INTERVIEW XI PART I “Sci-Non-Fi”

Vint Cerf

VP, Chief Internet Evangelist at Google

Co-Designer of TCP/IP Protocols

Technology Policy Developer 

[Recorded February 4, 2025 via Google Meet]

“We make our world significant by the courage of our questions and the depth of our answers.” - Carl Sagan

“In 1995, I had a career-changing opportunity to work as a young product manager alongside a team of amazing network and software engineers who were working to commercialize the Internet. The team was led by Vint Cerf, co-designer of the TCP/IP protocols and the architecture of the Internet. Vint had a mantra—‘IP on Everything’ or Internet protocol on everything. It captured the vision that someday, everything would be connected. This was a profound statement at a time when bandwidth was limited and access fees were charged per minute.

IP on Everything had a profound impact on me at the time. I began thinking about what this would mean in every aspect of our daily lives. Specifically, I began to think about how it would impact the ways we live and what it would mean for housing. This started a 30-year quest to figure out how to apply innovation to housing and the broader built environment. I quickly became familiar with the technology adoption curve and the diffusion lag that often can last decades, referred to as the productivity paradox.

Fast forward to today where we see a true transformation in our daily lives. The single biggest driver is the paradigm shift from life span to health span. How and where we live has a profound impact on our health span. There is a significant opportunity to leverage technology, specifically AI to keep us healthy and deliver healthcare at home. The challenge is to accomplish this equitably so that everyone can benefit.

My work has come full circle as I focus today on integrating health and healthcare into the home based on Internet protocols. I am grateful to Vint who answered my email three decades ago, when I sat in a hotel room in Seoul and reached out via email. I read an article in Fortune Magazine on my flight about the team that was being assembled to create internetMCI. Two months later, I was part of that team, and the rest, as they say, is history.

The Third Chapter Curious team had a chance recently to talk with Vint about his current work and how his projects today have come full cirlce from his early days in the field. We hope you enjoy part one of this interview recorded and written by Cameron Thompson.”

-Bruce Thompson, March 19, 2025

INTERVIEWER: 

“What stands out to you as a memorable project you worked on that remains relevant to your purpose today? What did you predict your work would lead to at the time and what has changed or remained true about your predictions?” 

VINT:

“I worked on the Apollo program that eventually led to landing on the Moon in 1969, specifically on the F-1 engines for the Saturn V rocket. My job was to analyze the data that we got back from testing the engines in the static jig to see whether they would survive until they ran out of fuel, after which we didn't care—unlike Moscow, who likes to reuse the first stage which of course, makes good sense.

The reason I bring this up is, I was 18 at the time in 1961. I was thinking by 1981, 20 years from now, I’ll be twice as old as I am now. By that time, we'll launch rockets and spacecraft every couple of weeks, there will be people on the Moon, and all these other potential evolutions.

1981 was 44 years ago. It’s only now that we're watching these events unfold with some frequency, thanks to programs like Space X, and we are anticipating getting back on the Moon within the next year. Sometimes progress takes longer than you expect.”

INTERVIEWER: 

“Yes, such as in life, right? And I’m sure sometimes you have to pause projects and come back to them later because even though the ideas were fully developed and technological capabilities were available, the implementation may not have been possible due to funding constraints or because something more pressing needed immediate attention.

We have an interesting present and future with space exploration and advancement and I’m excited to see what we will do.”

VINT:

“This is a major topic for me right now, having started work on the interplanetary extension of the Internet in 1998. That was 27 years ago and is only now reaching fruition. Of course, we anticipated it would take 25 years or so to return to the Moon when we would have some serious commercialization involved.

The consequence of commercialization will be quite interesting, and transformative because NASA is stating its plan to purchase lunar rovers, habitats, mining output, and so on. But, you're not allowed to own any property on an extraterrestrial body, according to the 1967 Outer Space Treaty, which was agreed upon before we even managed to land on the Moon.

I'm anticipating that there will be a considerable amount of turmoil in the next decade as we start seriously commercializing near space and perhaps more distant planets: Mars and some of its satellites, possibly asteroids. Legal frameworks and international tensions will have to be resolved somehow.

The Secretary of the U.S. Navy asked me to opine on what challenges the Navy would anticipate over the next decade. I turned my crystal ball to the Marines because they are always the first to take the beachhead. I asked myself: what are the pieces of training and doctrines we should adopt for a spacefaring Navy? What's the role of the Marines, and what tactical strategies should we have? How do we communicate? What weapons will we need? It's a pity that we have to talk about weapons in space, but space has already been weaponized whether we like it or not. There are spacecraft capable of interfering with the operation of existing satellites.

Here we are in 2025, which sounds like a science fiction year. In a way, it is, except we're living in that science fiction now, or some version of it. I feel like I'm 18 all over again except I'm 81. My body certainly doesn't think it's 18, however, the same sense of anticipation is very notable and present.”

INTERVIEWER:

“I’ve heard that your mind’s and physical ages are very different. You'll feel young mentally if you keep active and dialed into what’s new. It's comforting to hear as a 31-year-old.

When you were around my age, you began developing the architecture of what we now call the Internet. I grew up hearing about your work with Bob Kahn, and how influential you were and continue to be in your field. My dad had a t-shirt from when he was product manager at internetMCI that said, ‘IP on Everything.’ Every time he would wear it I would laugh because as a kid, that cracked me up. He would tell me his stories about working in that environment in the mid-90s and what a profound era that was for technology. 

This is a technology still in use today. TCP/IP. From my research, it's such a solid system that is very unlikely to be replaced by anything else, a rare occurrence during a period when technology changes in an almost unbridled fashion.

Did you have an idea of the magnitude and scope your work would have today when you and Bob Kahn started developing these protocols?”

VINT:

“Well, the honest answer is no, and it's important to me that you appreciate the many people involved in making this protocol work. I don't want to take excessive credit.

In the beginning, Bob started the program and asked me to join him. We worked very hard on TCP/IP, but many other people motivated the split between TCP and IP for example, because they wanted to have real-time communication like we’re doing right now. It's interesting how our aspirations to do what we're doing now were quite present 40 years ago—there just wasn't much capacity then. Slow data rates, and not very rapid video, all of which was black and white.

The real miracle is that there were so many people who contributed to the architecture of the system which could absorb all of these ideas and allow people flexibility and freedom to add new layers of protocol. The fact that the system was amenable to change and evolution is quite astounding. The one layer that has not been very amenable to change and adaptation is IP. You can tell that is true from looking at the IPv6 adoption rate, which is still only around 43%.” 

[IPv6 refers to the latest iteration of Internet Protocol. The percentage indicates the usage of IPv6 on a global scale by various organizations.]

“Part of the reason for that percentage is we haven't needed it as much as we thought we were going to, because Network Address Translation and client-server architectures have allowed us to grow the system, even though we don't have enough justice for usage by every device that's on the net. I hang on to the fact that millions of people have contributed to this program—it doesn't simply run itself. Everyday, people ensure it's still operating, and they fix it when it breaks. It continues to evolve with new protocols and the institutions it has spawned are relatively functional.

We have the Internet Society and all of its chapters. We have the Internet Governance Forum, we've got the ICANN [The Internet Corporation for Assigned Names and Numbers], we've got the IETF [The Internet Engineering Task Force], and the IAB [The Interactive Advertising Bureau]—these are all institutions that appeared when they were needed, or we introduced them when we thought they were needed. They are very loosely coupled. Sometimes there are still edges and tensions, but the amazing thing is that it all sort of works. However, it also breaks and we're in a period now where the penetration is so high that we're becoming very dependent on this thing working. That's troublesome, partly because when it doesn't work, all kinds of cascade failures happen, yet because of the penetration, it's given people an opportunity to do harmful things and amplify their capabilities. That's a very disappointing fact. After our call, I’ll be writing a two-page piece on looking back on Rousseau and The Social Contract. The question is, what should the digital contract look like? There is now a Global Digital Compact, which the UN has recently adopted. It's a very high-level thing. Getting down into the details, I'm sure there will be grungy places to be dealt with, and I want to write about all of that.

50 years after conception and 40 years after the program went operational, it's still serving, and now we're anticipating this extension into a solar system-wide communication environment. It will have similar characteristics to the terrestrial model in that multiple parties should be able to run pieces of it and have it be interoperable. It's very cool.” 

INTERVIEWER:

“It is very cool. That system, as far as I can understand, is triangulated between Earth, the Moon, and the International Space Station?” 

VINT:

“We've been operating with prototype software on the Moon since 2004. We are using the most recent standard phone protocol, Signaling System 7, on the International Space Station. We’ve done testing with other spacecraft that NASA has lent us. ESA, the European Space Agency, JAXA, the Japanese Space Agency, and KARI [now called KASA], the Korean Space Agency, are collaborating on implementing and testing these interplanetary protocols.

The Artemis program and the Moonlight Initiative from Europe’s ESA will interoperate with the same set of protocols.

The Chinese and the Russians are running separate programs, and haven't been collaborating much. There are some interesting issues arising when you think about a spacecraft, a habitat, or a laboratory on the Moon. Picking a concrete example, one might want to run the standard TCP/IP protocols, or the more recent QUIC protocols, which are an improvement over TCP, on the Moon because there's a lot of software and it's all available and should work. The problem is that it won't work between planets. A few people are arguing that we can tweak the parameters, and somehow it'll work. I'm not willing to buy the idea that either TCP or QUIC will work with a 40-minute round-trip time. The network management has to be very different from what we have typically done.

We're trying to figure out how to make TCP/IP interoperable with distinct and separate internets that are not connected. They can't connect in the same way that every element of the terrestrial internet can talk to every other element. We have to go through the Bundle Protocol suite with some kind of bulk carriage, and it's certainly not going to be interactive. In science fiction movies where some guy is in orbit around Saturn, and he's calling NASA saying, ‘Houston, we have a problem’ in a real-time conversation, well, we know that doesn't work. Quantum entanglement does not give you instantaneous communication.”

INTERVIEWER:

“Earlier you touched on our reliance on technology and the Internet. Our grid system, much of which is outdated, vulnerable, old-world technology, is struggling to keep up with our exponentially growing demands and new tech, and the systems don't seem to work well with each other. Are there safeguards to prevent catastrophic grid issues and restoration protocols?

Not in a doom and gloom sense, but is such a failure an inevitability, or something that can be avoided with our current capabilities?”

VINT:

“Well, boy that’s a tough one. There are scenarios in which the grid fails and it's very hard to recover. The best example of that is the Carrington Event from 1859, which of course occurred when we did not have the same reliance on technology as we do today, but even then, it created a big mess for what little we did have, in telegraph systems and so on. There's a book called Lights Out, which is an example of grid failure on the extreme side, but it points out that most of our power generation systems are purpose-built in a location with non-standardized parts. Many of the parts are standardized, but the configuration in the design and the size of the generators are often very site-specific. If we had a massive Carrington-type event and the grid had to be re-implemented, it would be a hard, long push and it would be really ugly. 

Some of us have been asking ourselves a slightly simpler question: Can we cold-reboot the Internet? What would happen if everything went down, and all the tables were wiped out? How do we get the thing up and running again? It's not a hundred percent clear what steps you can take in anticipation of that situation. I worry a lot about the grid. Some people were hoping that microgrids would allow for some sort of distributed production of energy and that they could be disconnected from each other so they don't have the same cascade failures that you see in so many of the major outages.

Take the Northeast blackout in 2003, for example, or the Southwest blackout in 2011 where a mistake in the maintenance procedure in Arizona propagated to the West coast in San Diego and Tijuana, up to Los Angeles, and the grids were down for two or three days. It was a huge flap, to say the least. The trend right now is geared towards local generation, whether it's solar power or wind power and there's a lot of interest in Small Modular Reactors as well.

I think the microgrid idea sounds attractive. However, there are people whose opinions I trust, who are very skeptical of microgrid systems being successful. I'm not that pessimistic yet. I remember working for a couple of years on something called a smart grid, which the Department of Energy and the National Institute of Standards and Technology collaborated on.

But I think we should be very worried, and even before we get to the catastrophic power failure situation, we have other likely scenarios that are still extremely troubling. I have a personal example of this. [My wife and I] both have Teslas. One day we hit a pothole, and one of the tires broke and had to be replaced, but the app was not working, no matter what I did, reboot the thing, reload the app, and got to the point where I wanted to order a new tire, and the app just couldn’t perform the right functions. The good fortune is that I live about 10 minutes away from a Tesla dealer that happens to be the main hub on the East Coast, so I was able to get there in our other car and talk to a human being. I told them that the car was in my driveway and needed a new tire and possibly a new rim, and we got it all worked out. I didn't have to rely on the app, but if I had been reliant on the app and it wasn't working, I’d be stuck with a dead car and a dead app, and there’s no phone number to call. Imagine if I had been somewhere remote, with no digital capability. That’s a big problem.

The more I think about those scenarios, of which there are many, I keep thinking we need to seriously start considering resilience in the way our systems work and the various dependencies that they have on each other.

There's a novel called The Machine Stops. It was written in 1909 by E.M. Forster, about a society that lives underground in individual pods and an omnipotent machine sends food to them and provides them with technology to communicate with each other from home. Sounds a little like the COVID pandemic, right? But this was 1909. Forster lays out this premise of this society that lives alone at home, never seeing each other face to face. Then, he posits that the machine stops working. Of course, that question is, what happens to that society? Did they remember anything from a life before the machine, or were they completely helpless?

I have been thinking more and more about how fragile it seems our technological society has become. We are incredibly dependent on devices of increasing complexity and on people who know how those devices work and can fix them when they break. Ask yourself, over the hundred years, what technologies have we become very dependent upon? It’s everything. That’s a chilling thought, and it's our reality.”

INTERVIEWER:

“Do we have any backup for those systems? Let’s say everything goes down. If and when we can reboot, what data and information will be lost?”

VINT:

“I take notes in a notebook with a pen, and it's old-fashioned.”

INTERVIEWER:

“I do too, and always have. I’m not sure why but it helps me process and retain thoughts.” 

VINT:

“Exactly, and I can't type very fast. There’s something reassuring about a piece of paper that could last at least a hundred years or more. We know that there are pieces of paper that have lasted 500 to 1,000 years, and there are written pieces on vellum, which last 2,000 years or more. I have a project with a small group of volunteers right now called Digital Vellum, and It's all about trying to figure out how we take digital content and make it preservable, accessible, and useful over hundreds of years or more.”

INTERVIEWER:

“I’m so glad you brought that up because I read about your research on digital obsolescence—that losing our archives could lead us to a digital dark age where we lose historical records and data storage. Preservation of the written word is something I think about a lot as a writer.

It's already happening as publications become defunct, their digital archives are erased and gone forever. I think the idea of having this blend of digital-analog systems is the perfect marriage of ancient and modern methods. It’s a very human solution to a technological problem. With that sort of interface and how we relate to technology, are you observing a trend in humans wanting to connect more face-to-face versus through a digital platform?”

VINT:

“Another good, tough question—I'm not hearing great news from parents observing their teenagers today. There's an excellent book on this topic by Sherry Turkle at MIT, and it’s called Alone Together. Some of it is devoted to human interaction with humanoid robots. People project onto the robots' social knowledge that they don't have, then get upset when the robots seem to ignore them or don't respond well because they feel like they're being dissed or something. The robots just have no clue about what's going on.

She also talks about teenagers who have become very accustomed to texting. Texting, telephone, and face-to-face conversations are extremely different. In the case of a real-time conversation, regardless of whether it's texting, voice calls, or face-to-face, you're confronted with an exchange where you might not quite know how to respond. In a real-time situation, there's this pregnant pause, and it can be awkward, whereas in texting, you're forgiven if you don't respond, because the presumption is that you got distracted, or there was something else on the calendar and were using texting in the interim. Kids have become accustomed to not responding when they're uncomfortable, but we're doing something in real-time right now, so let’s use this as an example. If I just suddenly stop talking…”

[Vint demonstrates his real-time social experiment by silently looking at the screen as I go along and do the same, both of us waiting for the other to break, which I did after 20 seconds when I started laughing.] 

INTERVIEWER:

“Ah, the long pause.”

VINT:

“I was wondering who would break first, and we made it about 20 seconds, which is pretty good! This is exactly what kids are trying to avoid. Nobody knows what to do in this situation.”

END OF PART 1 

INTERMISSION 

Recorded, Written, and Edited by C. Thompson for Third Chapter Curious, LLC, 2025