The early history of the Internet was affected by this control. As described by
John Naughton in A Brief History of the Future
(1999), an early design idea for the Internet was
proposed to AT&T by RAND researcher, Paul Baran, in the early 1960s. Resistance to his design
was strongest from AT&T. As Naughton reports, Baran recalls one particularly telling instance
[AT&T’s] views were once memorably summarised in an
exasperated outburst from AT&T’s Jack Osterman after a long
discussion with Baran. ‘First,’ he said, ‘it can’t possibly work, and
if it did, damned if we are going to allow the creation of a
competitor to ourselves.’5
This resistance is perfectly understandable. From AT&T’s perspective,
maximizing its control over its network was no doubt profit maximizing. And we should expect
corporate entities to behave in a profit maximizing manner. But this resistance was profit
maximizing only because AT&T was in control of the network uses. Or in other words, only
because the network was not “End-to-End.” Had the network been End-to-End, it would have
had no incentive to disable one use of the network it controlled in favor for another.
The same point about the relationship between innovation and the concentration
of control can be made more obviously about the Internet in foreign countries. It is no accident
that the Internet was born in the United States, since in practically every other nation, the
telephone architecture was controlled by state sponsored monopolies. These monopolies, no less
than AT&T, had no interest in facilitating the design of a network that would free individuals
from that control. For much of the 1990s, it was a crime in parts of Europe to connect a modem
5John Naughton, A Brief History of the Future107 (1999). Authors Katie Hafner and Matthew Lyon recount a
similar resistance in Where Wizards Stay Up Late 52-66 (1996).
to a telephone line. Even today, the franchise in Germany for public phones permits the provider
to control how access to the Internet occurs.
The Government’s Role in Creating the Competitive Environment for the
Internet
It is fashionable today to argue that innovation is assured if government simply
stays out of the way. The FCC’s hands-off policy to date appears largely to be motivated by this
prevailing ideological vogue. The view is that the best way for the government to guarantee
growth in Internet broadband is to let the owners of networks architect broadband as they see fit.
We believe this view is misguided. It ignores the history that gave the Internet its
birth, and threatens to reproduce the calcified network design that characterized our
communications network prior to the Internet. The restrictions on innovation that marked the
AT&T telephone monopoly were not removed by the government doing nothing. They were
removed by active intervention designed to assure the possibility for innovation. It was the FCC
and Department of Justice that cut the knot that tied innovation on the telecommunications
network to the innovation favored by AT&T. It was their action that eventually freed the network
from the control of a single strategic actor, and opened it up for the innovation of many.
Beginning with the Carterfone decision in 1968,6the FCC increasingly pursued a
policy that forced AT&T to open its network to competing uses and providers. In a series of
decisions, the FCC required that AT&T permit alternative uses of its network. In 1984, actions
by the Antitrust Division forced AT&T to unbundle its long-distance service from its local
telephone service. This unbundling was effected through a decree that lead to the breakup of the
largest monopoly in American history.
These actions together transformed the telephone network from a network whose
use was controlled by one company — AT&T — into a general purpose network, whose ultimate
use was determined by end users. In effect, they imposed a principle analogous to End-to-End
design on the telephone network. Indeed, though it masquerades under a different name (“open
access”), this design principle is part and parcel of recent efforts by Congress and the FCC to
deregulate telephony. The fundamental economic goal of the FCC in deregulating telephony is to
isolate the natural monopoly component of a network — the actual wires — from other
components in which competition can occur. By requiring the natural monopoly component at
the basic network level to be open to competitors at higher levels, intelligent regulation can
minimize the economic disruption caused by that natural monopoly, and permit as much
competition as industry structure will allow.
It is our view that but for these changes brought about by the government, the
Internet as we know it would not have been possible. Without these changes, the trend in
telecommunications was towards more centralized control over the communication network.
Network theorist Robert Fano of MIT, for example, wrote in 1972 that unless there was a change
in the trend in the computer-communications network, existing institutions would further isolate
computer and communications technologies from broad based control.7But by seeding the
development of a network within a different communication paradigm, and then opening the
existing communication network so that it might deploy this different communication paradigm,
the government created the conditions for the innovation that the Internet has realized.
6In the Matter of Use Of The Carterfone Device In Message Toll Telephone Service; Docket No. 16942; 13
F.C.C.2d 420; June 26, 1968.
7See, Robert M. Fano, On The Social Role of Computer Communications, 60 Proceedings of the IEEE, September
1249 (1972).
This is not to say that the government created the innovation that the Internet has
enjoyed. Nor is it to endorse government, rather than private, development of Internet-related
technologies. Obviously, the extraordinary innovation of the Internet arises from the creativity of
private actors from around the world. Some of these actors work within corporations. Some of
the most important have been associated with the Free Software, and Open Source Software
Movements. And some have been entrepreneurs operating outside of any specific structure. But
the creativity that these innovators have produced would not have been enabled but for the
opening of the communications network. Our only point is that the government had a significant
We do not claim that no communication network would have been possible
without the government’s intervention. Obviously, we have had telecommunication networks for
over a hundred years; and as computers matured, we no doubt would have had more
sophisticated communication-computer networks. But the design of those networks would not
have been the design of Internet. The design would have been more like the French “equivalent”
to the Internet — miniTel. But miniTel is not the Internet. The miniTel is a corporate,
centralized, controlled version of the Internet. And it is notably less successful.
The Relevance of Legacy Monopolies
As we have said, no one fully understands the dynamics that have made the
innovation of the Internet possible. But we do have some clues. One important element of that
innovation is a structure that disables the power of legacy monopolies to influence the future of a
By freeing the telecommunications network from the control of one single actor,
the government enabled innovation free from the influences of what one might call “legacy”
business models. Companies develop core competencies, and most of them tend to stick to what
they know how to do. Companies faced with a potential for radical change in the nature of their
market may recoil, either because they don’t know how to change to face changing conditions, or
because they fear that they will lose the dominance they had in the old market as it becomes a
new playing field. Their business planning is, in short, governed by the legacy of their past
success. These legacy business plans often affect a company’s plans about how to respond to
innovation. In a competitive environment, they will often disadvantage a company that fails to
respond rapidly enough to changed circumstances.
In some markets, companies have no choice but to respond to changed
circumstances. They either change or die. It is a mark of Microsoft’s success, for example, that
its chairman, Bill Gates, succeeded in radically altering the course of the company’s
development, in the face of changed competitive circumstances, despite the fact that such
changes resulted in the termination of projects at other times deemed central to Microsoft’s
future (MSN, for example.). In contrast, for example, commentators attribute Apple’s failure
during the early 1990s to its refusal to give up old models of business success. Legacy models
hindered Apple’s development; the refusal to be captured by legacy models was a key to
In an environment where a company has power over the competitive environment
itself, however, the rational incentives of a business may be different. If the business, for
example, has control over the architecture of that competitive environment, then it will often
have an incentive to design that architecture to better enable its legacy business models. As
Charles R. Morris and Charles H. Ferguson describe it,
Companies that control proprietary architectural standards have an
advantage over other vendors. Since they control the architecture,
they are usually better positioned to develop products that
maximize its capabilities; by modifying the architecture, they can
discipline competing product vendors. In an open-systems era, the
most consistently successful information technology companies
will be the ones who manage to establish a proprietary
architectural standard over a substantial competitive space and
defend it against the assaults of both clones and rival architectural
sponsors.
A company in this position can and will resist change, in order to keep doing what it knows best.
This was the problem with the telephone company prior to its break up by the
government. The telephone monopoly enjoyed substantial returns from its existing network
architecture. The fear of regulators was that these returns would make it unwilling to experiment
with other architectures that might better serve communication needs. As we have said, there is
at least some evidence that AT&T in fact resisted the emergence of the Internet because it feared
its effect on AT&T’s own business model. Certainly it resisted the development and
interconnection of other technologies to its telephone network. The regulators who pushed to free
the telecommunication network believed that the market would choose differently from how
AT&T, as controller of the network, would choose.
Time has proven these regulators correct. Once freed from the strategic control of
an entity that had a particular business plan to protect, the communications network has evolved
dramatically. The competitive process was enabled by making the network neutral about its uses,
and by giving competitors access to the network so that they could compete about its best use.
The same wires that AT&T used to send analog voice only are now being used to deliver stock
8Charles R. Morris and Charles H. Ferguson, How Architecture Wins Technology Wars, Harvard Business Review
86, 88 (March April 1993) (emphasis added).
quotes, music, fantasy games, reference information — in short, the whole content of the
The lesson from this explosion of innovation is critically important. An
architecture that maximizes the opportunity for innovation maximizes innovation. An
architecture that creates powerful strategic actors with control over the network and what can
connect to it threatens innovation. No doubt these strategic actors might behave in a pro-
competitive manner. There is no guarantee that they will interfere to stifle innovation. But
without competition or regulation to restrict them, we should not assume that they willsomehow
decide to act in the public interest.
The Proposed Merger of AT&T / MediaOne
The Threat Posed by Bundling
As we stated at the start, we do not question the merits of a merger between
AT&T and MediaOne in principle. AT&T’s argument that such a merger will enable much
greater competition in local telephony may prove persuasive; the efficiencies of a merger for the
supply of broadband access may prove persuasive as well. Our sole concern is the architecture
that AT&T and MediaOne propose for broadband access. As they have described in their papers,
cable broadband will prevent users from selecting an Internet Service Provider (“ISP”) of their
choice. Instead, access will be technologically bundled with ISP service. The network will thus
discriminate in the choice of services that it allows. This kind of discrimination may have
profound consequences for the competitive future of the net.
To see the problem with this architecture, we must first understand the importance
of an ISP. ISPs serve a number of functions in the existing narrowband residential market. Some
ISPs focus primarily on access to the Internet. Customers, through their local telephone
exchange, connect to the ISP; the ISP serves Internet access at speeds limited only by the local
telephone exchange. Some ISPs supplement this access with promises of user support — both
the support to assure the Internet connection is maintained, and in some cases, support with the
use of certain Internet applications. Some ISPs further supplement this access with server
capabilities — giving users the ability to build web pages on the ISPs servers, or support more
expansive Internet activities. And finally some ISPs provide, or bundle, content with access. The
most famous of these is America Online, but other ISP/content providers have included
This existing narrowband residential market is extraordinary competitive.
Customers have a wide range of needs that customers have in this market; the market responds to
this range of needs with different packages of services. Nationwide there are some 6,000 ISPs. In
any particular geographic region, there can be hundreds that compete to provide service.
The functions performed by ISPs, however, are not fixed. They have no inherent
“nature.” Hence as bandwidth changes from narrow to broadband, we should expect the range of
services offered by ISPs to change. As throughput becomes more critical in video services, for
example, we could imagine ISPs competing based on the caching services they would offer. Or
as the character of the content available increased, we might imagine some ISPs catering to
certain content (video content) while others specialized elsewhere (new users).
The functions of ISPs, then, must not be conceived of too narrowly. Their
importance, for example, has little to do with hosting “home pages” on the World Wide Web, or
the portal sites they might now provide. Their importance is in the range of services they might
bundle and offer competitively — from content (including video and audio services) to help
functions to reference functions to special caching needs. In short, ISPs are engines for
innovation in markets we do not yet imagine.
These ISPs thus could be important middle level competitors in the Internet
economy. They could provide an ongoing competitive threat to actors at their borders. In the
terms defined by Timothy Bresnahan, ISPs could become “vertical competitors” in an industry
marked by highly concentrated markets at each horizontal level.9Thus AOL, a traditional online
content and ISP, is now a potential threat to Microsoft in the operating system platform market.
This threat could not have been predicted three years ago. But by maintaining the fluidity of
borders, the net preserves the potential for new forms of competition.
This layer of potential competition is especially important given how little we
know about how the broadband market will develop. The Internet market generally has been
characterized by massive shifts in the competitive center. Hardware companies (IBM) have been
displaced by operating system companies (Microsoft); operating system companies have been
threatened by browser corporations (Netscape) and by open-platform “meta”-operating systems
(Sun’s Java). As Bresnahan notes, we have no good way to know which layer in this chain of
services will become the most crucial. Thus, multiplying the layers of competition provides a
constant check on the dominance of any particular actor. Again, Bresnahan: “Far and away the
most important is that competition came … from another horizontal layer.”1 0Thus as he
10Id. at 6.
recommends, “one (modest) policy goal would be to make the threat of [vertical competition]
The architecture proposed by AT&T/MediaOne for their broadband cable service
threatens this vertical competition. By bundling ISP service with access, and by not permitting
users to select another ISP, the architecture removes ISP competition within the residential
broadband cable market. By removing this competition, the architecture removes an important
threat to any strategic behavior that AT&T might engage in once a merger is complete. The
architecture thus represents a significant change from the existing End-to-End design for a
crucial segment of the residential Internet market. Further, there is in principle no limit to what
AT&T could bundle into its control of the network. As ISPs expand beyond the functions they
have traditionally performed, AT&T may be in a position to foreclose all competition in an
increasing range of services provided over broadband lines.
AT&T and MediaOne would achieve this change by bundling technologically.
The consequence of this bundling will be that there will be no effective competition among ISPs
serving residential broadband cable. The range of services available to broadband cable users
will be determined by one of two ISPs — @Home and RoadRunner, both of whom would be
allied with the same company. These ISPs will control the kind of use that customers might
make of their broadband access. They will determine whether, for example, full length streaming
video is permitted (it is presently not); they will determine whether customers might resell
broadband services (as they presently may not); it will determine whether broadband customers
11Id. at 18.
![[made with GoClick]](imgs/goclick.gif)
