Vinton Gray Cerf, one of the creators of the Internet, talks about the future of the technology
New Delhi: Vinton Gray Cerf, 69, is one of the creators of the Internet. In 1974, Cerf and fellow American scientist Bob Kahn wrote the protocol for computer networks to communicate with one another, making the Internet possible. Almost 40 years later, Cerf sees the World Wide Web as a curious youngster still—childlike in terms of its ability to absorb new ideas and inquisitive enough to discover new things. In an interview on Tuesday during a visit to New Delhi, Cerf, who holds the title of chief Internet evangelist at Google Inc., described proposals to levy an Internet tax as “bizarre" and said rising interference by governments in managing the Internet worries him the most. He also spoke about the future of the Internet. Edited excerpts:
Thirty years on, is the Internet facing a mid-life crisis?
It’s not as old yet in years as it might seem and there are a couple of reasons for that. When you’re a child, a lot of rapid development happens; you are acquiring languages, learning and getting physically bigger. So the Internet has been growing a lot, especially in the past three years, but it’s not out of childhood, at least in the following sense. It remains a highly evolvable infrastructure and it keeps absorbing new technologies, keeps supporting new applications partly because of the physical equipment available in the mobile space, tablets and so on.
Also, because the capacity keeps going up—there is optic fibre. So between this continued evolution of technologies, there is also a continued evolution of application. And that’s possible because the Internet is not built for any specific reason. It was simply designed to transfer packets and these packets don’t know what they’re carrying. It means that new applications could be invented without changing the underlying network. So nobody had to get a permission to do Voice Over IP (Internet Protocol), nobody had to take permission for doing the World Wide Web—nobody came and asked me, they just did it, just like Google, Amazon and Yahoo.
So, this permission-less innovation is the hallmark of the Internet. It’s very important to realize that we retain that feature and somewhere in our conversation we are going to have to talk about governance, privacy and other issues.
I have this ‘ant-hill’ theory of innovation on the Internet. If you ever have sat and watched an ant hill as a child, you will remember that during the course of the day, a couple of ants will find something interesting and drag it to the hill, just like the Internet. Because if you follow any person and his or her activities on a given day, some of them will have interesting ideas, some ideas go viral, some businesses will be born. That’s why the Internet is interesting, it’s like an anthill of creativity.
Many of the claims regarding the Internet appear to be science fiction.
It sounds like science fiction but that’s what engineering is all about. If you are a serious engineer like me you will take science fiction and turn it into reality. There’s this famous quote that goes something like this—any advanced technology will appear to be magic until somebody understands the basics behind it. You take some of the things we do today and a lot of it would have appeared magical sometime back. How about Google’s self-driving car, that’s actually quite a hard problem to solve if you think about it. It has to know all traffic signals, follow all driving rules. Information is being combined to allow these cars to navigate on the streets of San Francisco. I don’t want to over-think about these cars because the harder problem is not navigating on the streets, it’s actually going door to door. It’s about getting out of the garage, finding the office, you have to have the knowledge not only about the maps for these streets but how to find and park in an underground garage, it’s about telling step by step, turn here, turn there. Self-driving cars sound like science fiction.
The point I want to make is we are all experiencing a transformation that is moving from science fiction to reality. How about speech recognition? That would have been science fiction 10-15 years ago.
I want to emphasize that Internet environment and the platforms it supports, including big data and cloud computing, have all come together to provide functionalities you couldn’t have before because we now have a huge amount of computing power.
An Indian phone company’s founder (Sunil Mittal of Bharti Enterprises) recently suggested that there should be an Internet tax to be paid by firms such as Google. What do you think?
The problem is that the way Internet works is quite different from the way a cell phone works. The telephone industry said when you make a phone call you pay everybody that connects you. The Internet says you should pay the guy who provides you last mile of connectivity.
This taxation idea is bizarre because regardless of what the content is, regardless of the value of this content, it’s taking up the same bits. It’s almost like the post office proposing that we will charge you not for the cost of delivering something but on the value of what we deliver. Can you imagine them saying, so you are sending a cheque for two million rupees, we want 2% of that.
I don’t buy this (argument) at all. At Google, we already pay telcos and there are users paying on the other side—I can’t understand why there should be this insistence on having an Internet tax.
What are some things that the Internet could do in future but appear science fiction now?
The Internet is already seeing some of those becoming reality. One of them is about how sensor networks are evolving. They are being used for many things, one of them is security and another one is for environmental control. A third example is adaptation of this to understand a user’s needs and preferences. Imagine walking into a room that has motion sensors, the lights go on and if you are not there after a while, the lights go off. It can also get annoying if you are sitting, reading, without any movements and the lights go off. You then have to wave your hands or clap for the lights to come on.
But the sensor networks are going to be increasingly common and connecting them to the Internet will turn out to be very valuable because they will gather information about for instance, how we use electricity, use our resources. This information could be captured by sensor networks and available for analysis. So the idea behind Smart Grid is about appliances that report how much of electricity they use and when. So, at the end of the month you have an idea about consumption, and what behavioural changes are needed in terms of consumption patterns. This also helps electricity generating firms because they can tell when not to use power, or suppress some of the demand. This process of making smarter devices is now beginning to show up.
I used to tell jokes about Internet-enabled light bulbs that one day every light bulb will have Internet. I can’t joke about this anymore because somebody sent me an LED that has Internet and an IPV6 radio just this year.
Now these may cost around ₹ 1,000 but they last for 15 years—LED does not have any natural erosion characteristics. Even a lot of devices from LG, Sony and Samsung are coming with Internet-enabled functions.
Can you illustrate with an example?
Suppose you are a manufacturer of detergent for washing machines and your customer is faced with a terrible stain on a blouse and wondering what to do next. You can imagine a scenario where you go to a webpage that tells you which detergent to use and in what proportion. Then it says, ‘oh by the way, are your appliances (washing machine) online?’ You say yes, and the webpage asks you to click on a button that will set up correct parameters for washing and automatically cleans the stain. So now, this detergent has become a service because the Internet could help the soap-maker use the appliance.
You can imagine a variety of such examples wherein these Internet-enabled appliances offer different services.
Are you worried about increased government control in managing the Internet?
Well it worries me, does not scare me though. There are several reasons to be concerned about increased governance. One of them has to be the governments demanding information from suppliers of services on the Internet and at Google, we are frequently confronted with such demands. Our response is that as long as it involves proper due diligence and procedures followed it’s fine. One aspect of this is to protect people’s privacy from unreasonable attempts by governments. If due process is followed, then we will be foolish not to comply, our business model is not to go and break the law.
The other concern is what role do governments have in governance of the Internet. The term governance needs to be understood in an extremely broad context. So, the Internet is governed in some sense by its software, hardware, applications and also by business and legal rules around the world. So governments look at what’s happening on the Internet and say we have to do something to protect our citizens, and it’s a perfectly normal reaction. I am coming from an American point of view and one of the things we value very highly is this freedom of expression. And this is essentially embodied in the ...Declaration of Human Rights. So we think a good government needs to bear that in ...(mind) while at the same time trying to figure out how to protect its people. That is why we are very careful in sharing information on demand with governments.
The problem we are seeing is that governments are accustomed to having the last word. On the Internet, governance is a very large thing, there are a lot of different stakeholders who contribute—the standards making body, engineering and research teams contribute to Internet’s ability to evolve. It’s an intentionally multi-stakeholder- based approach. Civil society and governments, all have a say. I am a big proponent of this multi-stakeholder based model.
I am not suggesting that inter-governmental role shouldn’t be there in governing the Internet, but it has to involve multiple stakeholders in policy formation. The thing that worries me is that some governments’ inherent attempts to protect their citizens is done in a way that either ignores this multi-stakeholder part or damages human rights. There has to be a balance there. My big worry is that some governments really fear expressions on the Internet about governments. We see all information on the Internet, but users are not stupid, they have the ability to think and they can decide which information is what. Rather than suppressing the rumours, they should respond with more information.
My approach is quite simple. There is a website called Snopes.com that keeps debugging and debunking all rumours.
So what’s the real solution?
The solution to the problem is more information, not censorship. This is where I trust users of the Internet more than some governments do. Censorship can be potentially very abusive. The right thing to do is respond.
The previous mass media--television, radio and print--have often been controlled by some governments. Internet started out in an environment where everything was open, and that’s new.
I think we would make a serious mistake if we try to do with the Internet what has been attempted with other media. There will be seven billion people on the Internet someday soon, and trying to monitor them is very hard. The moral equivalent of trying to control people on the Internet is trying to control what people are talking on phones.
Can you imagine a phone conversation that gets dropped because you say something that a government does not think you should be saying?