Remembering Jon Postel — And the Day He Hijacked the Internet

One January day in 1998, Jon Postel emailed eight of the 12 organizations that served as address books for the entire internet. He told them to reconfigure their computer servers so that they pulled addresses not from a government-backed operation in Herdon, Virginia, but from a machine at the computing facility he helped run at the University of Southern California. And they did.

One January day in 1998, Jon Postel emailed eight of the 12 organizations that handled the address books for the entire internet. He told them to reconfigure their computer servers so that they pulled addresses not from a government-backed operation in Herndon, Virginia, but from a machine at the computing facility he helped run at the University of Southern California. And they did.

According to news reports at the time, Postel made the switch without approval from anyone. Some said it was merely a "test" meant to show that the internet's directory infrastructure could be repositioned as needed. But others said that Postel was making a statement -- that he was trying to show the White House that it couldn't wrest control of the internet from the widespread community of researchers who had built and maintained the network over the previous three decades. The White House was just days away from revealing a plan to reorganize the way the internet's directory system was governed.

Whatever Postel's intentions, the incident shows what a key figure he was in the rise of the internet. In 1969, he was part of the team at the University of California, Los Angeles that set up the first node on the ARPAnet, the research network that eventually morphed into the modern internet. He helped define the protocols that underpinned the network. And for years, he was the network's primary administrator, overseeing not only the RFCs -- the documents that define the operation of the internet -- but also the Internet Assigned Numbers Authority, or IANA, the organization that handled the net's naming system from the late 1980s through the late '90s.

Nine months after that January day when he redirected eight of the internet's "root servers," Postel passed away due to complications during a heart operation. But his place in internet history lives on. This past April, he was inducted in the Internet Society's Internet Hall of Fame alongside such names as Vint Cerf, Tim Berners-Lee, Ray Tomlinson, and Leonard Kleinrock.

The RFCs, or Request for Comments, are essentially the instructions for the internet. They were created in 1969 by Steve Crocker, another member of the UCLA that sent the first internet message, but when Crocker left UCLA soon after, he asked Postel to take over the editing of these documents. He did -- for the next thirty years -- and this bit of bookkeeping gradually expanded into a much larger administrative role.

Postel helped create the Stanford Research Institute's Network Information Center, which passed out network addresses on the ARPAnet in the 70s and 80s, and he later founded the IANA, which replaced the Network Information Center after the internet adopted the Domain Naming System, or DNS, which democratized the way the network handled addresses.

But Postel wasn't just an administrator. He was also a researcher who helped built the internet as we know it, authoring or co-authoring over 200 RFCs. "Without his contributions to the network design -- completely separate from any of the administrative or 'editorial' stuff -- I'm convinced that we would have no network today," say John Klensin, another Internet Hall of Fame inductee.

Vint Cerf and Bob Khan were the original designers of TCP/IP, the fundamental protocols that define the internet. But according to Klensin, Postel wrote many of the specifications that actually allowed the world to use those protocols.

He was one of many people who built the network -- at the lowest level -- but he was also someone who took the lead. "He was a real network researcher. But at the same time, he felt a lot of responsibility for what was going on with the network [as a whole], and that resulted in him volunteering for stuff that no one else was volunteering for," Klensin says.

"Jon tended to figure out where he needed advice or calibration and where to get it, but also tended to tell people that he had to be the final [or] official decision-maker, that he would have to take the heat if things went wrong, and that he thought he would get better advice if those advising him could be kept out of the spotlight and protected from any public attacks or fallout," Klensin says.

Here at Wired Enterprise, we know first-hand about Postel's commitment to the internet cause. Back in 1997, one of the reporters, Robert McMillan, was working for another publication when he emailed Postel about upcoming changes to the internet's naming system. Postel responded, but only to berate McMillan for using a digital signature with his email message. The signature, he said, was just a pointless way of clogging up internet bandwidth. He never did answer McMillan's question.

Despite the controversy that surrounded Postel's redirection of those eight root servers, he was slated to serve as chief technology officer of ICANN when he passed away. In 2011, Steve Crocker was named chairman of the board at ICANN, and in a way, this finally tied up a loose end of internet history.

"[Jon Postel] died almost at the moment that ICANN was formed," Crocker says. "It was unfortunate. But there’s a legacy that lives on, that’s rooted through him, all the way back to the earliest days of the ARPAnet. And in a certain sense, I provide a bit of continuity, reaching all the way back to that time."

Additional reporting by Robert McMillan

Image: Internet Hall of fame