Is it feasible to peer-review the Internet? A coalition of science agencies and Silicon Valley entrepreneurs is trying to do just that. They are launching what they claim will be an authoritative network of websites, where users can find trustworthy information on any subject. Top science organizations are signing up, but critics are sceptical about the project's rationale, and whether it can succeed.

The Digital Universe project is billed as a “network of web portals”, run by experts, on topics ranging from the Arctic and the oceans to the Solar System and the Universe. Users would navigate through the portals using a three-dimensional browser.

You could “fly over an accurate virtual Earth to explore the contours of the Grand Canyon, swim with the fish of the Great Barrier Reef and travel through the human body”, says an enthusiastic Robert Corell, chair of the steering committee for the Arctic Climate Impact Assessment and senior scientific adviser for the Digital Universe's Earth Portal.

The project also includes an encyclopaedia that will use similar technology to the popular online encyclopaedia Wikipedia, and Larry Sanger, a co-founder of Wikipedia, is helping to create it. But that's where the resemblance ends. All content in the Digital Universe will come from vetted experts, and articles will be reviewed by editors before going live. There will also be links to approved websites.

The Digital Universe: science you can trust on the Internet.

The driving force behind the project is ManyOne, a company headed by Joseph Firmage, who made a fortune in the 1990s from an Internet consulting company. He resigned in 1999 after the fallout from his book claiming that he had encountered extraterrestrials.

Firmage says he vehemently opposes the “anyone can edit” vision of Wikipedia. “Wikipedia is a very uninviting place for most intellectuals,” he adds. “I myself would not want to be writing articles that could be edited by somebody who does not necessarily have any expertise.” He hopes that peer-reviewed content will raise the standard of content on the web, which he describes as having “an intellectual deficiency of serious proportions”.

“The Digital Universe is an attempt to massively mobilize the scientific community,” adds Cutler Cleveland, director of the Center for Energy and Environmental Studies at Boston University and editor of the Digital Universe's Earth Portal and other portals on Earth's environment. “The information you see here you will know is trustworthy in a fundamental way.”

They're not in it for the money; actually, they're trying to save the world.

Many are enthusiastic. The US National Council for Science and the Environment, the University of California, Berkeley, Massachusetts Institute of Technology, the University of Oxford and the World Resources Institute have all signed up to help. The Digital Universe's Earth Portal, which is to be released next month, has contributors that include Gene Likens, the discoverer of acid rain, Thomas Kunz, a top bat expert, and Robert Costanza, founder of the field of ecological economics. Its international advisory board includes Rita Colwell, former director of the US National Science Foundation.

Critics interviewed by Nature were unwilling to speak on the record. But some believe that the project is over-complicated, and that much of its underlying technology — which still requires significant development — runs against the trend to distribute information in lightweight formats that can be accessed by cell phones or PDAs such as the BlackBerry. “If you have to rely on a high-bandwidth always-on network environment, on devices with a lot of storage, you are pretty much going in the wrong direction,” says one critic, an expert in Internet information systems. He is also unimpressed by the Digital Universe's concept of peer-reviewing material. “There's more than enough content on the web, even substantive content,” he says. “I'm not sure that generating new content is really a breakthrough.”

There are also questions over the business model, in which revenue would largely come from selling high-speed Internet access, with half the profits fed back into the work. “It's an odd choice; that's a dying business,” comments one observer familiar with the project, pointing out that in the future consumers will be unlikely to notice where their Internet access comes from. But he says he can't help being inspired by the idea. “They're trying to package science in a way that has some of the glitz and entertainment appeal of television, but that is also complete and correct,” he says. “They're not in it for the money; actually, they're trying to save the world.”