Engineers at Stanford University have invented a new technology that would give broadband customers more control over their pipes and, they say, possibly put an end to a stale net neutrality debate in the U.S.
The new technology, called Network Cookies, would allow broadband customers to decide which parts of their network traffic get priority delivery and which parts are less time sensitive. A broadband customer could then decide video from Netflix should get preferential treatment over email messages, for example.
The technology could put an end to the current net neutrality debate focused on whether broadband providers are allowed to prioritize some network traffic and block or degrade other traffic, said the researchers, Professors Nick McKeown and Sachin Katti and electrical engineering grad student Yiannis Yiakoumis.
Network Cookies, first described at a conference in Brazil in August, would put broadband carriers and web content providers on a level playing field when catering to user preferences, they said.
The technology puts the control in the hands of broadband users, Yiakoumis said by email. "Giving users choice is both feasible and beneficial," he said.
The technology adds transparency and "audit-ability" to network management processes, he added.
It's unclear whether broadband providers would support the new technology, however, given that Network Cookies would take some network management authority away from them. A spokeswoman for USTelecom, a broadband provider trade group, declined to comment on the proposal.
Yiakoumis defended the technology, saying it would be useful for broadband providers. "ISPs can differentiate from competition and better engage with their customers, in a way that is compatible with net neutrality principles," he said.
In addition, broadband providers would still control their infrastructure, and they could decide to provide a low-latency service, a high-bandwidth service, or free data with lower bandwidth, as some options, he said.
"They decide what network services to provide, with what guarantees, and how to implement these services within their network," he added. "But users can decide for themselves how to use these services and with what traffic. I believe this approach is easier for ISPs as well: Instead of trying to guess what users want they can let them choose directly for themselves."
But what if some broadband customers designate all traffic as high priority?
Some users may need incentives for proper use or be required to pick among applications, Yiakoumis said. ISPs could also use monthly quotas on high-speed connections or charge extra for low-latency services, he said.
"This is not very different from we what we have today," he said. "You can buy a 10Mbps or a 50Mbps connection plan. All we say is that it's OK to have both of them at the same time and decide what traffic to send where."
The researchers field-tested Network Cookies in home settings by working with Google. Users in 161 homes were given access to an application called Boost, and they decided to give priority to websites related to news, video, voice, and sports.