Submitted to: Request for Comments on Deployment of Broadband Networks and Advanced Telecommunications, Docket No. 011109273-1273-01, National Telecommunications and Information Administration
Hugh Carter Donahue, Ph.D.
Josephine Ferrigno-Stack, M.A.
University of Pennsylvania
Hugh Carter Donahue, Ph.D.
Josephine Ferrigno-Stack, M.A.
Annenberg School for Communication
University of Pennsylvania
3620 Walnut Street
Philadelphia, Pennsylvania 19104
This paper proposes a Quality of Service Monitoring (QoS) system for broadband networks and Internet Service and Access Providers as a policy vehicle at a time of regulatory devolution. While this proposal does not address any specific technical standards, the paper places the notion of quality of service metrics -- of use to the citizen and user, and of value to the corporation and regulator – in the historical context of regulatory devolution and information abundance. It suggests these historical circumstances both impel and enable users and citizens to take on many of the information gathering and discernment function historically exercised by regulators and journalists. Finally, it suggests that QoS monitoring will bring several benefits to broadband industries.
In response to this Request for Comment (RFC), Broadband QoS monitoring clarifies and unifies many of the issues raised by the NTIA. An initial discussion of Broadband QoS monitoring locates QoS monitoring in scholarly and public policy contexts. We then address several pertinent questions raised by the RFC.
Measurement and Standards Studies
Measurement studies properly begin in the United States with John Quincy Adams’s Report on Weights and Measures (1821) in which Adams, then Secretary of State, documents the evolution of standards from colonization, explains their incompatibilities, and finds that it is impossible to impose uniform standards or a metric system due to custom and habit.
Over the past one hundred years, the National Institute of Standards and Technology (previously, National Bureau of Standards) has made a mission of perfecting metrics for uniform measures and of incubating and nurturing wide ranges of standards setting initiatives. As a rule, these initiatives inflect industry practices to promote national commerce through accepted and adopted sets of standards, which enable competition to flourish.
In the genres of practical studies and guidelines for commercial application, Bell Labs, the Institute of Electrical and Electronic Engineers (IEEE) 802.11, and CableLabs have specified measurements for telecommunications and cable technologies, which become the standards for capital investment and infrastructure development.
In studies of the history of television, Fink chronicled the organization and workings of the National Television System Committee, Sheirs documents the technical development of television as well, Udelson traces competing technologies in the emergence of television, and Joel Brinkley narrates standards setting for terrestrial high definition television in the United States.
In the United States, an initial body of scholarship, created by Charles Francis Adams, called for sunshine commissions to administer the railroad industry. Adams’s “The Railroad System,” The North American Review, 1867 and “A Chapter on Erie,” The North American Review, 1869 initiate public affairs reporting of regulation in the United States and served as cornerstones for journalists and regulators.
By the turn of the twentieth century, muckraking journalists, notably Ida Tarbell’s History of the Standard Oil Company, 1904, exemplified a dynamic in which journalists called attention to corporate wrong doing and abuses of market power and Progressive Era reformers enacted legislation to create independent commissions charged with administering and overseeing markets. These palliative reforms were intended to insulate consumer and citizen from abuses of market power and to provide regulatory stability to foster capital investment.
Gabriel Kolko eviscerated the progressivist interpretation of betterment in his Triumph of Conservatism 1963 asserting that regulation and standards placed large, national corporations at competitive advantage over smaller, regional firms.
Marver Bernstein offered a less insistent tone in Regulating Business by Independent Commission, 1955 and Robert Britt Horowitz applies Bernstein’s thoroughness in The Irony of Regulatory Reform: the Deregulation of American Telecommunications, 1989 from a liberal-left scholarly point of view.
In The Corporate Reconstruction of American Capitalism, 1988, Martin J. Sklar points out that twentieth century regulation polices corporate abuse of market power after acts of egregious wrong doing and shies away from corporatist state/industry joint command of production and planning.
Ithiel deSola Pool’s Technologies of Freedom 1984 cautioned that the regulatory regime for one mode of expression should not be applied to others and argued passionately for first amendment freedoms of expression irrespective of modes of production and transmission in situations where there is no monopoly.
Lawrence Lessig, Francois Bar and others urged policy makers to carry the regulation of the narrowband telecommunications plant over to broadband communications.
Quality of Service and Service Level Agreement Studies
A common thread through the QoS literature explains Quality of Service monitoring as an implementation method for provisioning differentiated service offerings.
Kosiur 1998 notes that “[Service Level Agreements]...help keep conflicts between you and your service provider to a minimum by setting reasonable expectations of service.”
Huston 1998 similarly addresses issues of comity and measurement “the measurement of the SLA must be specifically generic so as not to create highly specific network tuning, but as the same time, the measurement must not be so generic that the technique and the reported outcomes become the subject to dispute.”
Odlyzko 1999 articulates the superiority of flat rate pricing to differentiated pricing, and views Quality of Service programs as “complicated...involv[ing] substantial costs in both development and operations.” There will be greater traffic with flat rate pricing and as such greater value in the network, he contends.
2 Public Policy Context
For American communications industries, a rate of return regulatory regime characterized capital investment and accumulation in fixed-wire telephony. Licensing regimes administered a broadcast radio and television industry supported by advertising revenues and a cable television industry supported by multiple income streams from subscriptions, advertising and programming. The Sherman and Clayton Acts authorized antitrust prosecution and oversight in cases of abuses of market power.
Capital Accumulation Mode & Regulation Regime
Telephone: common carrier
Rate of Return for monopoly
Commercial licensing and advertising
Cable: coaxial cable
Franchising Authority subscription plus advertising
Computer: binary core,etc
Antitrust/ Sherman and Clayton Act
Narrowband Telephone Systems
Common Carrier, mandatory carriage of all ISPs
High Speed Cable Systems
Commercially Negotiated Access; Cable Industry Discretion over ISP Carriage; F.T.C. Monitor
With the advent of high speed, distributed, broadband networks like the Internet, regulation is faced with new challenges, which do not submit to rate of return or licensing regulatory regimes.
“The industry, capital markets, and the government find themselves navigating between the matured, legacy communication system and the nascent innovation-driven Internet space of the future,” Federal Communications Commission Chairman Michael Powell commented in March, 2001.
To address these dynamics, Powell asserted that the Commission would promote broadband deployment, competition, innovation and investment, pursue universal service, harmonize regulations across industries, effect consumer protection, resist regulation of emerging markets, and resist the temptation to meld markets in an image of a particular industry.
Beyond these generalities, no firm regulatory course can be or has been charted because distributed, broadband networks are emerging contemporaneously with deregulatory devolution, and it is difficult to know with certainty how to proceed.
Partially, government wishes to stimulate investment and resulting tax revenues, so government is reluctant to act too soon lest it foreclose investment and taxes.
Partially, regulation no longer guarantees stability as it did in periods of copper wires and analog transmission towers.
And partially, the country is still feeling its way. For example, Congress vaguely defined broadband communications in the 1996 Telecommunications Act as capacities “to originate and receive high-quality voice, data, graphics, and video telecommunications using any technology.” The Act says nothing more precise. More recently, one bill defines `high speed data service' as “ any service that consists of or includes the offering of a capability to transmit, using a packet-switched or successor technology, information at a rate that is generally not less than 384 kilobits per second in at least one direction.”
The F.C.C. specified broadband as “the capability of supporting, in both the provider-to-consumer (downstream) and the consumer-to-provider (upstream) directions, a speed (in technical terms, “bandwidth”) in excess of 200 kilobits per second in the last mile.” This is little more than four times the speed of a 56K modem.
The closest regulation has come to a new regulatory regime are Federal Trade Commission and Federal Communications Commission consent decrees approving the AOL Time Warner merger.
In reaching its consent decrees with America Online and Time Warner, the Federal Trade Commission stipulated that the merged firm must reach commercially negotiated agreements comparable to one that Time Warner had negotiated with Earthlink with three independent Internet Service for carriage over its cable systems. A Monitor Trustee would oversee AOL Time Warner’s adherence to the implementation of these stipulations. An F.C.C. consent decree added conditions that allowed customers clear methods of choosing ISPs that do not favor AOL Time Warner affiliated ISPs, freedom from technical interference and choice of billing options.
Both consent decrees authorize a regulatory regime overseeing commercially negotiated access to broadband cable networks. Crucially for the regime change (from mandatory ISP carriage over narrowband phone lines to commercially negotiated Internet access over high-speed cable systems), the FTC and FCC expressly ruled out mandatory access obligations and place the government seal of approval on commercially negotiated access for multiple – not all – providers.
It is noteworthy as well in a period of regulatory devolution that the FCC adopted measures in October to overhaul accounting rules in its Uniform System of Accounts (USoA) and Automated Reporting Management Information System (ARMIS) ostensibly to reduce paperwork and simplify reporting requirements of incumbent local exchange carriers (ILECs). While certain efficiencies may result for the ILECs, the measures will make it more difficult for state public utility commissions to compare and contrast data, including data for service quality, in determining appropriate regulation.
Information abundance requires notice. Unlike earlier periods in our history when the capital costs of printing and broadcasting centralized news and information, information is more readily available. Its abundance enables people to educate themselves about the goods and services which they are purchasing. Specialized, niche Internet web sites, web newsletters, web news groups and chats offer opportunities for posting and receiving information in addition to information available through print and electronic publishers and broadcasters.
These conditions alter the public’s dependency on journalism to acquire timely and pertinent information about products and services and comprises a historic change, comparable in its own dynamics to regulatory devolution, in the means by which the public learns about products and services and problems associated with them.
Information abundance should not be construed, however, as some sort of electronic deus ex machina to suggest an end to the need for regulation and regulators of commercial markets – for information dispensed over the Internet is often manipulated by interested and in some cases malicious parties. Regulation and regulators, by contrast, must adhere to rules, procedures and metrics, which yield to verification.
We propose a system of hardware and software agents to measure the performance of the network from points at which outsiders have access to the network—namely, at the user's home or small business connections. The data collected by Broadband QoS monitoring will provide empirical data to clarify public policy questions, as raised in this RFC and to meet the Broadband Reporting Requirements as mandated by section 706 of the Telecommunications Act of 1996 and as codified by the Federal Communication Commission.
Furthermore, the data from this monitoring can be used by networks to differentiate service level offerings, by individuals to choose broadband service providers and monitor their own connections, by high-speed access resellers to monitor their performance to the end user, and by facilities-based high-speed access providers to distinguish their products from competitors.
Regulation of Broadband infrastructure is beset by a tangle of overlapping and confusing legislation and agency rules. The Federal Communication Commission (FCC) has traditionally monitored and regulated Internet infrastructure and connectivity issues, although Federal Trade Commission (FTC) is involved in monitoring and regulating antitrust and privacy concerns at the federal level and the NTIA also contributes to the policy debates. As evidenced by this RFC, there is still a great deal of debate as to the definition of broadband and the place of regulation. Broadband QoS monitoring provides a timely answer to several questions posed by the NTIA in this RFQ, most notably, those in sections B, C, D, J and K.
Section B, in summary, asks how should Broadband services be defined. Broadband QoS monitoring allows for more accurate measurement of the current broadband deployments, thus giving policymakers an empirical measure of what broadband services are currently available to consumers. Once an accurate examination of the variety of residential broadband services in the marketplace is established, a viable definition of what broadband services are and should be can be determined.
Section C seeks information to clarify consumer demand for broadband services and to discuss impediments to consumer demand. Broadband QoS monitoring and the distribution of Broadband QoS results to the public would work to level the information asymmetries between residential consumers and providers of broadband services. With the plethora of Broadband service providers and wide spread public confusion about these services, the publication of QoS metrics in easy to understand language would allow consumers to compare broadband services and prices, and make the best decision for their needs.
Section D asks whether the government should adopt universal access to broadband services as a policy goal. While Broadband QoS monitoring does address the normative issues of federal communication policy, the data gathered through widespread QoS monitoring can shed light as to what areas of broadband deployment are not responding to market initiative and clarify the policy steps required of the government. Such clarity would illuminate the digital divide with verifiable metrics.
Section J inquires as to how the broadband product should be defined. Again, as stated in response to section B, widespread Broadband QoS monitoring will allow policymakers to gauge the broadband services that are already in the market and the differences between these services. From this empirical evidence, broadband products can be more easily distinguished.
Section K asks for the essential elements of a unified broadband regulatory regime. Broadband QoS monitoring is essential to a unified regulatory regime because it monitors broadband services at the users end, regardless of broadband service technology.
The QoS monitoring, which we are proposing, also addresses a goal of Nancy J. Victory, Assistant Secretary for Communications and Information, to “promote competition through a technology-neutral paradigm” for it can be applied across modes of wireless, wireline and satellite modes of communication.
Please note that FCC measures modifying reporting requirements for its Uniform System of Accounts (USoA) and Automated Reporting Management Information System (ARMIS) make Broadband QoS monitoring a timely policy solution in order to provide regulators with service quality metrics.
In light of the uncertainty of the place of broadband services and policy in the United States, information abundance and asymmetry combine with regulatory devolution to make quality of service monitoring timely for citizens and users. Broadband QoS monitoring could emerge as a value added information services enterprise for consumers, policymakers and industry.
For the Public
Due to the limited steps the government is taking to administer broadband markets -- essentially monitoring the provisioning of multiple independent Internet Service Providers over the nation’s cable systems and the interoperability of Instant Messaging -- a quality of service monitoring system could serve the public interest as well as provide critical metrics to broadband industries.
The public finds itself in the curious situation of information abundance characterized by information asymmetry. As more citizens use broadband as a result of declining real costs and greater varieties of broadband service offerings, users will become more heterogenous.
Some will want to spend less time evaluating and selecting services, and will instead prefer time–optimizing, simplified information which generates confidence that they are getting what they are paying for. Such information could reduce consumer anxiety over choosing an opaque high tech service.
Others will prefer more detailed descriptions of broadband QoS. Many, like those involved in on-line, real time game playing, have already demonstrated their interest in service characteristics like low latency to make their experiences more compelling, and actively seek granular quality and performance information of the sort our QoS software can provide.
The ability of users to select a service offering will depend on how successfully system designers create easy‑to‑understand presentations of available choices. Even if software agents eventually weigh the offers made by various service providers, the user interface for configuring the network service "purchasing agent" will have to pose the options in readily comprehensible formats. A service that can be trusted to provide up‑to‑date assessments of the quality of service offered by diverse networks will be in demand as the variety of options for personal communications multiplies.
To address this situation, informational labels like those below, would enable the public to evaluate broadband services.
Figure One shows a user-friendly label evaluating a service for the applications most people use. Figure Two illustrates a comprehensive label for users with greater interest in service quality.
Figure One: Broadband Rating
Figure Two: Broadband Facts
Information presented in rating labels or on-line status reports would be collected by software placed on users’ computers. Such information includes:
Network Performance - which spans a number of dimensions:
Speed and data rate- a specified speed or speed range that will be provided to the customer
System uptime – average percentage of service up-time.
Latency – average end-to-end delays to a sample of sites on the Internet.
Jitter – variability in latency
Mail server response – time for mail server to respond to requests
Mail server speed – data rate attained by mail server when downloading mail to user’s computer
The following data would be retrieved by research of publicly available information:
Cost - a detailed description of the cost of the service, and the billing period,
Customer service - list of customer services contacts, and average support call wait time.
Privacy policies – implementation of published privacy guidelines as certified by a third party
Restitution - amount of money to be refunded in cases of network problems and/or lack of service.
Third-party arbitration - name and contact information where customers can lodge complaints against a high-speed access service provider.
If the government seeks to encourage market-based Broadband deployment solutions, QoS monitoring could emerge as an information services enterprise to supply broadband industries with:
Diagnostic capabilities for a fuller picture of the user experience than network operations data. With QoS monitoring, broadband network providers can scale bandwidth and establish price tiering for differentiated services. This intelligence will help to maintain and to increase revenues by yielding information on customer use, which can then be employed to optimally address consumer needs for bandwidth.
Data to use as a marketing tool to showcase broadband providers to competitive advantage. In markets where technically superior systems are deployed, monitoring results can be touted in advertisements to promote the broadband network provider.
Functionalities to enhance customer retention by providing easily understandable comparison and contrast among ISPs on broadband networks. The QoS monitoring simplifies the subscriber’s choice among affiliated and independent ISPs. It also reduces churn by providing customers with sufficient information to satisfy them that they are receiving adequate service via cable modem or DSL dedicated line.
Simplifying disputes between consumers and providers as to whether the provider has met its contractual obligations.
A process for offering and verifying QoS guarantees to consumers and service providers will encourage the development of new services like Internet telephony and real-time games. As network performance improves, these services can be marketed more easily. Without independent monitoring of QoS, service innovations will be slower to develop as uncertainty over the availability of reliable network services will deter investors and consumers.
Finally, reliable metrics of broadband service quality will clarify public policy goals.
Reliable, ubiquitous Broadband QoS monitoring is an essential policy tool to ensure the appropriate definition of broadband products and services. Furthermore, the widespread availability of this QoS data serves a tri-partite goal of stimulating public understanding and demand for broadband services, facilitating broadband policymaking and strengthening the broadband marketplace.
Charles Francis Adams, “The Railroad System,” The North American Review, 1867 and “A Chapter on Erie,” The North American Review, 1869.
John Quincy Adams, A Report on Weights and Measures, 1821.
Francios Bar, Stephen Cohen, Peter Cowhey, Brad DeLong, Michael Kleeman and John Zysman, “Defending the Internet Revolution in the Broadband Era: When Doing Nothing is Doing Harm,” Berkeley Roundtable of International Economics, 1999.
Marver Bernstein, Regulating Business by Independent Commission, 1955.
Robert Boyer, Theory of Regulation, 1986 and The Flexibility of Work in Europe, 1986
Alan Brinkley, End of Reform: New Deal Liberalism in Recession and War, 1995.
Joel Brinkley, Defining Vision: The Battle for the Future of Television, 1997.
Rhonda Crane, The Politics of International Standards: France and the Color TV War, 1979.
Federal Communications Commission, Broadband Today, Cable Bureau, 1999.
Federal Communications Commission, Summary of Report and Order in CC Docket Nos. 00-199, 97-212 and 80-286 and FCC News, “FCC Modernizes Accounting and Reporting Requirements,” October 11, 2001, Common_Carrier/News_Releases/2001/nrcc0139.html.
Donald G. Fink, editor, Television Standards and Practice: Selected Papers from the Proceedings of the National Television System Committee and its Panels, 1943.
Robert Britt Horowitz, The Irony of Regulatory Reform: the Deregulation of American Telecommunications, 1989.
Geoff Huston, ISP Survival Guide: Strategies for Running a Competitive ISP, 1998
Gabriel Kolko, Triumph of Conservatism, 1963.
Dave Kosiur, Building and Managing Virtual Private Networks, 1998
Lawrence Lessig, “Innovation, Regulation and the Internet,” The American Prospect, Volume 11, Number 10, 2000
David R. Linde, editor, A Century of Excellence in Measurements, Standards and Technology, NIST Special Publication 958, 2001.
Alain Lipietz, Production, Work and Territory, 1986
Thomas K. McCraw, Prophets of Regulation, 1984.
Andrew M. Odlyzko, “Paris Metro Pricing: The Minimalist Differentiated Services Solution,” Proceedings 1999 Seventh International Workshop on Quality of Service (IWQoS '99), IEEE, 1999, and “Paris Metro Pricing for the Internet” Proceedings ACM Conference on Electronic Commerce (EC'99), 1999.
Ithiel deSola Pool, Technologies of Freedom, 1984
Andrew Powell, Chairman, Federal Communications Commission, March, 2001.
George Sheirs, editor, Technical Development of Television, 1977.
Martin J. Sklar, The Corporate Reconstruction of American Capitalism, 1988.
Ida Tarbell, History of the Standard Oil Company, 1904.
Telecommunications Act of 1996, Public Law Number 104-104, 110 Stat. 56, 1996
Joseph H. Udelson, The Great Television Race: A History of the American Television Industry 1925-1941, 1982.
Tadeo Umaseo, “Johoka Shakai,” Choukoron, 1963
Nancy J. Victory, Assistant Secretary for Communications and Information, Department of Commerce, “Removing Roadblocks to Broadband Deployment,” A Speech before the Competition Policy Institute Conference, December 6, 2001.