Dr. Vinton Gray Cerf is recognized as one of “the fathers of the Internet” along with Bob Kahn as co-inventors of the internet protocol, TCP/IP. He has some 35 honorary degrees and won many awards including the National Medal of Technology, the Turing Award, the Presidential Medal of Freedom, the Marconi Prize. He currently serves on the National Science Board and as the VP and Chief Internet Evangelist of Google.
Sidd: You’re the Chief Internet Evangelist for Google. What exactly does that entail?
Dr. Cerf: Almost anything that comes to mind. Seriously, I didn’t even ask for this title. It was very funny, I joined the company in 2005 and they said: what title do you want? Then I said how about Archduke? Larry [Larry Page is the co-founder of Google] and Eric [Eric Schmidt is the CEO of Google’s parent company Alphabet] turned away and they came back and they said: you know, the previous Archduke was Ferdinand and he was assassinated in 1914 and it started World War I. They suggested a different title and it was Chief Internet Evangelist and I said okay, I can do that.
So I stayed in the research group but a significant part of my work involves internet policy on a global scale. So in addition to being concerned about domestic policy in the U.S., I care a great deal about national policies in other parts of the world and international policies. So I’m on the road about 80% of my time and some of that, significant part of that is spent essentially trying to shape policy primarily to promote the spread of internet around the world. We have about 50% – 51% of the world population online and of course I’m looking to get the other 49% as soon as possible.
The theory being that access to internet is increasingly important for everybody so that means shaping policy with regard to access network neutrality, with regard to the ability to get pretty much anywhere you want to on the net, no censorship, things like that. And it’s not going to work everywhere but you can at least advocate for that but it also means paying attention to things like investment policies and the ability of companies to get started so startup policy is in this picture as well. For education and that’s preparing people to make use of this technology in their daily lives, in their line of work and then I’m sure you could attest, just over time anyways, use of the internet is becoming increasingly vital to get daily work done. Mobiles and smartphones of course have simply accelerated that. So there’s that one angle of all this and I do spend a lot of my time on it.
On the other hand, Google has also been pretty generous with my time so I have various appointments outside of Google and they also consume my time. I sit on the National Science Board which I oversees the National Science Foundation for example and I was president of the association for computing machinery and served as chairman of the board of ICANN (the Internet Corporation for Assigned Names and Numbers). I am chairman of something called the Marconi Society and very active in a little group that Mei Lin Fung and I started up called The People-Centered Internet which tries to achieve more useful internet implementation… So those, literally, anything which looks like it would be useful for improving either the scope and access to the internet or benefit people who are trying to build on top of that platform sort of falls into my territory.
Engineers … the most important thing they can do is to learn how to sell
Sidd: How important has communication skills been to your efforts in trying to get both the ARPANET and internet adopted?
Dr. Cerf: Absolutely essential. In fact, one of the lessons that I tried to convey to engineers is that the most important thing they can do is to learn how to sell. And it’s kind of interesting because some of the reactions are, oh that’s awful, I would never want to do that. I explain to them that if the sales guys can’t sell what you’ve built then you don’t get paid, and that gets some attention. More important, if you can’t sell your ideas, you never do anything big because in order to do anything big, you have to have a lot of people who want it to happen and by good fortune, internet turns out to be a salable idea. I learned very early on in my networking career that this was an important skill to have. I don’t want to over claim credit here, there were lots of people with fingers in this pie. I’m sure you know, I was in a good position to try to advocate for investment in and implementation of internet while I was in the Defense Department and subsequently, in other jobs like in MCI and here at Google.
Sidd: I’ve seen so many scientists and engineers who are conditioned to do a data dump and they can’t do the presentation without the slides.
Dr. Cerf: Oh that’s interesting. Sometimes, I start my talks with I’m not going to use slides, I want you to listen to what I have to say, Power corrupts and PowerPoint corrupts, absolutely. So in fact, it is absolutely correct that slides that have a bunch of text on them are so distracting that people that will read the slides and won’t listen to what you have to say and so there are times when the best presentations often have no text at all, it’s just something to illustrate a point.
Or another very very good lecturer is Larry Lessig . Larry’s slides have, many of them, very little on them. When he does a presentation, he might use 150 slides but each slide has maybe one word on it or an illustration but nothing else and he will, as he makes the point that he wants to make, he shows that slide. And so his presentation on the laptop shows him what the next slide is and he knows when he says that word, he gets the slide that has that word on it. So it was an emphasizer but it isn’t, this is not intended for people to stare at the slide while he’s talking and I think that’s a very important skill to learn.
Sidd: What I have been trying to tease out subsequent to that speech was what is the bare essence that is required when we have to communicate technical information in a high stake scenario? And you were instrumental in getting the internet adopted working with Al Gore and other people, giving that, converting that defense project into essentially something that would revolutionize the entire world. The communication aspects of it for me, my standpoint looking back on it that that was a critical, communication was a critical component of getting the internet commercialized.
Dr. Cerf: That’s true. So let me suggest a couple of things. The first one is that the purpose, my purpose in communicating is to try to get the target of that communication to have the same model in their heads that I have in my head that’s driving me and that’s applied in a variety of different ways. It might be a business model that I’m trying to convey or it might be a technical concept that I’m trying to convey. In fact, I’m in the middle of writing a paper right now on the Internet of Things and that’s the same thought that is in front of my head, so to speak, is how do I get my model of what that internet is, what an ecosystem looks like into the heads of the people who are reading this paper. And I don’t even want to rely on illustrations necessarily because they are too limiting and I want people to build up a conceptual model as a consequence of reading text as if I were speaking to them. And I can tell you that I don’t find PowerPoint or other presentation tools to be fine. When I’m trying to explain to people the architecture of a system, the thing which I find the most helpful is to have a whiteboard where I can draw the architecture that I will do at a step at a time. I put a component up and explain what its role is and then draw a picture of the next one in front of it and show how it’s connected to the other. And work my way into a very very complex slide but I would never show that slide in its complexity at the beginning. I want people to walk through the zoo so to speak, visit each of the animals one at a time and see how this thing builds up.
So I remember lecturing newbies coming to MCI about how the MCI mail system works and took about 25 minutes or so to build up a view of the architecture of the system one step at a time but I love that style of presentation because if people get to encounter the design in the same way you did. So once you figure it out then you can figure out a good way to walk through it in a the rational sequence but I really like that style because I think it helps people understand and build up their own internal model of the concept you’re trying to convey.
Sidd: It’s really kind of amazing to me that that’s not an approach that’s widely used. It is incredibly powerful but nonetheless, I’ve seen so many academics and scientists and engineers basically like we said, doing that data dump approach. And the interesting thing to me was they don’t even seem to be, you can tell them that that’s not a particularly effective way and you can show them. I have an interview with a guy that was the, his name is Dr. Scott Huettel, who’s at Duke University. He was the president of the Society for Neuroeconomics. He’s studying the biology decision making and he was talking about how our brain actually, it’s very difficult for our brain to process all that information at the same time and if we could create the framework of information, it gives our brain the ability to organize information as we get it, right? Which is a much easier way of consuming a large amount of complex information. So in other words, we need to see the big picture, right? And then build complexity as we go along. But there seems to be a resistance amongst engineers and scientists and educators to really kind of address the communication aspect of their field.
Dr. Cerf: Let me give you an example of something that I’ve experienced a few weeks ago. There’s another organization that I helped to found, it’s called Innovation 4 Jobs, I4J. In fact, if you want to go to the website, it’s I4J.info. So I was leading a discussion at a round table of about 8 – 9 – 10 people about blockchain and there were some advocates of blockchain and I was the nominal skeptic. I had forced the conversation to solve a particular problem and I wanted to create a scenario and then I wanted them to demonstrate how the blockchain system solves the problem. The problem had to do with selling a book to someone else and wanting to demonstrate that the book’s ownership had transferred and to use the blockchain distributive ledger as a way of recording the fact that the ownership has transferred from me to the buyer. So it’s a transactional thing which is what blockchain conventionally is used to record. And what I insisted on doing was making the blockchain people talk to me, talk to us about what the conditions were that you needed to have for this system to actually work. I wanted to start out from the standpoint of this is me and the buyer doing this transaction and what is it that I need to provide to the blockchain system in order for it to record this transaction for me. And so I’ve forced them repeatedly to go back to specifics about how this was actually going to work and what things have to be in the ecosystem and what state that they have to be in and how do they get in to that state, how did they get configured in order to work? And so that’s a little different than starting with a high order level of abstraction and then working down into the detail. I wanted people to, be forced to explain to us in words that we can understand exactly what had to happen.
You might argue well, isn’t that starting with too much detail? I don’t think so. And the reason I don’t think so is that it was so fundamental to get the people around the table to understand what assumptions were being made and what conditions had to be satisfied in order for the system to work and only then could we then start to ask about well, how rapidly can transactions be recorded in the system, what are the constraints on how these things scale. So I was very proud of that discussion and I have to say that I got lots of positive feedback. Some people saying that they haven’t really understood blockchain until we have had that discussion. The people who were promoting blockchain later said that they felt like they were in the middle of a PhD oral defense which for all practical purposes is what I was making them do. I really liked that.
… Figure out at what level of abstraction could we explain the system so that people can get a kind of intuitive appreciation for what the parts are
He did say something else though which I want to endorse. There is a certain level of abstraction at which complex systems can be understood and you’re down in the weeds, it’s very very hard to go and understand the architecture. And so one of the most critical things that a designer can do is figure out at what level of abstraction could we explain the system so that people can get a kind of intuitive appreciation for what the parts are. When I was designing MCI Mail, I had this very consciously in mind. I took a sheet of paper and I drew on that sheet of paper the half dozen or so fundamental systems that were part of the design and then showed how they were interconnected and labeled the connections with the protocols that I was going to use to allow the information to flow between the various parts. And that picture would serve very very well for me as the system architect. When someone would come and say: I want to change x, I would pull out my sheet and say let’s look and see what the consequences are of changing x in to x star, what will be affected by that and that single sheet architectural view made sure I didn’t leave something else out of the analysis of the implications.
Sidd: You’re the perfect example, for me anyways, you have reached the highest, the pinnacle in engineering and at the same time, been able to be an extraordinary communicator and then that’s the reason why I felt very privileged to be able to talk to you about that because there’s just not a whole lot of people out there in the world that can say that they’ve done both.
Dr. Cerf: Well, I’ve tried anyway.
I think it is safe to say, by every standard, he succeeded.