We don’t need a backup plan for the internet, says Vint Cerf

Quite often I walk into my home, flip on the light switch, see the room suddenly illuminated and, in my mind exclaim: “It works!”

I don’t need to know every detail of the massive infrastructure that’s been installed between my finger on a plastic switch and the massive electro-grid and dam up north to appreciate how fortunate we are to be so comfortably nestled in the modern era. But I think about it sometimes.

Vint Cerf is one of the founders of the Internet, if anyone has a right to claim such a title, and he described the same feeling last week at a TED conference in California.

“Every time I load up a web page, I think, ‘Holy crap, it actually worked,” said Cerf. The audience, whose attitude towards a non-functioning web page might have been something like Louis C.K.’s bit about a pissed off airplane seat mate, laughed at the fact that he could be astonished.

Vint Cerf was rebutting an earlier TED talk by inventor Danny Hillis, who is making the case that the Internet needs a Plan B, in case something goes horribly wrong. The system is vulnerable and “there are a lot of bad guys on the Internet these days,” Hillis warns. Worried more that people will fret about the damage done to a bunch of infected computers and the other stuff that’s damaged in an attack (hydroelectric grids, air traffic control, the banking system, etc.), Hillis challenges us to focus instead on defending the Internet itself as a communications medium.

For Cerf, the very scale of the Internet, its decentralized nature, is a feature of its resilience.

Reminiscing about the state of the Internet in 1982, Hillis paints a picture of a system that was built on trust, mentioning the irony that “such a Communist principle was the basis of a system developed during the Cold War by the Defense Department.” Ignoring the fact that merely being “nice” or trustworthy has never been a defining characteristic of any actual Communist, there are other issues with Hillis’s solution to a very real problem.

“When you’re taking a plane at LAX, you really don’t think you’re using the Internet. When you’re pumping gas, you really don’t think you’re using the Internet,” Hillis tells his audience, finishing it off with the very real threat that, “We’re setting ourselves up for a disaster like the disaster we had with the financial system. We’re taking a system which is basically built on trust, was basically built for a smaller-scale system, and we’ve kind of expanded it way beyond the limits of how it was meant to operate.”

Again, a problem of definition: anyone who has ever held the concept of the banking system in their mind while managing to utter the words “built on trust” must be a unique specimen.

Cerf replied to Hillis by acknowledging that there are several very real risks, without mentioning that the largest of these are likely state sponsored. “It’s absolutely true that the Internet is getting bigger and bigger, is going to be in just about every appliance we can think of. Now, the question is, does that mean it’s all going to collapse? I don’t think so.”

Cerf: (The Internet) wasn’t designed to do anything in particular. And that’s why it’s been able to do almost anything we can think of, to program…It’s not a static system. And that’s why it thrives every day and will continue to do so, even post a major deliberate attack.

For Cerf, the very scale of the Internet, its decentralized nature, is a feature of its resilience, pointing out that there are always malware and targeted attacks coursing through the veins of the system. And yet it still works, well beyond the scalability that Hills says it was “designed” for.

A large-scale denial of service attack will likely be pointed at something quite specific. Cerf explains the Internet in the following way, “It wasn’t designed to do anything in particular. And that’s why it’s been able to do almost anything we can think of, to program.” It’s not a static system. And that’s why it thrives every day and will continue to do so, even post a major deliberate attack.

For what it’s worth, Cerf concedes that he’d be happy to consult on the idea of cooking up a Plan B with Hillis, even as he wonders whether it’s necessary. But he offers up a couple of alternatives. “I believe it would be prudent to do two things. One of them we’ve already started doing, this is called ‘clean sheet’. The question is, what if we started over and redesigned the network? What would we do differently?

“The second thing, I think, is to ask the question, what could I do to create a communication environment that’s even better than the Internet? I don’t know what the answer is, although it might turn out to be quantum communications, which the physicists told us are not possible.”

A lot of things that didn’t seem possible in the 20th century, even from the perspective of the scientific community, are commonplace today. History tells us that Cerf is safe in musing that what seems impossible today will be routine in the decades to come.

Below: Vint Cerf at TED. PS: If the link below is broken, we recommend contacting sysadmin.