A Practical Navigator for the Internet Economy

Future Is No Longer the "Public Internet"

It Is Role of IP Protocol in Global Telecom

COOK Report to Continue Focus on IP Technology, Economics and Policy in Global Communications

Introduction How to purchase this issue. $200 single copy or 600 group.

We stand on the cusp of major changes. In acknowledgement of those changes we are changing the name of this publication from The COOK Report on Internet to The COOK Report on the Internet Protocol: Technology, Policy, and Economics. This change in name renders explicit what has been implicit in our coverage for the past two years. The COOK Report matured by chronicling the rise of the commercial public Internet - in other words email and the Web connecting publicly reachable IP numbers to other publicly reachable IP numbers in perhaps 200 nations around the globe. The "public Internet" that we chronicled was and is only a single application of the Internet Protocol to communications. The public Internet was but one consequence of the Internet Protocol (IP) at its most fundamental level.

What is in the Name Change?

IP has become the predominant protocol used in packet switching. It is the predominance of IP added to the greater efficiencies of packet switching that unleashed the revolution that affects every form of computer and network communications and has rendered at least a trillion dollars of circuit switched plant and CPE obsolete. When the vast majority of packet switched applications began using the Internet Protocol, there emerged a commonality of approach to digital communications that quickly acquired economies of scale around the globe. This combination of IP and packet switching made digital communication, as opposed to analog communication, economically viable across all computer platforms and operating systems. It brought the full benefits of the integrated circuit, and of fiber optics to communications. By more efficient use of communications channels and commoditization of devices based on chips that could be turned out by the billions, packet switching made possible a huge communications pricing revolution.


In only a decade the Internet Protocol rendered the world's circuit switched analog telecommunications infrastructure obsolete and sent it to the verge of bankruptcy. The new Internet Protocol technology destroyed the viability of at least a trillion dollars worth of physical plant on a global level. Perhaps several trillion dollars worth. Spurring vast enthusiasm and investment, it grew a bubble that burst in the first months of 2001. We are now nearly three years into dealing with those consequences. In trying to make sense of what had happened and see where the future might lead, we have broadened the coverage of the COOK Report considerably. Why? Because short of the power grid and the physical highways on which we travel, the economic viability of the globe's most important infrastructural system was called into question by the IP revolution.
While the technology is not stopping its advancement, in every developed nation around the globe, the paramount issues quickly became the viability of the incumbent phone company. The technology destroyed the economic proposition on which an infrastructural system critical to the health of every national economy rested. The success of the new technology destroyed the economics of the old technology. The sirens heard were those of the national policy systems coming to the rescue. Policy is often slow to change but for the second time in a decade in the US we are now beginning to see some extremely important policy changes.

In this issue that closes out 2003 we will try to put into layman's language what the FCC is doing. We will also try to show what the landscape will look like at the end of the day. Major changes are about to happen. Some business models like those of independent ISPs are about to be destroyed. We are not happy to see such destruction. However the impact of packet switched technology will continue. The efficiency and power of the technology continues to make new business propositions possible. We shall continue to identify them. Nevertheless, we are making this name change to alert our readers to the fact that we shall continue to chronicle the opportunities created by the Internet Protocol and packet switching during a period of time where the "public Internet" may well no longer be at the forefront of change. The economic shoe dropped two years ago. We are now going to be treated to the policy shoe. Things will no longer be the same.

Restructuring will continue. Communications will become more dependent on IP not less as the major carriers are now announcing that they are converting their entire networks to run the Internet protocol. They have no choice but to do this because, as they make capital expenditures as a part of investing in upgrading their network infrastructure, by building that infrastructure on IP they get far more capability for each dollar invested. What is still unknown is how long it will take to complete the transformation of the global system in the US, Canada, Europe Asia and the world as a whole. We shall chronicle these changes on a global level.

The first part of this issue will discuss the Triennial Review, the first of several policy shoes (also known as "rule makings") that FCC will be announcing or "dropping" over the coming months. These rule makings will put a lot of folk out of business. They do not leave us happy. They are in fact ugly. But since the economic success of the Internet protocol has put the ILEC circuit switched infrastructure valued at perhaps 300 billion dollars in the US in jeopardy, we can hardly be shocked that the regulators stepped into the breach.

VoIP in Japan appears to be going full speed ahead as is a regulatory policy towards broadband that is not being fought by NTT the incumbent carrier. In the US unfortunately, the opposite is occurring. In the VoIP area over the past 30 days events are not encouraging. On October 1 a huge shoe dropped when California announced that VoIP providers have until October 22 to register with the state as a telephone companies. We will update the VoIP scene in early next year.
Meanwhile it looks like VeriSign may get away with its brazen wild carding of the .com and .net GTLDs on September 15. Although it backed down late on October 3rd, no one should expect it become any less arrogant. It describes what it has done as a "service." It is a service only to those who don't know any better. It also breaks many pieces of network infrastructure. VeriSign doesn't care. We see its behavior as one more piece of evidence that while many of the large carriers are beginning to go to all IP networks, what started it all - the original IPv4 public internet is public no longer as any kind of commons or public trust, but rather as a field of battle to be fought over by entities who are prepared to do whatever it takes for their own advantage regardless of the consequences for the rest of us.

In this issue we have three interviews on wireless sensor networks and pervasive computing. This will be an area of future coverage. We shall continue to focus on broadband on DSL versus cable and fiber to the home issues. We shall emphasize wireless at all levels. Voice over IP will be the subject of an issue we hope in time for spring VON. Peering, and changes of network architecture including viral and mesh networks is on our radar. The impact of regulatory decisions on this cannot be ignored. We will report on them and on their consequences for ISPs.

In the US the telecommunications marketplace may never again be as large. Nevertheless it is and will remain huge and globally it will likely continue to grow in size. The COOK Report on the Internet Protocol: Technology, Economics and Policy will continue to report on the business cases made possible by the on going shift in the three tectonic plates of IP.

Crunch Time in DC
Regulatory Socialism for the ILECs and Cablecos

As the financial destruction caused by the Internet Revolution continues, so-called "consolidation" is inexorably changing the shape of the public Internet on which we have reported for the past dozen years, we see more stark consequences for the future of the open edge controlled public Internet. Though we remain adamant in our belief that the virtues of the public Internet must be protected we must also admit that the public Internet as defined by openly reachable IP addresses is no longer where the future of telecommunications lies. The future lies with the Internet Protocol. Last spring ATT told Forbes: The 'Internet Protocol is like Pacman: It will eat everything in its way.' In acknowledgement of this fact we are modifying the name of the COOK Report from Internet to Internet Protocol, Technology, Economics and Policy. As we have for the past two years we shall continue to cover the global IP revolution in its broadest scope.

A significant aspect of that revolution is playing itself out in the United States in a series of major FCC rulings. These are modifying key portions of the Telecommunications Act for the second time in the past decade.

It seems that this country has adopted a strange definition of competition. Under the Powell led FCC, the Republicans are doing the Republican thing and letting the lobbyists of huge national phone companies set national telecom policy. The larger your cash flow, the more weight you can throw around.

The ILECs should have been allowed to follow WorldCom and Global Crossing into bankruptcy. Clear the decks. Junk their SONET gear and SS7 switches en masse. Put themselves in a position where they could deliver broadband now by rebuilding their infrastructure from the bottom up with the new technology that sooner or later they will need to acquire. Instead they are to be allowed to stumble forward - saved by their political muscle from having to bear the consequences of their slowness to understand what the Rise of the Stupid Network had in store for them. This FCC and this Republican Administration is showing its contempt for the risk taking small entrepreneurs who founded the ISP industry between 1992 and 1996. Building viable businesses on guaranteed access to ILEC lines, these ISPs are going to loose that access and the ability to compete with the phone companies which to the Washington politicians are too big to be allowed to fail. At present there are two viable broadband technologies for homes and small business. DSL and Cable. Each community has one source of cable and each, thanks to Powell's FCC, will soon have one source of DSL - the Local Phone company.

The policies of cableco and telco will be to keep similar prices and to go for elusive economies of scale by treating all customers as ignorant couch potatoes to be kept happy by high speed downloads of the web and scorned if they expect upstream bandwidth sufficient to email an issue of the COOK Report without a server time out due to packet loss. Customer support from a local ISP? Yes. From your local FCC sanctioned telco or cableco? You must be kidding.

Now into this duopoly, Powell is pushing wireless. He has many fans of what he is doing. Surely they feel it is a grand design. Wireless will knock the telcos and cablecos for a loop. Maybe. The problems is that to be really useful as we showed in our previous issue, wireless needs local economies of scale. The only way that it will get such is for enough ISPs to survive the telco FCC cableco onslaught to become wireless.

To be meaningful service must be provided by local business with roots in the community it serves and not by giant national corporations. Under those conditions the rest of us will have some choices. The scene in Washington is so pathetically conflict-ridden that we have little faith in any outcome beneficial to the creation of jobs or building of an infrastructure not controlled by distant, impersonal corporate giants.

By way of example consider Intel. Basically a good company, heavily invested in wireless, but one so big and with so many conflicting internal interests that it employs as Washington lobbyist Peter Pitsch - an ex ILEC VP who in Washington goes around embarrassing the wireless investment of his employer by advocating the position that if we just give the ILECs what they want they will save us after all.

The future of competition against the ILEC monopoly is further dimmed by Powell's success in getting the challenge to the Triennial that he helped instigate moved from the St Louis Circuit Court to the DC District Court of Appeals presided over by Judge Harry Edwards for whom Powell clerked less than ten years ago.

Triennial Review and the Problematic Future of the Independent ISP , pp 4 -9

The 568 page Triennial Review published by the FCC on August 22 dealt with the unbundled network. It sorts out the relationship between ILECs and CLECs. It does not deal directly with ISPs that have no carrier status. The Triennial impact on the ISP is indirect. If an ISP that is selling its customers DSL is using a CLEC service, and most are, then the CLEC is going to find that the UNE (unbundled network element) from ILEC that that enables it to provide DSL to the ISP, either disappears, or is no longer affordable.

The proceeding that does directly impact the ISP vis-a-vis the ILEC is the Wireline Broadband NPRM. This is the rule making is the one where FCC is considering throwing out required access to unbundled services that had been guaranteed by Computer II and III. A one two punch is coming. First the CLEC gets cut out of the ISP loop and then the killer punch for the ISP comes under the Wireline Broadband NPRM.

Computer Inquiries I, II and III, otherwise known as the Computer Inquiries, set up a regime where phone companies have to unbundle their services and provide them to ISPs on unbundled and non discriminatory basis complete with tariffed prices. Under the computer inquiries, if you are an ILEC and you provide service to your own ISP subsidiary, you have to charge the same price and be non-discriminatory in your provisioning and maintenance of circuits for independent and competing ISPs. The upcoming Broadband NPRM may well leave ILECs with no requirement to provide unbundled services to ISPs on regulated, nondiscriminatory terms using "reasonable" prices.

The combination of the Triennial Review and Wireline Broadband NPRM could mean that an independent ISP that relies on gaining wireline access to an ILEC's network to serve its customers is going to be at the mercy of whatever the ILEC wishes to charge for reselling its services. The ISP, wishing to compete with the ILEC's ISP subsidiary, is like a mosquito that has been feeding off of a restrained elephant and, now that the restraints have been taken off, expects to live. Unless it goes wireless and gets access to an IXC backbone, it will die. The elephant will no longer tolerate its bite.

The regulatory arguments involving all this are so opaque to outsiders that we shall probably wind up drawing diagrams to help out readers understand them.

We found Francois Menard's recent quote quite apt: The level of detail necessary to debate these issues is so immense that it takes more than dedication or logic; it takes passion that is beyond common reason.

The Communication Act requires certain things based on status. Title I entities have no formal obligations unless the FCC explicitly regulates them. Title VI Cable services are protected from various forms of regulation but are subject to different regulations.

Title II telecommunication providers cannot interfere with any transmissions over their lines, must interconnect with rival telecommunications carriers (what telecom lawyers call "non-discrimination"). Also, many Title II providers pay universal service fund (USF) fees. Title II telecom carriers are subject to CALEA and provisions of the USA Patriot Act that require telecom providers to build their networks so that law enforcement agents can monitor them for criminal or terrorist activity (with a proper warrant)."

COOK Report: Therefore, whether the Internet goes under Title I or II is really important.
While the Internet itself has always been under Title I the support structures necessary to provide Internet service have been under regulated Title II. To enable the telcos to capture the field, they are being removed from Title II and placed in Title I. But this is all being done piecemeal in a way that is almost impossible for an outsider to follow and with the result that ISPs will not be guillotined but die slowly, strangled one by one.

Congressional Shenanigans, pp.10 -11

Harold Feld: On September 17, 2003, the House of Representatives passed "The Internet Tax Non-Discrimination Act," or H.R. 49. The thrust of the bill is to remove the expiration date for the ITFA and make it permanent. However, in Section 3(c), it contains the following "technical correction."

The second sentence of Section 1104(5) and 1101(e)(3)(D) are each amended by inserting 'except to the extent such services are used to provide Internet access' before the period.
So the new definition of "internet access provider" will now say "such service does not include telecom service, except to the extent such services are used to provide Internet access."

What does this do? Well, at the least, it would appear that any telecom service used to provide Internet access is now free of state taxes. Since it is impossible to tell what part of a telephone line is used for DSL and what part is used for voice, this means that any household subscribing to DSL is exempt from state telephone taxes. The Multistate Tax Commission (www.mtc.gov) estimates that this "clarification" will cost the states between $4 billion and $8.5 billion annually by 2006. . . . .
Further implications of the change and how it effects the definition of broadband Internet access are unclear. Is it a Congressional endorsement of the FCC's cable modem ruling? Does it change the Communications Act definitions? Does it effect either the pending challenge to the FCC in the Ninth Circuit or the pending proceedings on defining DSL? This is, at present, unclear. Under one reading, nothing changes except that the states lose about $8 billion annually. On the other hand, supporters of defining broadband as an information service (the cable companies and the baby bells) will certainly argue to the FCC and the federal courts that this should be read as a Congressional endorsement of the FCC's current approach and a change in the regulatory status of broadband.

FCC to Independent ISPs Drop Dead, pp. 12-14

On August 22 the 568 page Triennial Review was finally issued. The rules mandated a transition from UNE-P line sharing to UNE stand alone loops. Grandfathering those who had them but imposing a 25% increase in price in year one, 50% in year two and 75% in year three.

ISPs without access to ILEC copper plant to deliver DSL would be effectively frozen out of everything except wireless broadband. The ILEC could charge its competitors whatever it wanted. Teletruth and the Small Business Administration had put a crimp in the FCC telco plans by a filing that said the FCC needed to consider the impact of its actions on small businesses that were affected by its regulations which the ISPs undoubtedly were. The FCC side-stepped that by claiming that since ISPs were not regulated, it could not be held responsible for how its actions affected ISPs. The impact of this would be to completely freeze ISPs out of offering broadband Internet service to new customers. ISPs would be left with dial up or wireless and if the ILECs did not have to tariff access to their copper local loops, they could in effect steal the business that as many as 10,000 risk taking entrepreneurs had built.

An independent ISPs' only hope for survival will be to go wireless. But without any need to tariff interconnections, the ILECs could price wireline backbone service to Internet Exchange Points at prohibitive levels. The WISP has to interconnect to Internet transit somewhere. The FCC may well have handed the ILEC as noose to string up the WISP.

Push the independent ISP domino down and find out how many of the remaining CLECs it topples. Likely quite a few. The ILECs seem to be on the verge of succeeding in a game plan to take out competition. With no competition up grades and concomitant capital spending will continue to decline. The industry will remain mired in depression and the nations communications system will become ever more backward. Business and jobs will migrate to countries that have been able to build adequate infrastructure.

Excerpts from The Digital Imprimatur by John Walker, pp. 16-28

Walker writes: Now let's consider what happens when this architecture is extended to the Internet. I believe the technological precursors described above will eventually be deployed in such as way as to turn the entire Internet into a Trusted Computing platform. What, precisely, will this mean? Well, just as a Trusted Computing system will load neither programs nor data files without a validated certificate whose signature matches their contents, neither will the Secure Internet transfer any document, in any standard protocol without such a certificate accompanying it. (And by the time this is rolled out, consumer Internet access will long have been restricted to a short list of protocols on standard ports: HTTP, FTP, SMTP, POP, etc. The peer-to-peer Internet, where any site could connect to any other on any port with any protocol will have passed into history as, indeed, is already beginning to happen due to NAT routers and firewalls which cannot be configured by the user to accept inbound connections.)

VeriSign and ICANN Fight, pp. 30 - 32

Late on October 3rd VeriSign announced its capitulation having the colossal nerve to claim "Without so much as a hearing, ICANN today formally asked us to shut down the Site Finder service," said VeriSign spokesman Tom Galvin. "We will accede to their request while we explore all of our options."

Wireless Sensor Nets and Pervasive Computing

This will be another area affected by the wireless revolution. It may well become an area of huge new growth for the industry - both IT and telecom. If it does, it will utterly change the way we live. We intend to cover it heavily in coming years.

Deborah Estrin on Wireless Sensor Networks, pp. 33-39

We were dealing with small wireless devices that we would use in an embedded context. This meant that they were not traveling around attached to someone's belt where they could get plugged into the wall once a day. It also meant that they were small and therefore that the physical size of the battery or solar panel also had to be small. Therefore energy was a primary resource constraint.

As you integrate the logic of the radio, the radio transmission, nonetheless, remains the dominant consumer of that energy resource. When you have a system where suddenly communications is very expensive because it is consuming a finite energy resource, then you have to design your systems differently. These constraints force you to process your data near the source. Instead of streaming all your time-series data back to a central location for processing you want to process the data locally to save transmission energy on the part of the radios. This is why you want to put computation with sensing on those nodes.

Energy and Scaling as Critical Constraints

Even without energy issues, you have scaling issues as the devices become smaller in size and begin to do more dense sensing. These issues also force you to move your processing closer to the data. This is in contrast to what would have been more traditional moves from 10 to 100 to 10,000 points of collection and measurement of data all being processed centrally.

Smart Dust, pp. 40 -43

We, as a company. are not a sensor company but rather a wireless sensor networking company. Our motes come with a microprocessor, memory, radio, a pair of AA batteries, a unique ID, and non-volatile memory for storing long term data. The software that runs on this thing is where the "magic" is. We give you 20 of those expansion boards and two gateways. The gateways have a serial port and a USB port on them. You plug the gateway into the PC and run the software that brings up the network of motes. With a graphical user interface, you get to configure the network. You get to pick the sample rates of the different channels including which channels you want to sample - depending on how many sensors you have hooked up. You get to pick the reporting interval. All this gets down loaded into the 20 motes that then automatically configure into a peer to peer multi-hop mesh network.

IBM and Pervasive Computing, pp. 44-49

Connelly: Our group is in what IBM calls the Pervasive Computing EBO or Emerging Business Opportunity. An EBO is an IBM group that is put in place to address new areas around which IBM needs to start to build a business. The EBO group is separated from the rest of the divisions in the company. The separation includes software and as well as services. It sits above those. In the Pervasive Computing EBO, we have a cross IBM mission. Our job is to try to understand the dynamics of the market. An important goal is to predict when the market will become mainstream. We want to be sure that we have been able to position IBM so that the company, as a whole, has the new products and services that are enabled by the new market.


Future Is No Longer the "Public Internet" p.1

Part One: Regulatory Issues

Imminent Demise of Independent ISP? We Examine the Impact of Triennial Review, Congressional Blindness, and Expected FCC Rulings ISPs Who Can't By Pass the ILEC Network Won't Survive p.4

What's in a Name? Plenty, Effects of the "Internet Tax Non-Discrimination Act" on Regulatory Classification of Internet Access Service, an essay by Harold Feld p. 10

FCC To Independent ISPs Drop Dead -- Agency Ignores Administrative Law Requirements as Small ISPs Are Unable to Fund Necessary Legal Action Visions of PC and Internet Enabled Entrepreneurs in Jeopardy p. 12

Will There Ever Be a Digital Imprimatur?
Can the Big players Ever Put the Internet Genie Back in The Bottle? p. 16

Pointers to Two Web Sites Valuable for Mapping Internet Connectivity p. 29
Verisign's SiteFinder Service Imposed without NoticeDestabilizes DNS ICANN Responds at First With a Toothless Meow -- Bankruptcy of Four Years of Self Governance Propaganda Revealed Public Internet Users Left with No Recourse p. 30

Part Two: Wireless Sensor Networks and Pervasive Computing

Wireless Sensor Networks
Getting Motes to Communicate as an Embedded System
Deborah Estrin Explains the Work of the Center for Embedded Networked Sensing p. 33

Smart Dust an Integrator of Small Form Factor Self Networking Motes to Which Users Add Their Own Sensors - Kris Pister, CEO of Dust Inc. Describes His Efforts to Commercialize the Technology p. 40

Pervasive Computing and IBM - Sensor Networks and Mobility Enhancements as Emerging Business Opportunities - an Interview with Letina Connelly p. 44

Interview, Discussion, and Article Highlights p. 50

Executive Summary p. 58