The governance debate over the last year or so has largely boiled down to two camps. The U.S. and Western governments endorse a multistakeholder model, one in which Internet policy is set collectively by representatives from technology, business, and public policy sectors. On the other hand, Russia, China, and other nations favor turning control of the Internet over the United Nations under the auspices of the International Telecommunications Union. The ITU option would put world governments in control of the Internet, with each nation having one vote.

While the “one nation, one vote” model may sound healthy and democratic, putting governments in charge of the Internet could have troubling consequences, Savage says. “The Internet’s capacity to empower individuals has caused a lot of governments to become alarmed. If they have the chance, many governments — potentially a majority of them — would prefer to limit the free flow of information and ideas. That would have negative consequences for individuals, businesses, and the scientific community.”

The best option for maintaining a free, open, and robust Internet, Savage argued, is a multistakeholder model of some kind. The problem, however, is that the concept of “multistakeholder” is often loosely defined.

“It’s a very fuzzy idea,” Savage said. “Advocates have expressed a desire for decisions to be made in a way that’s open, transparent, and inclusive, but beyond that there’s little conception of what a multistakeholder model would actually look like.”

Further complicating matters is the huge breadth of issues involved in governing the Internet. For some, Internet governance refers to the technical operations of the network, including traffic routing and domain names. For others, it means regulating content, combating cybercrime and terrorism, or protecting freedom of expression online.

In a recent paper, Savage and co-author Bruce McConnell outlined some broad recommendations for what shape multistakeholder governance might take. First, they suggested not conflating all aspects of Internet governance under one organization.

“The likelihood of reaching agreement increases if we can simplify the landscape,” Savage said. “That means disaggregating governance into a small set of important issues.”

In many cases, he argues, existing international organizations could be brought to bear in dealing with nontechnical public policy issues related to the Internet.

“So many of the issues people are looking to govern predate the Internet — crime, human rights terrorism, et cetera. For human rights, there exists the Human Rights Council. For crime, there exists Interpol. We think that these existing organizations should deal with those governance matters that fall within their respective areas.”

One problem is that many of these organizations might not be up to speed on the Internet. But that’s easily remedied, Savage says. Each group could assemble a multistakeholder advisory panel of Internet experts to help guide decision-making.

Technical issues including network architecture, domain names, and the like, could also be handled by a model that involves multistakeholder oversight, Savage says. But care must be taken to make sure that technical decisions are ultimately made by those most qualified to do so.

For example, the Internet Corporation for Assigned Names and Numbers (ICANN) has successfully expanded the domain name system to more than 3 billion users. There’s no reason it should not continue to do so, but adding a properly crafted multistakeholder oversight panel could bolster ICANN’s legitimacy. Such a panel should operate on the principle that policymakers can offer opinions concerning technical decisions, but that those with technical expertise should make final decisions on them.

“We think that a model of decentralized governance with multistakeholder oversight is a plausible structure for governing the Internet,” Savage said. “It’s a way of strengthening the multistakeholder model, which we think is key to maintaining a free, open, safe and robust Internet.”