MIT to train students to build a better Web

IDG News Service - A group of professors has formed a research collaboration to train students how to design future versions of the World Wide Web.

One of their first lessons will be how to strike a balance between better access to data and stricter rules about its use, said researchers from the Massachusetts Institute of Technology (MIT) and England's University of Southampton at an MIT news conference today.

The Web Science Research Initiative (WSRI) hopes to create a college degree program in "Web science" that combines disciplines including computer science, mathematics, neuroscience law and economics. It will also raise funding for doctorate students to study at MIT and the University of Southampton.

"Because we created the Web, we have a duty to understand it," said Tim Berners-Lee, a scientist in MIT's Computer Science and Artificial Intelligence Lab (CSAIL), director of the World Wide Web Consortium (W3C), and inventor of the protocols that control the Web. "Suppose among all the beautiful, wonderful things it's created, it also creates something horrible?"

The group created WSRI (pronounced "whiz-ree") because of two powerful trends -- the burgeoning amount of online information and the need for better social rules to control access to it. The number of Web servers has now reached 100 million, said Rodney Brooks, a founding member of WSRI and the director of CSAIL.

"There are 100 million servers, 100 million nodes, 100 million Web masters and mistresses, so when it comes to aggregating data, how do we do it? How do we get to a crisp mathematical question that can inform policy?" Brooks said.

One of the Web's greatest strengths is that the same rules govern small sites and enormous communities such as MySpace, Berners-Lee said. That "fractal" quality has supported the growth of the blogosphere, a complex space based on simple principles. Yet, the Web's growth has hobbled its evolution in some cases. For instance, Wikipedia has recently applied new rules for erasing spam postings instead of hewing to its original philosophy of "anybody can write," he said.

One of the first challenges WSRI faces is raising funds from private sources while sharing its research with the public. Companies including Google Inc. and IBM Corp. have applauded the effort and WSRI's mandate to share its research on an open-source basis, said Wendy Hall, another WSRI founder and a computer science professor at the University of Southampton.

Another challenge will be applying theory to real-world situations. The group is working with the British government to find a "policy-aware" model for sharing federal research such as census data and demographics in socially acceptable ways. Other applications could include ways to share information about clinical drug trials, real-time epidemiology to track the spread of a contagious disease and sharing counterterrorism databases while protecting citizens' privacy.

In this age of file-swapping tools, some of them illegal, finding a common definition of social concepts may prove to be a stiffer challenge. The issue is not a lack of laws, but how poorly they are represented on the Web, said Danny Weitzner, a W3C researcher and co-founder of WSRI.

"With all respect to Tim, who did a nice job with the Web in the beginning, there were some things left out," Weitzner said. "The Web is very bad at maintaining the social expectations and the legal standards and rules that we rely on in society."