The End-to-End Principles and Internet Policy

In "Rethinking the Design of the Internet" Blumenthal and Clark Pose Dilemma of Central Control vs Trust at Edge

We Ask Can an Analog of Error Checking Role of TCP Be Implemented in the Application Layers and Trust Moved to the Network's Edges?

pp. 1 - 5

Find out how to order single copy ($125) or group license ($250) for just THIS issue.

Marjory Blumenthal and David Clark have written an excellent paper Rethinking the design of the Internet: The end to end arguments vs. the brave new world, The paper, available as a pdf, is to be found at Rethinking

We have come to appreciate that there is really only one fundamentally basic and important thesis on which the Internet is founded. It is the end-to-end principal as expressed in the 1984 paper: J. H. Saltzer, D. P. Reed, and D. D. Clark, End-to-End Arguments in System Design, ACM Transactions on Computer Systems 2(4), November, 1984. The following URL will retrieve a copy of that paper. End-to-End

The TCP solution to the problem of a packet network that was prone to congestion, to error and indeed to 'packet loss' had profoundly revolutionary implications that were slow to be grasped. Even more than twenty years later their significance is not adequately understood by new generations of non technical Internet users. The location of a means of verification of correct transmission reception on each end user's machine rather than in the middle of the network made the transition from a single network to an INTERnetwork or network of networks possible.

Placing the TCP stacks on each user's machine ultimately meant placing trust and control in the hands of each user. Placing the stacks at the edge enabled end users to place applications directly on their machines ensuring the open and decentralized nature of the network. It also enabled end users to create and quickly replicate new applications like the World Wide Web because no central bureaucracy was needed to give approval.

In conversations with David Clark, David Reed (Reed is with Clark and author of the original 1984 End-to-End paper) and Einar Stefferud we focus on the issues of trust within the Internet asking whether there may be a means to do at higher levels of the protocol stack what TCP does.

Routing Infrastructure: Issues and Problems

Craig Partridge Finds that End of Era of Rapid Growth Provides Opportunity to Implement Fixes

pp 6 - 11

In an interview about the problems of Internet routing Craig Partridge, Chief Scientist at BBN, flays MPLS. He adds "BGP does need be replaced. The problem is you can't replace it in a vacuum. If you're going to head down this path, you will find BGP is just one part of a whole hierarchy of protocols that you have to begin to rethink. The problem right now is that we're creaking along with an aging infrastructure that has limitations in the ways in which you can dictate how your traffic is sent through the network."

"Never underestimate an existing technology. BGP, OSPF, and IS-IS can still run the Internet for another seven to 10 years. As time goes by, we will be increasingly dealing with their limitations and their challenges. We already are. The remaining mileage that we can get out of these areas keeps us from running off and doing something new. We will only do something new if that solves a whole class of problems that BGP, OSPF, and IS-IS do not.

Caspian Networks Bandwidth Study Questioned in Comments Prepared for COOK Report by Andrew Odlyzko

Larry Roberts Responds

pp. 12 - 17

With Dr. Lawrence Roberts (who as ARPA's Chief Scientist had in 1966 begun funding the work that gave birth to the Internet) as Caspian CTO and Chairman, Caspian, since its founding, has attracted much attention from a series of press releases that promised to deliver a very revolutionary product.

However, with the boom over and with the growth of the belief that there was a bandwidth surplus on the backbone, the founding rationale for Caspian Networks could be questioned. On August 13, 2001 Caspian released a study of bandwidth that claimed growth in demand was far stronger than believed. Unfortunately, since it was based on non disclosure data there was no way to obtain independent verification. We asked Andrew Odlyzko who with Kerry Coffman had done the most to debunk the earlier myths to look at Caspian's data and comment."

Odlyzko responded: "Although the Roberts study also comes in with growth rates far lower than the 8x or 16x per year that had been widely assumed, it makes a huge difference whether the true growth rate is 2x (as Kerry Coffman and I estimate) or 4x per year (as Larry Roberts claims). If 4x is the right growth path, then the Internet industry will soon see a huge increase in revenues, and this will feed back into huge increases of sales in equipment. However, if the growth rate is 2x per year, then the picture is much less cheerful, as technological progress, combined with competition and the overbuilding during the past couple of years, would keep revenue growth to modest levels for quite a while. In particular, which of our estimates is correct is likely to become very evident within a year or two, from watching what happens to the carriers' revenues."

Dave Hughes and the "Digital Divide"

25 Years of Building a Philosophy Assessing the Value of the Personal Computer and Internet as if People and Locally Controlled Grass Roots Economies Really Mattered

pp. 18 - 26

We began a long interview with Dave Hughes by asking: "There has been an unspoken assumption that computers are somehow better and that without too much thought, everyone can just apply them to speeding up the flow of everyday work. It seems to me there has been too little reflection about what should it mean to the economic political and social life of the community when you put the Internet and personal computers into the hands of the people. What have you learned in the nearly 25 years that you have been helping people make innovative adoptions of computer based communications? Is there any kind of methodology that you can apply to looking at these issues and to trying to decide how to chart a future course?"

Hughes: You have to start with the fact that nobody knows how to make a buck on the net. There is an article of faith that says if everyone were on the Internet, and went through the right steps somehow everyone could improve his or her economic situation . We really don't know to what extent this is true and it is certainly not clear that there is any one magic solution. When we talk about knowledge as the strategic resource of the so-called information age and the Internet as a means to move knowledge we are still groping for an understanding of how to apply these ideas in a marketplace economic model. And therefore there needs to be a little bit of skepticism applied to the assumption that if a community just goes through the "right steps" in hooking up everybody the effort will transform life there.

What people need to think about the basic kinds of economic transactions that can make a profit. Remember that electronic delivery of the knowledge product over the Internet can lower the cost of doing business more by that means than by other means for delivering a knowledge based product. And the cost will not always be just in dollars. But time spent - such as travel and meetings, Traditional activities that have a cost can be profitable via the Internet where rising costs affect more traditional ways of doing things.

The REAL ICANN Versus the Public ICANN an Essay by Dave Hughes and Gordon Cook

pp. 27 - 29


Postscript: Taking Advantage of the September 11 Attacks ICANN Moves Its Control Agenda to Front Burner

pp. 29 - 34

Mike Roberts September 14: Today, that idea [an at large membership] is out the window, along with a lot of other populist notions about any old terrorist around the globe getting to vote on how to run the DNS. When civilization takes a step backward, as it did last week, it usually means a period in which the people with the guns make the decisions.

On September 26, ICANN President Stuart Lynn announced that the planned agenda for the November annual meeting in Llos Angeles was cancelled and would be replaced with one on ICANN's role in protecting the security of DNS.

Roberts on Oct 26: But the legal date between the "old" Internet and the "new Internet was today, October 26, 2001,when President George Bush signed the anti-terrorism bill that was passed by the upper house of Congress yesterday with one dissenting vote.

This legislation brings the Internet and its developers, providers and users directly into the new war on terrorism. It extends extensive new power to law enforcement to find, capture, and punish those who use the network for terrorism or other criminal activity. It removes the previous barriers between foreign and domestic anti-terrorism investigations and establishes the principle that whoever you are, wherever you are, if you use the net for terrorism, you are in the sights of the FBI, the CIA, the NSA and their foreign counterparts.

Interview/Article/Discussion Highlights

pp. 35 - 52

Aproximately one third of the Clark Blumethal paper is included in this section.

Executive Summary

pp. 53 -54