Jonathan Zittrain on the Internet and Society

by Molly Lanzarotta, interviewed on July 16, 2013

In a matter of mere decades, the Internet has transformed how we socialize, learn, and do business. As with all revolutions, we hear that we are living in the best of times and the worst of times. Jonathan Zittrain is a professor of law at Harvard Law School and Harvard Kennedy School, the co-founder of the Berkman Center for Internet & Society, and a professor of computer science in Harvard’s School of Engineering and Applied Sciences. His research focuses on the legal, technological and culturally transformative aspects of the Internet.

Q. In your 2008 book “The Future of the Internet and How to Stop It,” you discussed the differences between early generative technologies that encouraged experimentation, and what you called “sterile” appliances, iPhones and Xboxes, “tethered to a network of control.” You warned that the Internet’s trajectory could be one of lost opportunity. Five years later, where do you see us in this dynamic of potential creativity and sterility?

Zittrain:I drafted my book on the future of the Internet in 2006-2007 and it was just at the time that iPhone had come into prominence. Prior to that time, we were still using regular old flip phones that could barely get online and I bracketed the introduction and conclusion of that book with some observations about the introduction of the iPhone and what it could be representative of.

At that time the iPhone didn’t allow any outside code to run. It had a cool little screen, but no third party applications – there was no Apps Store – there was just what Apple wanted to put on the phone. At the time that was an avowed Apple strategy. Steve Jobs said, “You don’t want this thing to act like a PC.” Because you load three things on your PC and before you know it, it’s crashed. And that’s not the way a phone should work; it should be trustworthy, like an appliance. And there’s a lot of truth to that.

Interestingly, just as my book was coming out, the iPhone had opened up a little bit. It had introduced the Apps Store and suddenly third-party coders – the kind who would be writing software maybe for Microsoft Windows back in Windows’ heyday – were writing code for the iPhone as well. But there was an interesting difference between that era of PCs and Windows. The iPhone didn’t run just any code – you couldn’t just load it up as the holder of an iPhone – instead you had to go through the Apps Store, with extremely limited exceptions. And what that meant was that Apple could serve a gatekeeping role on what kind of code could run on the phone. They asked outside coders to register with them, to disclose their identities, to be certified, and left enough wiggle room in their software development agreement so that they could basically veto things that competed with Apple functionalities.

There were very few – especially in the early days – alternative email programs on the iPhone; you had to go with Apple Mail. It was very difficult to get a fully functional browser on the phone; you had to go with Apple Safari. These are things that would have been unheard of with Microsoft Windows, even during the days of their antitrust lawsuit brought by the U.S. government. And now it’s commonplace with the phone. Maybe that’s the best of both worlds or maybe that’s the worst of both worlds – that you entice a bunch of programmers in, you make the platform very attractive, but at the end of the day there’s this opportunity for control. That was a wonderful, open question just being born at the time that my book came out.

I think what we’ve seen occur in the intervening five years or so is a mixed bag. There are opportunities not just to control the code that runs on the phone, but also the content. Apple is able to take its 30% not just of sales from the Apps Store but of in-App content that might be sold. So if I have a newspaper App, it’s going through Apple Newsstand, or it’s otherwise something that Apple can say “this is the kind of content we don’t want.” For example, there was an instance of a Pulitzer-prize winning cartoonist who had his cartoons banned on the iPhone – a level of control unusual for a technology vendor. It’s the equivalent to buying a home with the provision that you can only furnish or paint the home with products that are approved by the contractor. That would be very unusual, and the fact that there’s some competition – you can get an Android instead of an iPhone – might not fully cure that because once you’re committed to one platform, you’re sort of stuck there.

It’s been so interesting to see how these platforms aren’t about mobile in particular. These are the sorts of things that have found their way into Apple’s iOS/OS X or into Windows 8 such that there is this opportunity for gatekeeping and it's thanks to the fact that there’s so much saturating network around us that what used to be a product – a phone, or a computer, an operating system – is now a “service.”

So I believe we are in the era of cloud, and cloud necessarily entails an on-going relationship and, potentially, control by the vendor with whom one interacts, and that’s true even when we think we’re buying an object. That object is phoning home all the time and getting guidance.

So those are some real opportunities for control and for limits on our freedom, either by the vendors themselves or by any government that can persuade or force the vendor to surveil or shape what happens on or through the device. That is the sort of thing I was worried about in 2008 – even as I have an iPhone. And I enjoy it, too.

Q. I’ve heard you discuss some wonderfully creative uses of technology in teaching, from “intellectual playlists” of live-annotated texts to “rotisseries” of questions about a common text that professors and students can electronically share around the globe. Can you say more about this?

Zittrain:This is an amazing, fascinating time for education relating to technology. We’ve seen ways in which there’s some reason that institutions of higher education are a little bit afraid – they’re afraid they’re going to get their lunch eaten by cash and carry online degree and certificate programs for which the consumer – the student – says, "I don’t need the residential experience, I don’t need the activity fees, I don’t need any of that. I just want to be able to pick modularly what I want to learn about." I think for some skill sets, this is absolutely a terrific way to go and we’ve seen, in this corner of the world at Harvard, an oar really being dipped in the water with EdX, of which Harvard is a part, and HarvardX, which is that component that Harvard does. It is interesting to note that EdX is a nonprofit. It’s not doing this for profit the way that a number of the current – circa 2013 — competitors like Udacity and Coursera are.

I worry a little bit that if we develop an overly consumerist a model for education -- particularly for undergraduate education -- we will lose the opportunity to bind certain students together and have them meet one another because we’ll be thinking of it too much as a product that one buys and gets tested on, and then we move on to the next person. So I’m interested in both online distance learning and integration with our current classes at places like Harvard Kennedy School or Harvard Law School or the School of Engineering and Applied Sciences. I’m interested in ways that course materials can be gathered and made vibrant rather than “here’s a stack of stuff that I’m supposed to read as part of the course.”

So the way we’ve been experimenting with that is through a concept of playlists. I happen to run the Harvard Law School Library, among other things, and under the auspices of that library, we’ve created a first year legal curriculum on our education platform called H2O that is free and open and organized like a playlist. So I might have a concept “self-defense in court” that I’m about to teach, and there’s a playlist of cases and other materials that I’ve edited that I can offer the students. Because of the way the system works, it doesn’t actually delete anything; it only applies and overlays. What that means is that when the students hit something that I have cut out of the case – there’s just an ellipsis that says “oh, yeah, there’s more stuff here but don’t worry, you don’t need to see that” – the student can at any time click on the ellipsis and see what was cut. And that means that any time he or she wants some content that may not be contained in the case -- some context – there’s an opportunity with a single click to see what’s there, to make annotations, to share those annotations with a study group, or with the rest of the class. When I’m in class, I can display on the screen the texts that we’re talking about, and we can be highlighting it all together on one big screen rather than our noses buried in our respective books.

Now for the students so far in version 1.0, it has been up and running – they feel happy because they are not having to pay the usual fee. A typical legal casebook costs $200 per student per class, and the class only lasts a term. That adds up. This doesn’t cost them anything and, very prosaically, it’s not very heavy. It’s something digital that they can read on a Kindle, on an iPad, on an Android phone, on a PC, you name it – and, of course, they can print it out. Now the real benefit comes when other professors who are looking to teach a course of whatever kind – whether it’s in law or public policy, where there’s a lot of documents, government edicts, policy papers –can look at other professor’s playlists and borrow as much or as little as they want to shape or augment their own courses.

So instead of our having to chose a case book or a text book, it means that as teachers we can drag and drop from all sorts of sources, and mix and remix and offer that back to the commonweal, which means that a curriculum in transportation policy, in defense policy, in tort law, can evolve one professor at a time, one small change at a time, rather than relying on a handful of professors who generate these case books in their umpteenth editions and which change just a little bit over time. My strong suspicion is we’ll see a much more vibrant curriculum and one that can be readily localized around the world.

In addition, it’s nice to be able to ask the students a question around the readings, and because this system is aware of lots of different professors and courses drawing upon a common set of materials, that means that if I ask my students a question through the system and they answer, the system may know that there are some students in Singapore looking at exactly the same text. So essentially the first professor could ask the second professor, "would you like to arrange for your students to have their answers critiqued by the Singaporean students, and visa versa?" And if the other professor is game you can end up having on the fly intellectual interactions around the common text that simply wouldn’t have been possible before a system like this. So developing this under the auspices of universities seems to me exactly what universities are called to do. Rather than getting an off-the-shelf “Lectern 3.0” software and leaving it to someone else to write it and maintain it and charge us for it, we can build our own sets of materials and active verbs around those materials in a way that reflects the best of what a university is supposed to be.

Q. You’ve been studying the Internet for your entire career. What are the trends you are following most closely now? What are you most excited about?


Zittrain: Well there have been some interesting trends in cyber security, cyber warfare – lots of words beginning with cyber – and I’ve talked about some of those in “The Future of the Internet” book, but right now when I look at tomorrow’s issues that may be under-scrutinized, I think in 2013 we’re on the cusp of a lot of interesting privacy issues outside the norm of traditional privacy concerns. You can see this legalistically as a Fourth Amendment constitutional issue in U.S. law or you can see it more broadly in world-wide surveillance. And there is concern about corporate surveillance – cookies, and what happens when I visit a website: what do they know about me, what kind of advertising is tracking me? Those issues are largely well-known and we know how to fight about them intellectually and in the policy arena.

I might add that one of the interesting things that may come up with corporate surveillance is not just learning information about you and targeting ads to you that you might actually like but, rather, price and service discrimination. Meaning that when I walk into the supermarket, they may have a sense for just how willing I am to spend money and on what items and offer me a higher price than somebody else who is a bit choosier as a shopper. And that’s something that we haven’t contended with before – except in the case of automobiles and airline tickets – that different people pay a different price for the same item. And I think we will be seeing more of that and service discrimination as well. Home Depot or Lowes or any other big box store may get a sense for just how valuable a customer you are and, if you need help, you may find it less forthcoming if your profile indicates you’re a lot of talk and not a lot of “bought.” That’s the kind of thing we could see coming into the fore.

Aside from government surveillance and corporate surveillance that leads to privacy issues, of most interest to me are peer-to-peer privacy issues. 2013 saw the introduction of Google Glass and there will be more like it in the future. We see lots of start-ups wholly apart from Google. There is one called Looxcie with a little cylinder that you attach to your eyeglasses and you can watch the world as you go. There is now an app for an iPhone circa 2013 that just like a flight black box data recorder continuously records everything going on around the iPhone and leaves a five, ten, 15 minute window. And if at any point something happens to which you say “huh, I want to save that tape,” you can click “save it” and now the past 15 minutes of your life are preserved should something interesting have happened. And once that gets going and concatenated with facial recognition and tagging and live streaming, we’ll see what we thought of as a private sphere, while out in public and interacting with others, becoming a very public sphere, especially if anything out of the ordinary is to happen. And that’s going to have implications for teachers in a classroom, for every day incidents in a supermarket or on the road, as all of that is suddenly, potentially fodder for viral or other coverage online.

Another set of issues I’ve been interested in – and this is what my next book called “Cog” is going to be about – are the ways in which we’re applying so many of the principals of cloud computing to human labor. So there are services like Mechanical Turk in which you can hire somebody for five seconds at a time. Five seconds worth of labor integrated across millions of people can yield very interesting results. I could put out an all-points bulletin so that if anybody happens to be walking by a particular house, there’s an opportunity to take a picture or video of that house and suddenly that house is under a form of surveillance far more thorough and cheap than hiring a private investigator or having two law enforcement officers sitting outside the house and monitoring it. Or if there’s an anonymous person – I don’t know who it is, but I have a photograph – maybe I’m the government and I want to know who a protestor is who was seemingly anonymous – I could hire tons of people to juxtapose my national I.D. cards for which there may be millions of I.D. cards, drivers license photos, and compare it to that one unknown protestor. With enough people spending five seconds at a glance, I can find that needle in a haystack, for better or for worse.

So there are lots and lots of implications of this field that we now call “human computing,” treating the human mind as if it were a processor that can generate results in a consistent fashion. What happens when you can plug the human mind into any step of an otherwise completely computational process raises some really fascinating questions which I hope to answer.

Q. Do you have any advice to offer to potential students considering coming to Harvard?

Zittrain:I would say to anyone considering becoming a student here and studying internet technology: it’s a wonderfully generative space in which to think about these topics holistically. The intersection of the entrepreneurial and policy -- of the tinkerer and the systematic thinker -- is part of what makes this field so fun. It’s great to have such an intersection and at this university. There is some great cooperation among the Kennedy School, the Law School, the Engineering School, the Business School – really looking deeply into some of these multifaceted issues from the multiple angles that they call for.

John F. Kennedy School of Government 79 John F. Kennedy Street
Cambridge, MA 02138
617-495-1100 Get Directions Visit Contact Page