* * *
It’s pretty easy to argue that the Internet has been a transformative
technology — although some people argue that it hasn’t been. That for
all its shiny and sexy features, it isn’t so much a new new thing as it
is a sort of booster shot for existing technologies. Email and other
Internet communication may be faster, more flexible, and cheaper than
snail mail and the landline phone and the telegraph, but it’s still just
communication.
You may be extremely reliant on Uber or Spotify or Amazon, but we
already had cars, and music, and catalog stores. You may be downright
addicted to Facebook or Twitter but we already had — well, mouths, and
ears, and friends and enemies. How transformative has the Internet been
for you? Your answer probably depends on a lot of factors — at what age
you started using it; exactly how and how deeply you engage with it; and
maybe the degree to which you want your Internet engagement curated,
whether by Facebook or Google or, as a user’s agreement might put it, by
any and all platforms now known or hereafter devised.
The Internet will of course keep changing, and it’s already changed a
lot. How did it get to where it is? How closely does the modern
Internet resemble its original design? Those are some of the questions
we’ll be trying to answer today. The episode was inspired by a recent
issue of
Daedalus, the Journal of the American Academy of Arts & Sciences. The issue was called “
The Internet,”
and it featured essays by some of the scholars you’ll be hearing from
in this episode. Let’s begin at the beginning of the Internet, with
Freakonomics Radio’s senior producer
Christopher Werth.
CHRISTOPHER WERTH:
David Clark
first got his hands on a computer in the mid-1960s, when he was a
student at Swarthmore College. The IBM computer the school bought was
this big, clunky contraption. It filled an entire classroom. And he says
everyone took turns using it.
DAVID CLARK: At the time I was an electrical
engineer, and when this computer showed up, it actually came with a book
with all the schematics. And I sat there and I read them, and I studied
the operating system, and I was hooked.
WERTH: Not long after that, Clark switched to computer science and
went on to get a Ph.D. at MIT. In fact, he’s still there, as a senior
research scientist at the
Computer Science and Artificial Intelligence Lab.
But, in the 1970s and ’80s, Clark was part of a small, loose-knit group
of engineers in academia and at government agencies in the United
States and Europe who fantasized about linking all those big, clunky
computers together so that people all over the world could collaborate.
They tinkered, they experimented and ultimately figured out what this
network that would later be called the Internet should look like; how
it should work and function. And to this day, Clark says, everything we
do online sits on top of this largely invisible, underlying structure
they created. For example, Clark was a key figure in developing the
basic language of the Internet, which passes information back and forth by breaking it up into these little pieces called “packets.”
CLARK: Basically, what it does is it numbers all the
packets in sequence. When they’re received, it puts them back in the
right order. If there’s a missing packet it gets sent again. So
eventually, the sequence of packets gets reassembled and then that’s
handed on to the application at the receiving side.
WERTH: The important thing to bear in mind, he says, is that there
was no central blueprint for the Internet when it was being developed.
Engineers, like himself and others, just had this notion of connecting
computers together over long distances. And he says it really could have
gone any number of ways.
CLARK: We were making this up from scratch as we
went. And there weren’t any design principles; there weren’t any
guideposts to help us. This was a new, uncharted territory. So
recognizing that this is an engineered artifact — it’s not something
that just happened — helps you to understand that it could have come out
different.
WERTH: That is not to say the Internet wouldn’t have happened at all.
In the 1960s, there were technologists who were already dreaming about
this stuff.
CLARK: One of the really notable visionaries in this area was a man called J.C.R. Licklider. Everybody called him “Lick.” And he was a genuine visionary.
WERTH: “Lick” was a psychologist and engineer at MIT who was very excited about the computer’s prospects.
ARCHIVAL TAPE OF J.C.R. LICKLIDER: Computer
technology has been moving in a way that nothing else people have ever
known has moved. Here’s a field that gets a thousand times as good in 20 years.
WERTH: Licklider was also a director at the
Department of Defense’s Advanced Research Projects Agency, or ARPA. It was responsible for the creation of
ARPAnet, a precursor to the kind of Internet Lick envisioned in the 1960s.
CLARK: Some of the papers he wrote back then are
amazing for their predictive quality. He talked about interactive chat.
He talked about the Web — although he didn’t call it the Web, of course.
We didn’t have that name for it. But he talked about services online,
banking and financial services. He talked about games, shopping online.
All of these ideas were floating around as potential applications once
we built this global network of computers, and that was well-articulated
in the 1960’s. What we were doing in the 1970’s was trying to build it.
WERTH: But, people like Clark didn’t have any experience in building a
network for sharing information at that scale. At the time, the only
thing that had come close had been built by telephone companies. And he
says they had a very specific use for that network.
CLARK: It was for carrying phone calls. And there’s a
major distinction between the Internet and the telephone system, which
is that telephones are pretty stupid.
WERTH: Meaning that a telephone itself doesn’t really do much; it’s
the networks they’re connected to that have all the smarts. Telephones
are really only good for making and receiving phone calls.
CLARK: All the intelligence in the telephone system
is in the switches. That’s where phone numbers are translated. That’s
where calls are forwarded. That’s where resources are allocated.
WERTH: Computers, however, aren’t stupid. They can perform a lot of
different functions.They’re what engineers call “general-purpose
devices.”
CLARK: And those of us who were building the Internet
wanted to build a general-purpose network to hook general-purpose
computers together.
WERTH: A network that wasn’t focused on one specific company the way
the phone network at the time was built around AT&T or focused on
one specific function. They wanted it to be versatile. For example, the
first system for sending emails was invented by a computer programmer
named
Ray Tomlinson
in the early 1970’s. He designed it for the people working on ARPAnet
so that they’d have an easier way to communicate with each other. And
for a long time after it was commercialized , Clark says email was the
innovative application.
CLARK: Back then to say that you were on the Internet
meant that you had an email address. So we could have built a network
that was dedicated to email, but we didn’t want to do that. We wanted to
build a network that was general, because that made it possible to
introduce new applications without modifying the network.
WERTH: And to accomplish that, Clark says they needed to do the
opposite of what telephone companies had done. In this case, it was the
network, not the devices, that needed to be stupid.
CLARK: In fact, we wanted to build what we have
called the “dumb net.” And our whole goal was to make sure that the
network was completely ignorant of what the user was trying to do. It
just sees data come in one side, and it forwards the data and delivers
it to the other side. It’s up to you what that data is. And, in that
respect, the core of the Internet is very, very simple because we
assumed that all the smarts — the application code and stuff like that —
was running on these computers at the edge. So in some sense, we took
the design principle of the telephone system and turned it inside out.
That’s basically what occupied us during the 1970s. What happened in the
1980s is that we had to deal with issues of scale. We had a sense of
hooking up all the computers in the world. The thing that confused us
was we misunderstood how many computers there were going to be. And then
the personal computer comes along. And there was this moment when we
realized, “Oh, there are not going to be hundreds of thousands of
computers. There are going to be hundreds of millions of computers.”
WERTH: Hundreds of millions of computers. The 1990’s, of course, saw the introduction of the
World Wide Web.
And we should be clear: the Web and websites that you visit online are
accessed through the Internet. They are two different things, even
though we often use those terms interchangeably. By 1995, according to
the
Pew Research Center, 14 percent of Americans were using the Internet. And for the people who ventured online in those early days …
YOCHAI BENKLER: It was exhilarating.
WERTH: That is
Yochai Benkler. He’s a professor at Harvard Law School and a director at the
Berkman Center for Internet & Society.
BENKLER: The experience of being a user in that
earlier moment was very much one of discovery, of constantly being able
to stumble on new things. That part of it was exhilarating.
WERTH: People were writing on the
web, creating their own websites and new online communities, all of
which began disrupting the institutions — media and commerce and so on —
that had been in place for decades.
BENKLER: There were well-defined control points or
tollbooths that someone could sit at and decide: “You get to innovate,
you don’t get to innovate. You get to communicate with people you want
to communicate, you don’t get to communicate.” That changed dramatically
with the Internet, and you saw a radical decentralization of the
ability to speak and innovate and communicate without permission. Power
was decentralized.
WERTH: And that’s basically how we think about the Internet today:
disrupting, and sometimes replacing, traditional institutions and
industries. But as the Internet has grown, and the number of global
users has topped
3.2 billion people, Benkler
says there’s been this really worrisome shift, with a handful of
companies — Facebook, Apple, Google — that now have an outsize influence
on how we all use the Internet.
BENKLER: What we have seen in the last seven or eight
years is that around those open spaces we created in the first two
decades, a new set of control points is beginning to emerge on the net.
TUFEKCI: We’re seeing the birth of a new center of power, real power. We
depend on these technologies that have been in many ways wonderful and
fascinating, and they’ve greatly enriched my life, but these companies
have major power, and they’re making significant decisions unilaterally.
So there’s a whole host of new questions that haven’t even been
explored.
WERTH: Like whether we’re trading that open and exhilarating Internet
for a more insular, siloed Internet experience — like the way Facebook
prefers the Ice Bucket Challenge over police protests, showing us things
it thinks we’ll like instead of what might actually be important. Or,
Yochai Benkler says, take the seemingly innocuous shift from the desktop
to the mobile Internet.
BENKLER: More than half of Internet access, if you think of time and particular uses, is now moving to the smartphone.
WERTH: Which has also prompted Internet users to migrate from
general-purpose web browsers to single-purpose apps: your Facebook app,
your Twitter app, your
Wall Street Journal app
.
BENKLER: When browsers first came out in the 1990s,
the idea was: here is a universal standard for describing what it is you
want to say or show and if you use that universal language, then anyone
using any device can implement this reader. And essentially what the
browsers did was they decentralized power from the operating system. If
you were writing for Windows, you needed to write for Windows in this
way. If you were writing for Apple, you needed to write for Apple in
that way. Once you could write something on the web, you could write to
this general purpose reader, the browser, and anything could run. What
happened with the app is that you got special-purpose containers, if you
will, for every kind of content. So you shifted from a general-purpose
platform that’s based on open standards and anyone can write what they
want, to a platform that says, “Write a very special program that fits
only your data.” It’s a complete transformation.
WERTH: To understand that transformation, Zeynep Tufekci says, just
look at the very-proprietary iPhone ecosystem that’s been created by
Apple.
TUFECKI: It designs the phone, and then it designs
the operating system, and then it decides what apps are available on the
operating system or not.
WERTH: iPhone users have to go through Apple’s App Store to add
almost anything to their devices. And if you’re one of the hundreds of
thousands of outside developers who produce apps for the iPhone,you
can’t just put your app out there. There’s an
approval process. And Yochai Benkler says Apple has been known to reject apps that it finds objectionable:
BENKLER: For example, someone developed a game
that essentially criticized the manufacturing conditions and the worker
conditions at the Chinese company Foxconn that was putting together the
iPhone. That got banned on the App Store and removed from the App
Store.
TUFEKCI: There was another app that was blocked that sent out a notification every time a drone had been used
to kill people. So, it was a political statement to say, “Look, this is
how much we’re using drones to do this.” And the App Store wouldn’t
approve this.
WERTH: Apple, we should mention, has about a
40 percent share of the market for mobile-operating systems in the United States and roughly 25 percent globally.
TUFECKI: Which means if you’re not in the App Store you’re done, because you lost every Apple phone out there.
WERTH: And while these may sound like fairly minor examples, Benkler
says, consider the deal that Apple made to restrict the use of
Skype when it first became available on the iPhone
in 2009. At the time, AT&T was the iPhone’s exclusive carrier. And
AT&T essentially told its customers, “by the way, we have this rule
about Skype.”
BENKLER: “You can’t run Skype unless you are on a Wi-Fi network.”
WERTH: Skype was not allowed on AT&T’s cellular spectrum, which
Benkler says was unfortunate because Skype was this breakthrough
technology. It had finally figured out a way to carry voice calls over
the Internet.
BENKLER: Skype innovated in a way that
basically was by many people thought theoretically impossible. They
didn’t need permission. If they had needed permission from somebody who
runs voice services to put the application on, they would have not
gotten permission because it would have been considered too poor, too
weak, can’t work. And yet it did. But when Skype comes on the iPhone the
first time, AT&T doesn’t want it to be run in competition with its
voice service.
WERTH: And to be clear, this app-approval business is not confined to
Apple. Benkler says Google has a long list of reasons it can reject an
app for Android, its own mobile-operating system. And the same goes for
Facebook, which has a large community of developers who build products
and services around Facebook.
BENKLER: You have to persuade Apple or Google and Facebook to let you in. That’s the critical point.
CLARK: And this is a very different sort of industrial structure than we imagined when we built the Internet. So there’s a kind of regulation or curation of the Internet experience.
WERTH: And he admits that curation
can have advantages. More control can make it harder for malicious
software to get through. Clark says security is perhaps one thing he and
others should have done differently when the Internet was first being
developed.
CLARK: One decision we made is that we wanted to make
it possible for any computer to establish a connection to any other. On
the other hand, it opens up a range of problems when all those other
computers in the world try to attack you. Maybe it wasn’t as important
as we thought that it be equally easy for every computer in the world to
talk to every other one. And in fact, I would say we sort of
over-empowered the user, if I can say that.
WERTH: And maybe those security concerns — whether they’re from
hackers or terrorists or government institutions — are what’s driving
some of the shift towards a more regulated, curated Internet
environment. The concern, Clark says, is that the degree of control that
offers will simply allow companies to exploit their market power to
make more money.
CLARK: And what we see today with the powerful players — and I consider Google and Apple and Facebook to be quite powerful players — is, I would say, a carefully thought-through and not-surprising attempt to build durability into their dominance.
BENKLER: And then the question becomes, where does the next
Google, where does the next Skype, where is the next platform going to
come that won’t need permission of somebody who is already there in
power to let them in?
* * *
