A triple issue on Commoditization, Telecom Economics, and Voice as a Network Application and the economic consequences of the disappearance of premium tariffs for voice

The Paradox of Commoditization

Studies in VoIP and Telecom Economics

Find out how to order single copy ($400) or group license ($800) for just the April - June 2003 issue.

Contents:

Voice Becoming Just Another Application

Increasingly Rapid Deployment of VoIP Will Break Most Vertically Controlled Telco Business Models

Japan & Korea Broadband Build Outs Successful

While FCC Action Leaves Municipal Networks as Only Reasonable US Alternative pp. 1- 4

Part One An Introduction -- Technical Issues

SIP, H.323, and Megaco/H.248 How the Basic Protocols Fit Into a Changing Picture p. 5

Current VoIP Architectures, Protocols and Business Models -- Is Delivery of Real Time Voice to Be Provisioned as Service or as a Customer Owned and Controlled Application? pp. 6 -17

VoIP to VoIP Call Routing Outside the TDM Network (PSTN), p. 7

Bellhead VoIP versus Internet VoIP, p. 7

Quality of Service a non Issue? p. 9

H.323 Scalability, p. 10

Is the NGN Concept Viable? p. 10

Not SIP Versus H323 But Rather is it Application Neutral? p. 12

Is ITXC Going SIPing? p. 12

VoIP as an Enhanced or Information Service p. 14

Current Functionality of SIP --SIP Proxy Server Design and Business Functionality Issues Voice as Just Another Application May Have no More Incremental Value to Network Operations than E-Mail p. 18

The Session Initiation Protocol (SIP): -- A Key Component for Internet Telephony by Jonathan D. Rosenberg & Richard Shockey (June 2000) p. 22

Verizon Plans Walled Garden SIP --Critics Claim it Refuses to Give up Control p. 25

Just Do It! NANOG Discussion Finds QoS Generally Not Needed for VoIP p. 29

Part Two VoIP Arbitrage of International Rates

Tom Evslin Describes Operation of ITXC in the Context of India, China and West Senegal p. 33

CLEC Economics in Alaska -- Access Charges Established at the Time of Divestiture no Longer Reflect the Costs of Service -- Local Service Subsidies Will Be Bypassed p. 40

Black Market VoIP in the Third World p. 44

Legal Bypass in Europe -- Rate Differential Between Cellular and Wireline Leads to many Unintended Consequences p. 46

Part Three Broadband and VoIP in Asia

Nobuo Ikeda Describes the Opening of Access to NTT Fiber and Copper and the Rise of Softbank, p. 50

Izumi Aizu Describes Relationship of Players in DSL Market in Japan, p. 56

NTT Fiber to the Home Problems and Prospects, p. 60

Broadband in Korea Musings on Izumi Aizu's Study and an Update in Conversation with Kilnam Chon p. 62

Part Four Regulatory Issues

FCC Feb 20 Decision Grants LECs first step in building unregulated "fiber" monopoly Ruling Internet as Info Service Gives LECs Opportunity to Keep VoIP off their Loops Raising Opportunity for Cable Co.s p. 67

Computer II and III Next on FCC Hit List? Apparent Strategy Designed to Cut off Independent ISPs from Anything Except Dial Up Access to ILEC Networks p. 73

Open Spectrum - Thinking about Spectrum? Behavior of the New Radios Doesn't Map to Traditional Views of Spectrum, p. 78

June issue

Some Ideas on How to Move Ahead with Edge Based IPv6 p. 84

Part Five Economic Policy Analysis

A Roadmap for Decision Making --Evaluation of Present Choices Depends in Part on Evaluation of Options Lost

We Examine How Telecom Economic, Investment and Regulatory Decision Making Fit into a Possible Array of Future Choices, p. 87

A Roadmap for Decision Making -- Roundtable Discussion Thoughts on the Economic Foundations of Telecom p. 89

A Roadmap for Decision Making -- The Paradox of Commoditization Trying to Save What is Inevitably Lost, We Lose What Could Otherwise Be Gained p. 102

Interview, Discussion, and Article Highlights p. 104

Executive Summary p. 123

Introduction

Over the next year the telcos will have to face the fallout of voice increasingly becoming nothing more than a data stream. An application with no more impact than email on bottom lines. This Issue of the COOK Report explores the many causes.

Some Data Points

* Flat rate pricing for phone service. Begun by MCI last year with $50 a month local and long distance. In some US cities. Experimented with by ATT with 20 dollar a month unlimited to other ATT customers. Pushed further by Vonage with $40 unlimited North American Long distance. On March 5, Scott Moritz writes that Verizon and SBC are preparing to launch their own flat rate plans. http://www.thestreet.com/tech/scottmoritz/10072101.html

* IBM moving its workforce to VoIP. It's ripping out token ring and installing Ethernet organization wide in order to do it. http://www.nwfusion.com/news/2003/0303ibmvoip.html?net

* Foreign governments that try to maintain official PPT tariffs find traffic disappearing off their official TDM networks drawn away by VoIP.

* An Alaskan CLEC that is reluctant for political reasons to use VoIP to bypass termination charges of local carriers investigates using VoIP over cable to accomplish the same end.

* A technology solution to cellular wireline rate arbitrage in Europe is found to enable free calling from the US on tie lines to European cell phones. The same technology solution is then found to enable the by pass of disparate international tariffs. In Asia this technology can insert calls from the wireline into the cellular networks and in so doing avoid the local tariff differentials. All one need do to get an understanding of these arbitrage possibilities is to look at the differences in Vonage wireline and cellular international calling rates.

* A month after starting to use Vonage calls to Canada go from five cents a minute to free. New Telephony announces that Vonage and EarthLink will announce a joint marketing agreement.

* The recent FCC decisions by threatening use of UNE may educate users into turning to the Cable Cos for telephony services rather than risk getting caught between warring phone companies. Developing a suite of content, internet access and telephony services is likely the only way that cable cos can get enough income to keep their stock price from collapsing.

* In Japan there has been spectacularly successful unbundling of access to NTT Fiber and copper ignites Japanese broadband DSL market where VoIP is bundled free with Internet access by Yahoo!BB. The only ones harmed appear to be NTT. See http://www.forbes.com/forbes/2003/0317/102.html

* SIP makes voice an application and puts it into end user hands

What Makes VoIP just an 'Application?'

There are three components to an end user controlled and accessible VoIP application. (1) Broadband IP - the transport layer, (2) a VoIP gateway (that is a hardware device), and (3) SIP or H323.

When an end user wants to run VoIP, the end user needs a SIP or H323 proxy service (application) that may come now packaged in various forms. The user must also have broadband IP and a VoIP gateway hardware device or system. The device might be a ATA-186 in the case of residential users or in the case of enterprise users, a blade in a PBX or a box attached to the back of a PBX such as a Cisco 2600.

But with enterprise services or with ones like Vonage, it is SIP (or H.323) acting as an application that establishes the end user as customer and channels that customer's calls to the PSTN or to other VoIP destinations. SIP (or H.323) is the "starter" that makes the VoIP engine run. While H.323 is currently the dominant application by which VoIP is delivered, readers of this issue will see the many reasons why SIP is expected before long to replace H.323.

VoIP, as just described, is the means by which centrally managed and controlled voice service as provisioned in the PSTN is being transformed into an end user controlled application. VoIP can be configured as either a hosted application similar to WEB hosting (like Vonage or Packet8) or directly managed by the enterprise or individual depending on their technical competence or appropriate cost-benefit analysis.

The Telco's Vertical Business Model Broken

This is the fundamental revolution in Voice Communications now underway. Carriers are losing control of the ability to provide voice services since these are now nothing more than applications that you can buy off the shelf or purchase on a hosted basis from anyone that has sufficient IP transport capability. The cost differential of deploying SIP versus classic circuit switched technology is worth pointing out. An Ethernet port needed to run VoIP is maybe $10.00 while a Class 5 Switch port costs about $175.00 not counting maintenance contracts.

Voice is becoming another asset owned and controlled by the user. Most voice has not yet succumbed to end user control. There is now virtually nothing standing in its way. The damn is breached and pace of by pass will only increase.

The telco in short is no longer in control of all its services. When it tries to control what is customers can do, there are now many work-arounds. A work around in one area often produces cascading work-arounds in related but not immediately expected areas. However the telcos will worsen their situation by remaining in denial.

The old wire line ITU blessed voice pricing tariff regime has been largely destroyed by VoIP and opportunities for arbitrage

The old system of different prices and built in subsidies no longer sustainable. The world will increasingly turn to flat rate charges. If the FCC ruling of February 20 survive the only course for innovation in the United States will be seen increasingly to be asset based municipal builds where cost of financing over 20 year periods will be sustainable in ways that the debt leveraged telco assets are not.

Perhaps the future of telecom will be one where municipalities and enterprises will choose between doing business with increasingly weak vertically integrated old-fashioned giants or building their own modern systems that they operate and control. They will build with the assistance of technical specialists that will offer the edge based asset owners specialized consulting and project management services. The edge located asset based enterprises with go to regional exchanges where they can interconnect with the new version of the carrier. These will be inter asset based transit providers connecting the enterprises and the municipal networks. Level3 and Global Crossing come to mind.

Then, if the Canarie view of the world turns out to be correct, the main operational question will be the means of inter municipal and inter enterprise transport. Namely how many light waves will there be and what will their granularity and patterns of control look like?

Meanwhile the FCC heads in a direction that most agree will land everything in court, will not encourage competition, and will raise prices. As the discussions presented in part four of this issue show, it may well be rife with unintended consequences. The FCC decision is clearing the way for the LECs to restrict access to their networks in such a way that the cable co's may become the only wire to the home where innovative services like Vonage's VoIP can survive. The cable cos already lead the LECs in delivery of broadband in the United States. The FCC decisions could help the Cable cos to increase that lead. In threatening independent ISPs the FCC in trying to assist the LEcs has likely made it clear that municipal networks and wireless are the only remaining alternatives for modern telecommunications. The LECs may come to find themselves quite isolated.

Meanwhile we can hope that the 802.11 family of standards has too much invested in it for any politically motivated move to harm it. Dan Gilmore's column about Canopy is a hopeful example of where the misguided policy of the FCC is pushing things. See: http://www.siliconvalley.com/mld/siliconvalley/5352348.htm Wireless networks stitched together by intercity carrier fiber (Level 3 for example) could eventually replace the LECs local loops. The question is how much economic damage do we do ourselves before this happens?