need for a Digital Universe encyclopedia, and even to suggest that the quality of Wikipedia articles will outstrip that of Digital Universe encyclopedia articles. This suggestion is incorrect, as I will explain.

The Digital Universe encyclopedia will feature a very similar wiki-based editing system; like Wikipedia, and unlike the Britannica, many hands will be able to keep articles maximally complete and correct. But unlike Wikipedia, the project will be led by experts, who will have the final say in how articles presented to the public will read. A strongly collaborative project like this, led by experts, has the very best chance, in my opinion, to absolutely eliminate identifiable errors. After all, the scientists who reviewed articles on behalf of the Nature study--who determined whether something was an error in the first place--are exactly the sort of scientists who will be involved in the Digital Universe. Moreover, unlike Britannica articles, Digital Universe encyclopedia articles will be open to editing by multiple scientists and scholars, in real time. It seems highly likely that, if the Digital Universe succeeds in attracting adequate expert participation, it will have the lowest humanly-identifiable error rate in history: that is the beauty of the idea of marrying strong collaboration with expert participation.

Dr. Larry Sanger
Director of Distributed Content Programs
Digital Universe Foundation