A Practical Navigator for the Internet Economy

GLOBAL SURVEY: PACKET NETWORKS & FIBER INFRASTRUCTURE RACING TO MEET DEMAND

Farooq Hussain Discusses Policy Issues and Business Models Impacting Convergence and Relationship of Regulation to Innovation

pp. 1 - 9

We interview CIX Board Member Farooq Hussain about his recent presentation on Internet growth issues to the Asia Pacific economic Council. Farooq points out that circuit switched telcos find themselves hard pressed to adjust to a world where data has already or is about to become the majority of their business. The exponential growth of the data market is expanding to ordinary ISP's like AboveNet the ability to by IRUs on dark fiber, something that would have been unthinkable two years ago. The ICAIS group within the Asia Pacific Economic Council is making a detailed analysis of data charges by route. They find city pairs internationally and figure out what capacity is available, what the cost was 18 months ago and what it is today.

AboveNet, Verio, PSI, Frontier Global, Exodus and others have all bought international fiber capacity as IRU's or leased it higher up the food chain. Each new trans Atlantic cable more than doubles, or in some cases triples, the previously existing total capacity across the Atlantic. Over the lifetime of a cable, paying $8 million now for an OC3 for 25 years is a very good proposition for the cable infrastructure provider because they're saying: We're going to have 15 to 20 times more OC3s available. And we can put those additional OC3s on the same physical strand of fiber!

At least 5 new entrants (Level3, Qwest, Williams, Frontier, Enron) are very serious national level infrastructure players in the United States. On the international scene, there are very significant players, Gemini for example, over and above Global Crossing who can sell you trans Atlantic bandwidth now. Huge increases in capacity are being met by huge increases in demand. The unknown is the extent to which they will balance.

The Internet industry does not need some ingenious reinvention of the current circuit switched settlement model to apply to packet networks. The industry needs a model which will enable it to continue to grow and support the stable development of large carriers, medium providers and small start-ups. The telco regulatory map doesn't provide for this. Until deregulation, the telco regulatory map provided for the conservation of large players only.

The industry does need to move everything over from circuit switched to data networks. The faster and more efficiently it does this, the more successful it is going to be in today's rapidly changing environment. In the midst of such rapid technology change if the government or anyone else seeks to impose a regulatory settlement environment between circuit switched and data networks, it will kill everything off. All of the innovations and all of the new, emerging companies that are driving the information infrastructure economy are coming in from the data world. Seeking a regulatory environment that preserves the status quo is the last thing the industry needs to do, especially given that the circuit-switched world has already made business decisions that they should be buying companies in the data network world.

There are in fact a very large number of ISPs and CLECs in the United States. Relative to 3 years ago, this is an astonishing picture. There are over 6,500 ISPs in the United States alone. The number worldwide is larger, of course. A significant part of everybody's [AT&T; MCI WorldCom; Sprint] long distance business is marketed through resale channels. The portals are leveraging their strategic relationships with service providers and content providers. So there's a tremendous amount of mixing going on here. There's cable and broadcast media, traditional long distance carriers, new emerging companies, ISPs and CLECs. Each of these segments has accompanying industries of very varied sizes.

In addition, more and more communications of all types are moving into packet-switched networks. As this happens, new companies, architectures and business models are being made possible. These new business models can't be mapped into the older infrastructure because they don't acknowledge the old boundaries found in circuit-switched networks.

The industry looks like it's headed for both flat-rate and service specific utilization-based pricing models. Take, for example, Net Zero, a company in California. There are a number of these types of companies springing up now. They offer free Internet access to subscribers in return for their filling out a profile form which is really not that different from the form people fill out to get free trade magazines. The form asks what you do, who you work for, etc. Dixons in the UK is taking advantage of the fact that you have to pay metered usage for local access. The customers consume local access minutes for which they must pay BT. BT turns around and gives a portion of the local access call fees to the ISP which is thereby funded to provide the free service. On the other hand, Net Zero is selling advertising.

Two years ago with all the flak about UUNET cutting off its peers, Hussain was definitely worried. But the industry survived that moved with even more players and a more diverse infrastructure. The fundamental issue remains the same. Can a company get itself started, deliver profitable service and grow in the environment we have? The answer is yes today but it probably was very much more ambiguous a couple of years ago. Why? Because the answer then was still yes, but you had to buy connection to an upstream Tier 1 provider first.. Finally with the changing data models and the liberalization of the telecom regimes, a lot of the traffic will eventually be carried on the European and Asia-Pacific exchanges. The European issue has already started to move away from MAE East. The links have grown in traffic volume enormously in the last year. The European players have changed their rules of membership to be more open. There are additional new exchanges points in Europe which are acting very, very constructively. A lot of the regional traffic is now staying within the region.

NANOG DISCUSSION EXPLORES QOS PARTICIPANTS SEEK TO DEFINE WHAT IS HYPE AND WHAT IS REAL

Ask Whether Bandwidth May Be Used As QoS Tool

pp. 10 - 15

A Quality of Service discussion on NANOG in mid May revealed that for all the discussion and hype surrounding QoS, while there is no unanimity regarding its use either structurally within the internet or in terms of business models, some tools are becoming available and being used. Interest in the subject in both its technical and economic aspects remains high.

AT WAR FOR THE FUTURE OF THE INTERNET

ICANN IGNORES ITS BY-LAWS, SWEEPS OPPONENTS ASIDE, CLAIMS CONTROL OF ALL DNS AND PREPARES FOR SHOWDOWN WITH NSI UNDER GUISE OF ENDING NSI MONOPOLY

Actions Create Regulatory Body Designed To Make Internet Safe For Content Control, Legacy Technology and Fortune 500 E-Commerce,

pp. 16 - 25, 28, 30

The Internet has leapt from success to success daring the technology, government and commercial empires it threatens to do something about it. It has directly challenged the most powerful political and economic interests in the world. They have come up with ICANN to protect their own viability under the ruse of "protecting the stability of the Internet". A charter to perform the old IANA technical coordination functions and trumped up need to end the NSI monopoly has served to give Vint Cerf, Mike Roberts and Esther Dyson carte blanche to use the worst tricks in the corporate law books. What they have done is to create an unaccountable body that, barred by its own by laws from making policy, is now creating a mechanism that ostensibly is a defense against government regulation but in reality places the corporations whose interests are being threatened in control of their own regulatory regime. This regime lacks any of the procedural safe guards that we have come to expect from democratic and open governments.

People who both know Esther Dyson well and understand how she is selling out the interests of small businesses and entrepreneurs via her imposition of ICANN's regulatory, IBM big-business-friendly regime over the internet have wondered why she is doing this. the answer came on June 10 when Esther began to collect her reward from IBM in return for the flak that she has put up with. Not content to be the guru of industry insiders, she has embarked on a broader path of publicity, on a program of instant recognition for her personna courtesy of IBM. She's the star of an IBM e-commerce commercial. Esther as chess playing athlete brought to you by IBM e-commerce solutions. So there's the grand tie in as Esther draws the corporate nooses around our necks and she is beamed into 50 million homes as part of the NBA playoffs. E-commerce? Think IBM. Think Dyson. One of the strongest pushes behind ICANN is the perceived need on the part of the Global corporations for the "creation of a global legal framework in which laws work across national boundaries to reinforce the upward spiral of value that Internet is capable of creating." in the words of Vint Cerf. However, interoperability is desirable only in the context of a balance between public and private interests which is utterly lacking with ICANN.

Our article takes material largely but not entirely from a private mail list where some participants in these events are developing an understanding of where these issues came from and the ways in which ICANN and ISOC are working to impose their "rule" on the internet.

Dave Hughes has summed things up well: "The danger of ICANN, of course, is an 'internationally' governed Internet - meaning everything from IP names to 'permissible' content, who can be an ISP, who can put up Web Sites (be licensed, with fee), who can even POINT one web site to another (one foreign court just ruled it is illegal for a web site to even post a hyperlink pointing towards a web site owned by Scientology, a step far beyond citing them for publishing their content). i.e. we seem headed for an Internet which is government controlled, without freedom of electronic speech, and run to the satisfaction of the largest corporations for their benefit, not yours. Under such an internet we will likely see the imposition of "standards" in a way to deliberately inhibit the introduction of new technologies in order to allow those with the vested interests to amortize their installed plant investments over the longest possible period at the expense of the newer technology companies and the general public. ICANN aims at nothing less than becoming a Internet United Nations, a Treaty-bound organization, where the US government would be bound by 'its' rules, arrived at by votes of other countries, with taxing authority (ICANN wants every DNS domain name taxed $1 a year to pay its expenses)."

We are at war for the future of the Internet. Ironically for those who hate NSI, it may be only they who have enough muscle to derail ICANN.

DAVE FARBER'S VIEWS ON ICANN'S ORIGINS AND SOLUTION,

pp. 26-27

In a June 6 interview Dave points out that the legal threats that Jon Postel faced in 1998 allowed his pro bono attorney, in order to protect Jon, to create an ICANN that was unaccountable and could not be sued thereby laying foundation for the problems that have outlived Jon. Farber also offers useful opinions on what he sees as the need for uniform international standards on matters like e-commerce.