The Internet is Broken!

So just build a new one! - but there exists a lack of wider perspective in Stanford University�s
*Clean-Slate Design for the Internet white paper

[*Note1: The Standford University White Paper referred to here is no longer available on their web site. Even so, this discussion raises many important issues surrounding new internet design, and there is nothing currently on Stanford's Clean Slate web site to indicate a shift in policy regarding their research program since the white paper was published.]

[*Note2: Jan 2008: Someone was kind enough to send me a copy of the PDF]


Imagine that your telephone worked like the internet. 90% of all incoming calls would be people trying to sell you pills to enhance your sex life, selling pornography, giving you �hot� stock tips, or calls from Nigeria asking your help in recovering large sums of money for a share of the proceeds. Many of these scam artists calling you will identify themselves using names of people you know.

Your answering machine would soon become infected by a virus, and report to some criminals in Russia who control vast networks of answering machines. It would begin sending recorded messages spamming everybody you know during the night while you sleep.

When you pick up the phone and dial your bank, without realizing it instead you will be connected to an imposter who sounds just like your bank manager, asking you to confirm your account and password details. The next day, you would discover all the money withdrawn from your account.

There may be times when it is impossible to dial out from you company�s lines because of endless incoming calls that just hang up when you answer, for days on end. You will then receive a threatening letter in the mail saying that if you ever hope to able to receive calls from clients again you must send money to some offshore account. You will quickly decide to give in and pay the extortion rather than be forced out of business.

Sexual predators will call your children for a �friendly little chat� if you are not constantly watching over their shoulders when they are near the phone.

There will be times when your telephone conversations will be frequently interrupted and times when the line goes suddenly dead in the middle of a call. Even worse, you never know when someone may be listening in on your call.

Spam, scams, bot nets, phishing, denial of service (DoS) attacks, identity theft, unreliable service, hacking attempts - the problems grow day by day. We shut down one vulnerability, and the criminals just find another exploit.

Worst of all, we have increasing allowed our entire economy to become dependent on the internet in spite of the fact that it has simply become too vulnerable to be relied upon. The next time there is a serious international conflict, the internet will become a target, and our economy will be crippled in the first attacks. Even a concerted effort by a large group of terrorists could bring down the internet and force our economy to its knees.

The things we use such as Firewalls and NATs to adapt the 30 year old internet to modern realities inhibit new applications or at a minimum require convoluted work-arounds for things like voice and video communications, applications that were inconceivable when the internet was born.

We do have a new internet standard coming on line - called �IPv6�, but this only addresses a few technical problems like the fact that we are going to run out of internet address in the next seven or eight years. It cannot begin to solve the all the problems alluded to above and many others too technical to discuss here. The internet is broken.

Why not rebuild the Internet from scratch? Now that we have some thirty years of experience with the old one, what a difference we could make in designing a new one, while at the same time having a much better understanding of how to build a network that will sustain continuing evolution on into the future. Stanford University has come out with exactly this bold and exciting idea in their white paper - "Clean Slate Design for the Internet" (PDF - 18 April, 2006) - that I strongly advise everyone to read. It is just too important to be left up to academics and big business.

From the overview: �The proposed program will focus on unconventional, bold, and long-term research that tries to break the networks ossification.� �In the spirit of past successful inter-disciplinary research programs at Stanford, the program will be driven by research projects from the ground up. Rather than build a grand infrastructure and tightly coordinated research agenda, we will create a loosely-coupled breeding ground for new ideas.���our goal is to be flexible, creating the structure and identifying and focusing funds to support the best research in clean-slate design.� �The program will collaborate with, and be funded by, approximately seven industrial partners with interests in networking services, equipment, semiconductors and applications.� (such as Cisco Systems and Deutsche Telekom).

There are a few essential things missing from the Stanford proposal. I didn�t see anything to suggest that they are looking for this to be a truly international collaboration. If it isn�t, that would be a very short sighted omission. I only saw the name of one woman listed among 17 faculty and staff members dedicated to this research. I would suggest that they could benefit from a more balanced representation of the human race. Also needed are the inclusion of social scientists capable of researching the making of value judgments and decisions about how the proposed new internet can encourage social inclusion and break down the digital divide, and political scientists who can suggest how the proposed new internet can guarantee the free flow of ideas and information, enhance democracy and promote international harmony where this does not conflict with democratic ideals.

Without input from a broader perspective that includes social values and the fact that in the end, the new internet must serve people, we will end up with a technological Frankenstein. This is too important to be left up to big businesses that have no other motive beyond maximizing profits for their shareholders. By failing to give a prominent place for the issues I have just mentioned right up front in the white paper, it seems certain that these questions are doomed to either being given a low priority or even to being completely overlooked in the final design.

In Stanford University�s white paper, they state Stanford is particularly suited to do this research because of their unique tradition of inter-disciplinary research. Though they are clearly proud of this, it turns out that the disciplines involved in this program are all technical or business and management oriented. The faculty draws from four departments: Computer Science, Electrical Engineering, Management Sciences and Engineering, and the Graduate School of Business.

I have no doubt that they will do some exceptionally innovative and exciting research in these areas. However, I will argue that this is not nearly sufficiently inter-disciplinary for the task of building a new internet.

As an example, an important part of the discussion in the white paper is about security issues. �Any future network design must be built with security in mind from the start.�

�[A] component of future networks may include tools to help law enforcement identify the origin of an attack. This is a controversial topic and the extent of such tools must be limited by an appropriate privacy policy�" "Support anonymity where prudent, and accountability where necessary�.

This requires a closer look. There will be many times where security will involve making trade-offs between �anonymity� on the one hand, and �accountability� on the other. This requires making the value judgment, what is the value of anonymity as opposed to the value of accountability? They admit �This is a controversial topic�.

Who is to decide the how these trade-offs should be made? Can computer scientists, engineers, or business leaders make value judgments and set privacy policy for us? If not, who is to decide? I leave that to the reader to ponder.

How can the internet contribute to strengthening democracy? How can the internet ensure that freedom of speech will be preserved? How do we deal with regimes that will want censorship capabilities built into the infrastructure? How do we deal with regimes who will want to use the internet to collect information on their citizens? What is the importance of building international collaboration and consensus?

What is the role of the internet in society? How can the needs of big business, government, and the military be balanced with the needs of society or the individual? What are the moral and ethical questions that need to be asked? How can the internet contribute to social inclusion? How can it better help third world nations out of poverty? Are these last two valid questions to ask?

Reading the Stanford white paper alone, we could easily become alarmed. It simply does not address these issues. I began to dig more deeply into the subject...

Further research revealed that the Stanford program builds upon earlier research stretching back over seven years at least, and that many well informed and well intentioned people have devoted a lot of thought to many of these thorny questions.

As far back as July, 2000 a seminal paper �Developing a Next-Generation Internet Architecture� discusses goals and directions for a research effort aimed at developing a next-generation Internet architecture. It addresses most of the issues I have identified and more. The paper recognizes legal and public policy issues including intellectual property law, encryption export law, police surveillance, privacy and free speech, telecommunications laws, charging, and taxation.

It concludes �but our job is to concentrate on the technical requirements within this broader context. We note that the proposing team brings both Internet industry structure and economics research credentials and pragmatic experience within the changing Internet environment to the table, and has a track record of effectiveness working within this space.�

To answer my perception of a lack of international participation, the paper recognizes �these issues are all subject to national variation, since the Internet is worldwide", and states that we must be aware of these issues, but it goes on to say...

�Because the Internet is open, this is a classic role for a neutral party such as a government-funded research effort. Such an effort could easily be justified because the government is itself a sophisticated user that directly benefits from the generality of a coherent architecture and the platform that results. Once a draft architecture is completed, it will be useful to involve a larger subset of the research community. The team may catalyze this process by organizing workshops devoted to discussion of the Internet architecture. This task might be undertaken by a research group of the IRTF (Internet Research Task Force).�

Another paper, �FINAL TECHNICAL REPORT Future Generation Internet Architecture� came out a few years later, at the end of 2003. This document is the final report of a collaborative research project to explore the architectural fundamentals of the Internet. The goal of this project was to consider the following question: if we could now design the Internet from scratch, knowing what we know today, how would we make the basic design decisions? In this paper, they recognize �The emergence of conflicting interests�...

�Today, we see that the stakeholders in the Internet space often have interests that are adverse or conflicting. Users want to have a private conversation, while law enforcement wants the ability to carry out lawful intercept. Users want protection from spam; spammers try to circumvent their protections. Conspiring users share music; the rights-holders try to prevent this. The tussle among these various interests occurs in many ways: in debate over protocol design, in law and regulation, in technical choices made by different groups of users and providers, and so on. These conflicts, [are] perhaps intrinsic to the nature of society.�

�Design for tussle� must become an overarching objective of all architectural decisions we make for a next generation Internet. We must recognize and catalog techniques for designing for tussle, and make these techniques a basic part of the designer�s toolkit.�

They recognize that �There are often very important social and economic benefits of allowing anonymous behavior, but since it is hard to police a world with no accountability and there is little accountability without identity, systems that permit anonymous operation may have to impose much more restrictive technical constraints up front, to prevent rather than punish misbehavior.�

�Of course, this is a tussle space, not a given right, but many designers see the individual user as potentially at the mercy of �big business�, or �repressive government�, and tend to make design choices that favor the end-user when they can. We need techniques to tilt the playing field toward the end-user as we design a server architecture.�

�Include a concept of identity, and consider giving the receiving user control over how restrictive he wants to be in what he accepts. Sometimes, anonymous communication is socially or economically valuable.�

�However, a more general resolution of the security problem is hampered by a more fundamental limitation�a lack of agreement on what is meant by �security�, and what problem is to be solved. Security is not a simple goal where �more is better�, it is a complex tussle of requirements, defined by deep conflicts of values.�

The National Science Foundation (NSF) put out a funding proposal under Networking Technology and Systems (NeTS) in Fiscal Year 2006 inviting research proposals about all aspects of networking research, including future end-to-end Internet architectures. Future Internet Design (FIND) Funded projects will seek to design a next-generation Internet which they call the "Future Internet". One proposal is the following...

"NSF's Design for Utility and Social Needs: Multi-disciplinary research on social and policy issues in the design of mechanisms, such as balancing privacy, identity and accountability; designing for regional difference and local values, architecture for trust, and linking mutual trust to limits on activities; recommendations on policy and mechanism design for surveillance, identity tracking, and response to malicious behavior, in the context of multiple jurisdictions; design for utility in times of crisis including mechanisms for priority access, robust fall-back modes for the network and for applications, access to emergency services; design for usability such as self-configuration of network management and diagnosis at the consumer level, approaches to providing user-comprehensible error reporting. Proposals with a focus on societal needs should include researchers with expertise in networking as well as social science."

The Computerworld article, �Spam, viruses, botnets: Can the Internet be saved?� by Gary Anthes (February 12, 2007) states �Not all the projects focus on technology. One is studying the interactions and economic and social motivations of users, hackers, service providers and regulators, with the goals of improving service, security and user utility. � (initiative is part of a family of FIND projects at Stanford.)� An interview with Kahn, who�s now CEO of the Corporation for National Research Initiatives in Reston, Va., indicates he has concerns about privacy and censorship. �And it�s worth noting again that not all of the NSF�s FIND and GENI projects deal solely with technology. Another interviewee says �We are suggesting that a lot of the security projects should have a social collaborator,�

In conclusion, after further research, I am somewhat relieved. Clearly a rich discussion and investigation of essential political and social issues is firmly embedded in the body of research underlying the Stanford initiative, even if they don't feel it is worth addressing in their Clean-Slate program. Their research agenda appears to be entirely driven by big business � their seven corporate sponsors. Though they emphasise an inter-disciplinary approach, it turns out that the disciplines involved in this program are all technical or business and management oriented. They have not included disciplines appropriate to investigate social and political issues. Stanford�s Clean Slate Design for a New Internet has no soul or social conscience.

Stanford University states their research program can be characterized by two research questions: "With what we know today, if we were to start again with a clean slate, how would we design a global communications infrastructure?" and "How should the Internet look in 15 years?"

A new internet architecture such as proposed will open vast new markets and endless business opportunities � in short � a potential gold mine for the seven industrial sponsors. The fear is that the Stanford research program will trade off attention to social and political issues for expediency in the impetus to get the new infrastructure up and running sooner.

How do we ensure that those questions don't get switched around to, �...if we were to start again with a clean slate, how would we design a better conduit to more efficiently funnel revenues to our sponsors?� and �How should their profit margins look in 15 years?"

I think there should be a single focal point dedicated to social and political research, public discussion of these concerns, and solicitation of public input. There should be some sort of Ombudsman appointed to ensure that these issues will continue to receive proper attention right on through to implementation of the new Internet infrastructure. I feel the most appropriate seat for a �social conscience� would be at the heart of where the bulk of technical research is taking place, where social scientists can interact with engineers on a daily bases. In my mind, that would be a truly inter-disciplinary approach.

A new internet architecture can open up new possibilities for democracy by for example, allowing internet voting and instant referendums on current issues. On the other hand, it will open vast new opportunities for taxation and government control. Censorship and collection of information on citizens will become easier than ever imagined.

We need to shape political opinion before the politicians get involved, so that when they finally do, they will be guided by an already established body of enlightened thought rather than reacting to their lobbyists or whims of the day. Political scientists need to anticipate the political issues well ahead of time and illuminate our collective consciousness so that we will be better prepared to guide our political representatives or be able to react quickly when legislation is proposed that goes against what should be pre-established principles.

I feel these issues require broad discussion. The time to begin discussion is now. It is my hope that I have just initiated such discussion with this blog.

George W. Taylor
(aka TropicalCoder)
March 15, 2007

Your comments are welcome.