In an ironic digital twist, the US federal government - the driving force behind the creation of the Internet - has not been able to fully embrace the second incarnation of the Internet, according to a study released late last month.

Hampered by a slew of legislative requirements for publishing documents and other factors, federal agencies have been relegated to the sidelines of a Web 2.0 world where users are taking raw data from feeds and other sources and services to create focused content, according to the Princeton University study.

The report, from the university's Center for Information Technology Policy, charges that the federal government focuses too much on its role as a Web site publisher and not enough on finding better ways to make the underlying data it creates available to citizens.

David Robinson, associate director of the center, said that instead of focusing on Web site publishing, the government should task agencies with making the data they generate available in RSS feeds, XML or other open formats so private sector operations can publish the data.

"[The government] has sort of wanted to make it a priority to get data all the way to the end user," Robinson said. "We have the sense that government could have greater success in making government data available and useful to citizens if it were to focus more narrowly on making available the underlying data. People have had to really struggle to recreate data sets that are behind some Web sites or some difficult interface that [was] erected with the best of intentions."

For example, the report - called Government Data and the Invisible Hand - notes that the Federal Communications Commission has a Web site based on a structure that has remained basically unchanged since 2001. To make any practical use of the system, users must know a docket number for a proceeding. In addition, materials on the site can only be searched by a few criteria, like the date of a submission or the name of a submitting attorney. The site does not allow users to search the actual content of filings, even when they have been submitted in a computer-searchable file format, the report says.

Another site, Regulations.gov , a government wide docket publishing system launched in 2003 and used by nearly all departments and agencies, was put into place with a limited search engine and limited browsing capabilities, the researchers found. However, it began releasing its underlying data in RSS to allow users to create alternative views of the data, which led to the creation of OpenRegulations.org. The site competes with the government side by offering "pared down, easy to navigate" listings, the report added.

"Each time a new law makes a new requirement for government Web sites, it sort of gets piled onto this bunch of other rules that already exist," Robinson added. "There doesn't really seem to be anybody who is able to look at these series of statutory mandates and make them reasonable as a group. Private actors have a lot more flexibility in how they can approach those questions. What we've seen from the technically literate community, from the group of Americans who know how to build Web sites and use data is that there are many people who are interested in sites that take advantage of government data."

If private entities were to play a role in publishing government data they could take advantage of technology like RSS feeds, links to information sources, mashups, discussion forums and wikis or collaborative filtering and crowdsourcing - technology that the government has been slow to adopt because of various statutory barriers, the report added.