The Paradox of the Best Network

David Isenberg and David Weinberger

The communications revolution has been thwarted

Just a few short months ago, it seemed that humanity stood on the edge of a communications revolution. New technology promised to topple barriers of space and time. We were on the verge of inventing new ways for the world to work and play together. We were giddy with possibility. Now a grim face replaces yesterday's optimism. Prospects of new connectedness recede as capital markets tighten, existing telephone companies back off on capital expenditures, established communications equipment suppliers falter, and ambitious new telecom companies fail.

But not by technology

Despite the darkened outlook, new communications capabilities are within reach that will make the current Internet look like tin cans and string. The technical know-how exists. Radically simplified technologies can blast bits a million times faster than the current network at a millionth of the cost. These are sitting in laboratories undeveloped, in warehouses undeployed, and in the field underutilized.

Or even by greedy telcos

It's not even that the communications revolution has been derailed by inept or self-aggrandizing behavior by incumbent telephone companies and their government regulators. Something more fundamental is at work. The situation has been shaped by a paradox inherent in the very nature of the new technology:


The best network is the hardest one
to make money running. 1

A paradox has stopped us.

This is the Paradox of the Best Network. It lies beneath the sudden stoppage of infrastructure innovation and growth in 2001. It provokes incumbent companies to mass lawyers and lobbyists to thwart the development of a competitive communications market. It causes investment capitalists to drive their stakes into firmer economic ground far away from telecommunications.

The best network has the fewest added features and functions

The paradox arises from the meaning of "best." If "best" meant, "generate the most cash for the network owner," there would be no paradox. But if we accepted this meaning of best, we'd have to be content with the tightly-controlled, relatively thin stream of bits that the telephone companies currently grant us. Communications networks have a more important job than generating return on investment — their value comes from their connectivity and from the services they enable. Therefore, the best network delivers bits in the largest volumes at the fastest speeds. In addition, the best network is the most open to new communications services; it closes off the fewest futures and elicits the most innovation.

The best network just moves bits

Designing a network that is intelligently tuned (optimized) for a particular type of data or service — such as TV or financial transactions — inevitably makes that network less open. As software engineers say, "Today's optimization is tomorrow's bottleneck." Thus, the best network is a "stupid" network that does nothing but move bits.2 Only then is the network truly open to any and all services that want to use it, no matter how innovative or how unexpected. In the best network, the services live at the edges of the network and use the network to transport bits; they do not rely on any special characteristics of the network itself.

But a stupid bit-moving network is a commodity

The Paradox of the Best Network comes about because as a network gets stupider, connectivity becomes a commodity. Those who own and operate the network have less to charge for. After all, they're just moving bits. The high-value services, the ones that command premium prices, reside at the edge of the best network. Because the best network is simple, it is low-cost to operate. In a competitive market, this means it is low priced. Low price also lowers barriers to innovation at the edges of the best network.

The telcos like smart networks

The telephone companies are impaled on the horns of this dilemma. Historically, their high-margin services have been built into the middle of their network, which has been optimized for a single application — voice. Their business is based on this special-purpose network. They know that implementing the new commodity network threatens the very basis of their business.

The Internet succeeds because it's stupid

In contrast, the Internet is not optimized for any specific application. Its strength is its generality. It's designed simply to move bits across all kinds of wired and wireless infrastructures. As a result of this simplicity, the Internet has proven to be the most scalable, most robust communications infrastructure humans have ever built. It has proven itself effective at encouraging innovation:  of all the winning networked applications of the last decade — email, web browsing, instant messaging, chat, music sharing, streaming audio, ecommerce, etc. — every one appeared on the Internet. Not one was invented by a telephone company. And not one needed any special mechanism within the network itself

The Internet's success threatens the telcos

This fact frightens the telephone companies. It should. The Internet's bits-are-bits simplicity even threatens to turn their cash cow — voice telephony — into something anyone can do just by installing simple software onto an everyday PC. Hook a PC to a high-bandwidth, always-on connection and anyone can make high-quality Internet phone calls without telephone company involvement. Further, innovations like document sharing, collaborative whiteboarding, and add-on video conferencing, which are difficult on the old telephone network, are relatively easy additions to an Internet telephony program. Because the Internet is a commodity network, Internet telephony is cheaper. Because it's a stupid network, innovation is easier. Further, the value is added at the edge of the network, outside of telephone company purview.

But the Paradox is the real threat

But, the real threat to the incumbent telephone companies isn't the Internet. It's the Paradox of the Best Network. The paradox means that companies that run the old, closed, special-purpose telephone network have an unfit business model for running the new network. No amount of technological upgrading will fix this. To survive, the incumbents must become different businesses. But there's no guarantee that they'll be the best companies to run the best network.

Cable and DSL aren't the answer

Established communications companies have tried to distract our attention with DSL and cable modems, as if these would complete the new network. But these are crippled compromises at best, touted precisely because they are not disruptive. They milk already-depreciated assets without overturning established business models. And that is precisely why the current communications companies are pushing them so hard.

Bandwidth could be plentiful and cheap

There are alternatives. Incumbent communications company clout has forestalled delivery of a variety of radically simplified, extremely affordable technologies — from software-defined radios arrayed in self-organizing architectures to Ethernet-over-fiber-optics — that are storming the gates of the telephone companies' existing network. These promise every home more bandwidth than a medium sized town uses for all of its conventional telephony — for about the price of a monthly bus pass. These will be developed and deployed wherever established companies hold less sway.

Bandwidth threatens content providers, too

Telephone companies are not the only institutions goaded by new network technology. We can see from the reaction to today's Internet that the Paradox of the Best Network is not kind to the recording industry, to book publishers, or to any other group that makes its living by controlling access to content. These groups have already called in the lawyers and lobbyists to protect their current business models. Nor will the new network be popular with any institution — economic, political or religious — that seeks to shield itself from conflicting cultures and ideas.

This is about politics, not just business

In fact, the best network embodies explicit political ideals — it would be disingenuous to pretend it didn't. The best technological network is also the most open political network. The best network is not only simple, low-cost, robust and innovation-friendly, it is also best at promoting a free, democratic, pluralistic, participatory society; a society in which people with new business ideas are free to fail and free to succeed in the marketplace.

It's about our destiny

More than the fate of telephone companies is at stake. We must not allow the short-sighted self-interest of the incumbent telecommunications industry to thwart the connectedness that will enlarge us as social creatures. Our destiny as a species has always been to be connected. The new network — open, fast and out of control — will change what is most important about us lonely humans: the way we join together with others to become more than we are alone.

So, how do we do it?

But the best network is the hardest to make money running. So who builds it? Who runs it? Who fixes it when it breaks? And who develops the next generations of faster, simpler infrastructure?

Let the government do it?

Arguably, building the best network is a Public Good. It will boost the economy, open global markets, and make us better informed citizens, customers and business people. So, perhaps we should let the government do it. Perhaps we should insist that the government do it.

Big government is bad

But governments tend to make mistakes. Big governments tend to make big, costly, persistent mistakes. We do not want government to lock us into particular technologies or certain ways of doing things, no matter if they seem to be the most promising technologies and methods today.

A purely open market approach won't work either

A purely governmental solution, therefore, is too risky. But so is a pure reliance on the invisible hand of the market. Left to itself, the market would favor larger network owners both because they benefit from economies of scale (the more connections you provide, the lower each connection costs) and because they have financial resources to withstand the low operating margins of a commoditized market. Even starting from a mythical "level playing field," larger network owners would acquire smaller ones. And once again, large carriers would become monopoly-like, with little incentive to hook up less-populous and poorer areas.   More important, these regenerated monopolies would be as loathe to open their network or to invest in new technology as the current crop of telephone company incumbents.  The Paradox of the Best Network is not resolved by the free market; indeed it is a consequence of it.

Here are some ideas

So, we are stuck between the Scylla of big government and the Charybdis of free market dynamics. We need to find wise ways to proceed. If we don't, telephone company lobbyists will write the next chapters of the communications story.


We propose that:

Set a goal

The U.S. government should set a national goal: every citizen will have high bandwidth, open access to the Internet within five years, beginning with schools and public buildings, then businesses, other private enterprises and homes.

Separate wire and services

Just as the Internet separates transport from service, the incumbent telephone companies should be separated into transport companies and service companies. The transport companies would be have government incentives (e.g. , assured return on investment), to make fiber, pole attachment, and right of way available to all service providers. The service companies, in contrast, would be completely deregulated and freed to compete with other service providers in the newly revitalized marketplace at the edge of the network.


In addition, since we do not yet know what kinds of initiatives will succeed in bringing high speed connectivity to our homes and businesses, a variety of experimental initiatives should be encouraged. One such proposal, end-user financing, springs from the fact that because high-speed Internet connectivity costs just a few thousand dollars per home, home owners should be able to buy their own connectivity. Another proposal would make special homeowner loans available for this purpose. Still another, condominium ownership of fiber, would permit private groups to install, own and operate their own connectivity.

Get out of the way

In addition, municipalities, utilities, and other non-traditional entities should be encouraged to provide high-speed connectivity. Some states have laws that prohibit such entities from owning or operating communications networks; these should be swept off the books.

Help the municipalities

Some municipalities and rural areas may need additional help providing connectivity where economic conditions are less favorable or low population density means higher per capita infrastructure costs. The Rural Electrification Act of 1936 provides one demonstrably successful model for this.

Keep the IP stupid

Even without high speed connectivity, the Internet Protocol has been a powerful source of innovation. As the Internet has assumed economic importance there have been calls to alter the Internet Protocol, e.g. , to increase security or strengthen copyright protection. In virtually every case, such concerns can be satisfied at the edge of the network or by adding more bandwidth to the network. The Internet Protocol itself must remain simple, stupid and best.

Restore the role of copyright

These steps to keep the infrastructure open will be in vain if vested interests use other means (legal and technological) to expand restrictions to content and reduce free flow of ideas in the public domain. The laws governing copyright should be brought back to their original purpose to ensure the free flow of information that is critical to the functioning of a free, open society.




Steps like this must be taken if only to save the incumbent network from meeting the fate foretold by the Paradox of the Best Network. Our destiny is one short mile from us. We need only connect.

Fail Fast

Click here for the open letter to the FCC urging that the telcos be allowed to fail fast





"Rise of the Stupid Network"
D. Isenberg

End-to-End networks
J.H. Saltzer, D.P. Reed and D.D. Clark

Connectivity: What Is It and Why Is It Important?
Bob Frankston

The Stupid Network: Essential Yet Unattainable
A. Odlyzko

Privacy, Economics, and Price Discrimination on the Internet
A. Odlyzko

"The End to End Argument"
D. Reed
Industry consortium

"Who's Holding Back Broadband?"
Lawrence Lessig



# # #

1 We first encountered this formulation in the September 2001 issue of Roxane Googin's High Tech Observer. She wrote, "The perfect network is perfectly plain, and perfectly extensible. That means it is also the perfect capital repellant, [which] implies a guaranteed loss to network operators, but a boon to the services on the 'ends'." 

2 See "The Rise of the Stupid Network" by David S. Isenberg, Computer Telephony, August 1997, pp. 16-26,