|Broadband: Bringing Home the Bits Committee on Broadband Last Mile Technology|
source ref: ebookbrand.html
Broadband, the darling of techno-sophisticates and an object of interest to a growing number of politicians and government officials, as well as to the general public, is often misunderstood. The term itself, originating in the characterization of a communications channel's capacity (in contrast to narrowband), has come to be used as, among other things, a marketer's label for advanced cable television service, the 21st-century incarnation of the early 1990s "information superhighway," and one element of the next stage in the development of the Internet. Broadband has been a beacon for investors and a stimulus for entrepreneurs and mainstream businesses, and it has intensified debates about the public interest in information and communications infrastructure. It is as an enhanced means of access to the Internet that broadband has begun to have real traction in terms of actual deployment and use, and it is in this sense that the term has become commonly understood. Years of assertions by technology gurus, business executives, and marketers about the potential promise of broadband have given way to a small but rapidly growing U.S. population who are using a first generation of broadband for faster Internet access from their homes.
Today, people are asking about when most, if not all, of the population will partake of it. Basic questions about who pays for and who benefits from broadband are being discussed in communities, state and regional government entities, and in the Federal Communications Commission and Congress at the national level1--as well as in trade associations, consumer advocacy and other public interest groups, and by civic organizations, telecommunications and Internet policy scholarship forums, and groups concerned with international economic development. That such a diversity of organizations share a common interest in broadband underscores how much the Internet has become accepted as important to the U.S. economy and society, a marked contrast to the early and mid-1990s when the federal Information Infrastructure Task Force worked hard to proselytize the Internet and related services.2 Various factions point to broadband as a compelling reason for shifts in telecommunications policy; these include proponents of both more and less government intervention. Understanding the nature of broadband and what is involved in getting it to more consumers is one of the major goals of this report.
Broadband, in the sense of high-capacity communications channels, is already present throughout much of the communications infrastructure. Fiber optic links with very large capacity are already commonplace within the networks of telecommunications carriers and are available for local access in many locations, albeit at high costs affordable by only larger businesses or organizations. The broadband challenge on most people's minds today is how to make higher-capacity connections available on a more pervasive, affordable basis. In particular, how can one best extend high-speed connectivity to users in homes, small businesses and smaller offices of larger organizations, local governments, and so forth? Widespread use--marked by new patterns of information flow--not only would benefit the individuals connected, but also could lead to qualitative changes in how people interact with family, community, and the workplace, with potentially profound social and economic implications. Broadband is viewed by some as a double-edged sword: networking could promote economic development, yet electronic commerce also has the potential to displace local businesses. (Present and potential applications and impacts are considered in Chapter 3 in this report.)
Extending the reach of broadband generally implies building on the existing communications infrastructure base, either incrementally or through significant investment in new infrastructure. It is an expensive undertaking to deploy broadband to households and to small businesses and other small organizations. Each of these is a small economic unit, and most premises are located a significant distance from a so-called point of presence--which is a location where a communications service provider can get economies of scale by aggregating traffic from many customers onto its core, high-capacity links that connect to other parts of the network, including access to the Internet. (In the telephone network, these points generally are located at central offices, while in hybrid fiber coax cable systems, these points are known as head ends.)
The link between the point of presence and the customer--using either existing communications infrastructure or new facilities--is frequently referred to as the "last mile," because it represents a bottleneck that constrains the benefits the consumer gets from the rest of a network, which is literally at some distance. Greater difficulty and cost are associated with dispersed populations, whether they have low density, with homes being far apart and farther from the local point of presence, or are remote, with an entire community, whatever its density, being many miles from the nearest existing aggregation point.
A major goal of this report is to examine whether broadband deployment is working and what, if anything, needs fixing. There are very different perspectives on what is happening and how well the process is working. In the course of its work, the Committee on Broadband Last Mile Technology learned about several different views described below, all of which shaped its thinking:
Finally, a note on perspective as it influences terminology. While "last mile" is the more commonly used term for the local access network, the challenge is frequently put in terms of building the "first mile"--that is, building out from end users to the network. From this perspective, the last mile might be thought of as a more supply-driven concept, while first mile refers to a more user-centered view that emphasizes social and economic benefits at the local level. This report uses these terms interchangeably.
The character and evolution of the existing communications infrastructure provide lessons applicable to today's broadband deployment challenge. Some options for broadband decrease costs by using existing infrastructure for new purposes. And the business models, policy, and regulation developed in the contexts of older communications infrastructure continue to be applied to the new technology. That base has many faces: there are multiple communications networks, which have emerged, evolved, and coexisted in largely self-contained fashion. Thus, it is helpful to briefly review the communications past and its evolution over recent decades to a digital infrastructure capable of supporting broadband.
Traditional circuit-switched telephony is, with the exception of high-capacity lines leased by large customers, a way of providing analog narrowband last mile access. The public telephone network was originally built for voice communications but for some four decades has been used increasingly for data communications. At its core--in the channels that aggregate communications from many users--the telephone network has long had large, and growing, bandwidths. At the edges, with the exception of a few customers (chiefly larger organizations) that have been able to lease high-bandwidth private lines, customers have historically been connected through relatively low capacity twisted pair links (this is often referred to as the "local loop"). Analog telephony has evolved to digital, with high-speed digital signal transmission commonplace throughout the public telephone network except in the last mile segment. Digital transmission over the last mile has been possible for many years with the addition of dial-up modems. Integrated Services Digital Network (ISDN) never took off in the consumer market, and more recently, DSL technologies add equipment at both the customer premises and the central office to leverage the existing last mile twisted pair infrastructure in order to provide higher data rates.
Two-way wireless networks, which originated as consumer services to provide analog cellular telephony, complement these other infrastructures. Digital transmission also extends into the last mile in the current generation of cellular telephony (although analog cellular networks continue to be used). In recent years, a second generation of digital services has been deployed in more populated areas; these services support very limited data communications, including access to some Internet services. Second-generation services have grown dramatically in popularity as service prices have dropped and handsets have improved in size and battery life. So-called third-generation service, which will offer greater bandwidth (though less than the hype would suggest), inspires considerable speculation. Even with today's generation of technology, various forms of access to Internet-like services--most notably text messenging, but also including access to information and commerce--have proved popular in countries other than the United States, fueling speculation about their potential in the U.S. market.
Broadcasting in the form of radio and television has been in place for many decades. While radio and TV have very large bandwidths and may make use of digital signal transmission, none of these services fits today's common understanding of broadband. This is in part because, unlike the more general purpose, generally Internet-based broadband offerings of today, they integrate physical- and higher-layer functionality. That is, the services are aimed at particular types of communication or content (e.g., broadcast radio or television), much as the public telephone network has been designed to support a particular set of voice communications services, and they have emerged, evolved, and coexisted in self-contained fashion. Some proposed applications are data-centric, however, and may play a role complementary to the digital communications services discussed in this report. Since the 1980s, direct broadcast satellite has used satellite transmission to provide many channels of service over very wide areas, and this technology has been further developed to provide two-way broadband service delivery.
Cable television networks started out as a way of extending the reach of the broadcast networks. As this analog, one-way infrastructure grew, it began to distinguish itself from broadcasting through development of its own content. In more recent years, cable networks have joined telephony networks in the public debates about broadband because recent and ongoing upgrades have added support for two-way communication at comparatively high bandwidths.
The widespread use of computers in the home--most notably and visibly today in the form of the general-purpose personal computer, or PC--has helped to transform expectations for consumer electronics, which historically focused on entertainment and specific personal services. Computers in the home provide a broader base of support for work, education, and other nonentertainment activities within the home. The general-purpose nature of PCs implies both breadth and a multitude of options for future use. The home PC, in turn, has proliferated in part because of its potential for connectivity outside the home (today chiefly through the Internet). Capability (e.g., storage, processing speed, and support for audio and video) in home PCs and the richness of the content available through the Internet--and therefore the bandwidth needed to access it easily--have increased in tandem. These trends have been the most obvious drivers of those seeking broadband in the last mile. Forecasts of future demand involve extrapolations from the rapid growth in adoption and speed that has been experienced thus far. An important caveat is that these figures derive from the early adopters: home penetration by PCs, which grew relatively quickly in the 1990s, has been leveling off, and there is reason to believe that the explanation goes beyond simple costs.
In each of these instances, there have been well-defined roles for regulators at all levels of government. In the case of broadcasting, the airwaves were declared public property and the Federal Communications Commission (FCC) was assigned the task of standardizing frequency bands and interference masks for terrestrial transmission, and it licensed the use of spectrum segments for particular well-defined services. In telephony, geographical boundaries served as a foundation for defining the respective roles of state and federal regulators.3 Cable franchising was deemed to be largely a local responsibility.4 In a series of FCC proceedings known as the Computer Inquiries, computers and related data processing functions were deemed to be an area that would be largely unregulated.
Today's debates about broadband have their roots in the 1990s, when the growth of the Internet and increasing experience with conventional approaches to data communications triggered broad discussions about national (or global) information infrastructure (NII). These general concepts, as well as development of the first generation of broadband local access technologies, set the stage for considerable speculation about a broadband future and invoked a vision of rewiring America to take advantage of these new capabilities. Although the 1980s and 1990s saw the deployment of a series of experimental broadband trials, it was not until 1998 that deployment began in earnest.
While not all of the original NII visions have not come to pass, what has been proved since the mid-1990s is the impact of the Internet--a single, general-purpose communications platform capable of delivering a wide range of content and applications. But interest in Internet access has not been enough to resolve other uncertainties, and incremental approaches have been successful. So the wholesale rewiring of America that had been hoped for by many in the early and mid-1990s did not come to pass. Instead, leveraging of existing wiring has dominated the broadband scene, accompanied by limited investment in wireless infrastructure and only spotty investment in new wireline infrastructure deployment.
A major change since the mid-1990s is the nature of the regulatory environment, occasioned primarily by the Telecommunications Act of 1996. The 1996 act asserts the goal of marketplace competition, but early experience has shown problems in achieving competition and raises questions about how much competition is realistic to expect (see Chapters 3 and 4 in this report). An important question today is whether the framework established in the act deserves rethinking in light of the subsequent evolution of communications technologies and services.
Still, despite these changes in the environment, many contemporary issues related to broadband are, in fact, not new, as demonstrated by a review of three earlier reports by the National Research Council's Computer Science and Telecommunications Board (CSTB) (see Box 1.1). For more than a decade, broadband in the last mile has been understood to be a key to maximizing the benefits of the communications and information infrastructure. Now, as then, there are multiple technologies and industries that can advance the infrastructure in general and broadband in particular. Optical fiber systems continue to promise the most bandwidth, but at the highest cost and risk. Private investment is still viewed as an essential ingredient, but it continues to be inhibited by uncertainty about what consumers will buy and what business models will succeed. Today, as then, it is understood that government action (or inaction) has the potential to both inhibit and promote investment.
FCC data based on reports from carriers show rapid growth in broadband subscriptions by residential and small business customers, with the total growing from roughly 1.8 million in December 1999 to 5.2 million subscribers in December 2000 (see Figure 1.1).5 Looking specifically at residential users, an October 2000 report from the National Telecommunications and Information Administration (NTIA) indicates that 4.4 percent of all U.S. households, or 10.7 percent of households with Internet access, had broadband access as of August 2000.6 Market research reports from 2000 provide consistent figures; for example, a November 2000 study by the Cahners In-Stat Group found that roughly 9 percent of U.S. households that access the Internet use a form of broadband Internet access.7 Reports from mid-2001 show further growth: a total of more than 5 million cable modem subscribers and more than 3 million DSL subscribers.8 There have, however, been hints that growth has been slowing, at least in some market segments: second-quarter 2001 reports from Verizon and AT&T Broadband show growth below the 2000 rate, and early 2001 also saw contraction in the competitive local exchange carrier (CLEC) business.9 Penetration rates can be much higher than the average in markets where broadband has been available for several years. For example, in Portland, Maine, an early test market for TimeWarner Cable, about one-quarter of households have become cable modem subscribers.
Access rates--the fraction of households that could subscribe to broadband if they chose to--are substantially higher than current subscription rates. A survey of Internet users commissioned by the General Accounting Office indicated that some form of wireline broadband access was available to 52.4 percent of Internet users--25.4 percent via both cable and DSL, 16.9 percent via cable only, and 10.1 percent via DSL only--as of May 2000.10 As of mid-2001, approximately 60 million homes reportedly had cable modem service available, and 52 million homes had DSL service available.11 Availability and use of broadband are both correlated with town size. Nielsen/Netratings found that broadband users in the top 25 local markets (by size) constituted nearly two-thirds of all broadband users in the United States.12
Past communications infrastructure has been the subject of efforts by industry and governments to extend benefits to ever-larger segments of the population. In the case of telephony, the effort has been in the form of interventions expressly aimed at reaching all Americans regardless of location or income level. At the same time that broadband services have reached many, the differences in access to bandwidth among segments of the population are becoming pronounced--including bandwidth disparities between work and home or between urban and rural areas.
The economic challenges in extending communications services to households can be seen in the history that led to today's baseline. The penetration of communications technologies takes time, reflecting the time it takes to build the necessary infrastructure and attract subscribers. As Figure 1.2 shows in displaying the penetration of several communications technologies over the period 1970-2000, broadband is in the early stages of this process. For example, broadcast radio and television allows a network service provider to reach wide areas using relatively few transmission facilities (terrestrial or satellite). Yet even this kind of infrastructure does not reach quite all households and areas, reflecting broadcaster choices about where and in what to invest (decisions that benefit from the rights to a service area associated with a government license). Broadcast television reaches many households, though reception is of varying quality. Broadcasting has been largely dependent on advertising support,13 and expanding access has depended in part on consumer electronics advancing so that radios and television sets became increasingly affordable (in the pre-cable-TV era). Cable service grew faster. It is available to roughly 97 percent of homes, and 68 percent of households subscribe.14 Direct broadcast satellite service covers the full continental United States (assuming the household has an unobstructed view of the satellite). It benefited from attractive economics: high capacity meant that it could offer a wide range of content via a subscription model, and once the very high fixed cost of the satellites and other transmission facilities was paid, an unlimited number of receivers could be added within the satellite footprint with minimal incremental investment.
It took decades to achieve a reach to over 94 percent of households,15 using copper wiring in the last mile, but the technical path to deployment--pervasive copper wiring backed by a centralized circuit switch--was, at least in retrospect, straightforward compared with what one observes with broadband today. This near-universal reach reflects both costs to network service providers and regulatory programs that promoted so-called universal service and provided financial support to providers to help realize that objective (regulatory history is briefly described in Chapter 5 in this report). The web of internal support flows that historically have characterized universal service policies has been complex; it evolved over time, notwithstanding the relatively homogeneous and slowly evolving technology that was the subject of these policies.
The oft-cited penetration numbers for different communications services can mask considerable variations in the nature or quality of the service. Both basic and higher tiers of capabilities have arisen in all of the traditional or conventional communications infrastructures. For example, touchtone and multiparty capability emerged as options in telephony, FM emerged as a "higher-quality" option compared with AM in radio, and color emerged as a "higher-quality" option compared with black-and-white TV. Cable came along to provide yet another TV option, but over a different infrastructure, with service tiers determined by the nature and amount of the programming content. Even dial-up Internet access is subject to variation. The speed ceiling is limited by the audio bandwidth of the public telephone network, but the floor depends on the quality of the individual telephone line. As a practical matter, what has been universally available is a lowest-common-denominator service, although upgrades in provider networks tend to raise that floor over time (for example, touchtone service has become the norm in telephony, and party lines have gone by the wayside).
As is seen in many markets, saturation--or at least high and slowly growing levels of penetration--tends to prompt the introduction of enhanced, higher-value and higher-priced offerings in the hopes of increasing provider revenues. Sometimes this is in the form of new content or applications, and at other times it reflects upgrades to the hardware and software of the communications network itself. And, over time, new communications technologies arise that complement the existing ones, offering new or significantly enhanced capabilities. The experiences of the 20th century suggest that few new technologies completely replace earlier ones, and that whatever substitution occurs does so over a long timescale. As demonstrated by the rapid rise of direct broadcast TV, new technologies can, however, pose a significant challenge to the existing players.
Complicating an understanding of broadband in the last mile is the evolution of the community context. On the one hand, the community can compensate in part for a lack of home access by providing access to broadband capabilities in publicly owned spaces (e.g., schools and libraries) and through private organizations (e.g., businesses as employers). On the other hand, the community can stimulate home access demand by generating content and opportunities for civic interactions--as well as supporting organizational use of the Internet, telecommuting and distance learning, public awareness and education about how to benefit from broadband, and so on. The lowest-common-denominator level of communications infrastructure--basic telephone service; television via broadcast, cable, or satellite; and dial-up Internet access--may have been taken for granted in all but the areas hardest to serve. However, differences among communities raise questions about how broadband relates to local economic development for purposes of attracting broader economic opportunities (and a higher quality of life) for citizens (through access in homes as well as schools, hospitals, and libraries). There is, of course, a tension between the well-connected home as a place from which its residents can access people, information, and economic activity that may be based physically quite far away--and that same home as an entity with a specific physical location, which drives needs to access people, information, and economic activity locally. Today's islands of broadband have the potential to intensify other differentiators of home and community experience, which has led to the invocation of broadband in the evolving consideration of a "digital divide." Communities compose the social level where interhome (and intercommunity) differences in access, burden, and benefit are most apparent, and they are on the frontlines of the integration of business and nonprofit influences on local activity. Although 1990s public debates over information and communications infrastructure took a national perspective, communities may play a larger role in moving forward in the new century.
Because many costs are per-house-passed rather than per-customer-served and because of economies of scale, individual or isolated consumers cannot induce providers to deploy expensive infrastructure at a reasonable price. Thus, investors and industries search for clusters of consumers who will pay for service before the investors and industries will commit to upgrade or deploy infrastructure. One key recurring question is what types of service the customer will pay for and at what levels. Both the reach of different services and their enhancement paths reflect differences in industry and provider business models. Consumers usually pay for the in-home hardware (and software) required to use a communications service, but it varies as to whether they purchase it directly or lease it from a service provider. Cable television reflects consumer willingness to pay for access to programming content, in contrast to broadcast content, which is for the most part paid for by advertising. Consumers who place a telephone call to an Internet service provider (ISP) pay that provider for Internet access service, in addition to what they pay for telephone service and for the equipment they use, and in addition to what they may pay to access certain kinds of content or services through the Internet. What people will pay regularly for information and communication services remains an open question as the choices proliferate. Providers of traditional services approach this question with concerns about preserving traditional revenue streams as well as cultivating new ones. Providers of new content and services have been struggling to find the right mix of advertising and subscription revenue models. Related questions arise about who owns and/or manages the technology and content used in the home--consumer or service provider--and what are the life-cycle costs and benefits from the perspective of consumer and provider.
Thus, debates continue about the role of information and content--notably entertainment--as a stimulus to broadband deployment. The version in the late 1990s and early 2000s has centered on open access: what does it take to ensure that consumers can access information from any source available through the Internet? Do preferential arrangements between ISPs and entertainment programming providers undesirably constrain consumer choice? Do such arrangements play an important role in stimulating investment? The entertainment industries have proven business models for serving consumers (e.g., broadcasting or cable television), but the questions revolve around how and with what consumers are served in a more complex world. These issues have brought consumer advocates and Internet-related public interest groups into the discussion as well.
The Unpredictable Certainty16 had challenged the conventional wisdom of the time that called for a "killer app," a single use of communications infrastructure significant enough in scale--that is, generating enough revenue from customers--to bring down the risk in infrastructure investments. The experience of the late 1990s confirmed the absence of a new killer application, and the experience of the early 2000s--the recognition of weak performance or nonviability among many dot-com ventures--showed that even broad categories of seemingly successful applications (such as various instances of electronic commerce) are not, in isolation, sufficient to drive investment in last mile technologies. Only two proven "killer apps" have emerged. One, e-mail, is the same application that proved of enduring value in the data networks that preceded the commercial Internet. The other, World Wide Web access, is in fact not really a single application, but rather a general-purpose platform supporting a wide variety of content and services. While ISPs have been able to derive considerable revenue (if not always profitability) from providing Web access, few companies have demonstrated success in deriving substantial revenue or profits from Web-delivered content and services themselves.
Although popular discussions of broadband tend to focus on PC applications, typically uses of the Web, these are only one of many possible applications of broadband. The growing trend toward digital storage and communications--one facet of what is frequently termed convergence--is also stimulating interactions with other kinds of equipment and services. Today, as described in Chapter 2 in this report, home networks are largely about connecting PCs together and sharing a link to the outside but will increasingly also involve interactions among computers, entertainment devices, and other appliances.
Recent developments have underscored the political aspects of business and policy decisions related to broadband and have added to already significant regulatory and financial uncertainty, except in instances when court decisions and administrative decision making have been clarifying. The popular debates over open access illustrate how politics can have influence, but future, especially long-term, policy making relating to broadband should have more solid analytical foundations. Providing them is one mission of this report.
Recent developments also raise questions about how future information and communications infrastructures may deviate from the classic Internet as experienced in the mid to late-1990s. Certain wireless technologies and certain approaches to the ISP business provide illustrations of something that falls short of a "full-service" Internet. Viewed as a "half-full glass," such offerings may expand Internet access and other activities to new groups of consumers; viewed as a "half-empty glass," they may deprive consumers of choices with various economic and social benefits associated with unfettered access to all Internet content. The long-term implications are unknown, but it is important to understand the nature of the technical alternatives and their economic and other consequences.
In the meantime, the investment climate for major telecommunications infrastructure upgrades is uncertain. Company and investor disappointment over some previously touted technologies may be one factor. For example, ISDN, which was implemented slowly and was perceived as having significant shortcomings, has been adopted by only a modest number of users. The failure of the much-anticipated Iridium satellite telephone service underscored concerns about the financial risk of bold communications infrastructure investments. Many of the Internet start-ups that captured attention in the late 1990s tended to leverage more than add to the infrastructure. The optical networking start-ups tended to focus on the core of the network and access for very large business users. Start-ups that focused on local access, primarily DSL-based competitors enabled by regulatory developments, found themselves foundering in 2000-2001, and a more general slowdown in the telecommunications sector was apparent in mid-2001. Even in the face of this uncertainty, incumbents have invested in infrastructure upgrades to meet what they see as continued demand for broadband, and a new player in the form of geosynchronous satellites has introduced a new option for broadband. It is unclear at present to what extent uncertainty in some segments of the business could contribute to an overall slowdown in investment, and caution should be exercised in projecting long-term prospects for broadband based on the present downturn.
The world of broadband is significantly more complex than that of the traditional or conventional communications infrastructures. This report is designed to assess the nature of broadband, its deployment and acceptance, expectations about future deployment, and the potential longer-term technical and social implications of broadband access. The challenging and sometimes necessarily speculative nature of this analysis makes it inherently imperfect. Notwithstanding these limitations, an additional goal of this report is to make useful recommendations about how best to maximize the potential impact and rewards of broadband, generally exploring what will be required to achieve ubiquitous broadband access, in the sense of both expanding the geographical reach of broadband facilities and addressing affordability issues.
The focus of this report is on providing access to fixed (i.e., nonmobile) users, but the label "last mile" also applies to mobile applications, and the report touches on mobile issues as appropriate. For example, to the extent that one wishes to support near-seamless communications across fixed and mobile locations, the two are coupled.
Broadband poses many analytical challenges that go beyond the complex technical and economic landscape. One problem is the lack of a common information base: detailed information is often proprietary; information presented in governmental proceedings is tailored to those proceedings; and systematic detailed data on deployment are hard to come by. Perhaps an even bigger challenge relates to the generally poor track record of both those within industry and outside observers in forecasting information technology (IT) developments or its economic parameters. Finally, basic assumptions will continue to shift. For example, changes underway include the nature of installation (do-it-yourself/off-the-shelf versus professional installer) and the specification of what can be shared (e.g., a shift from built to leased wireless towers or from single-user to shared conduit or fiber bundles). The future may well see change in licensing requirements (e.g., per-site versus per-system licensing of wireless service). Finally, the roles of the key players--consumer, community, communications-industry sectors, and all levels of government from town to federal--continue to be in flux.
Notwithstanding these and other challenges, the Committee on Broadband Last Mile Technology has attempted to put forth, by consensus, views about the broadband last mile that seek to have value in the 2- to 10-year time frame. While such a time frame might seem to be daunting in the face of the rate at which some of the basic technologies are advancing (Moore's law and its kin), the processes of deployment and acceptance have always proceeded much more slowly, and there seems to be no particular reason to expect a significant change in these time constants going forward.
1 Broadband-related bills introduced in the 107th Congress include S. 1056 (Community Telecommunications Planning Act of 2001), H.R. 2139 (Rural America Broadband Deployment Act), H.R. 1542 (Internet Freedom and Broadband Deployment Act of 2001), H.R. 267 (Broadband Internet Access Act of 2001), S. 150 (Broadband Deployment Act of 2001), and S. 88 (Broadband Internet Access Act of 2001).
2 For the flavor of that period, see CSTB's 1996 report The Unpredictable Certainty and its 1994 report Realizing the Information Future, National Academy Press, Washington, D.C.
3 In broad terms, federal regulators have primary responsibility for long-distance service, while state regulators have primary responsibility for local service. State commissions also regulate intrastate long-distance service, and the FCC regulates local telephone companies insofar as they provide origination and termination services for long-distance calls. The division of responsibilities between state and federal regulators is determined by the Communications Act of 1934, as amended.
4 Cable franchising authorities are local (municipal or county), except in states where a statewide authority has this power.
5 Interpretation is complicated because the figure includes small business as well as residential customers and because ADSL is separated from other forms of DSL, which are lumped under "other wireline."
6 National Telecommunications and Information Administration (NTIA). 2000. Falling Through the Net: Toward Digital Inclusion. NTIA, Washington, D.C., October. Available online at <http://www.ntia.doc.gov/ntiahome/digitaldivide/index.html>.
7 Cahners In-Stat Group. 2000. Broadband Consumers--Profiles and Strategies, Report No. BBWIS00-05SP. See also "Broadband Subscriptions Will Rise 77% Through 2004," Broadband Week, available online at <http://www.instat.com/rh/bbw/is0005sp_story.htm>.
8 The National Cable & Telecommunications Association (NCTA) estimates 5.5 million subscribers as of August 2001. See <http://www.ncta.com/>.
9 AT&T Broadband reportedly signed up 259,000 customers during the fourth quarter of 2000, but only 131,000 during the second quarter of 2001. Verizon signed up 90,000 customers in the fourth quarter of 2000 and 120,000 during the second quarter of 2001 (Christopher Stern, 2000, "Broadband Market Growth Slows," Washington Post, August 28, p. E01).
10 Determining the fraction of homes with DSL service available is more complicated than in the case of cable. DSL availability depends on central office equipment, line length, and the characteristics of the specific loop, while availability of cable modem service for the most part depends only on whether the subscriber's system has been upgraded.
11 Remarks of Robert Sachs, NCTA, June 11, 2001; Cablevision Blue Book, Cahners Business Information, New York, June 2001.
12 Nielsen/Netratings. 2001. "New York Local Market Dominates Broadband Usage, According to Nielsen/Netratings" (press release). Nielsen/Netratings, New York. May 15.
13 While public broadcasting has increasingly relied on commercial sponsorship, it still relies on tax subsidies, charitable support, and listener donations.
14 Cable TV Financial Databook, 2000, p. 10, as cited in National Cable & Telecommunications Association, 2001, Industry Statistics, available online at <http://www.ncta.com/industry_overview/indStat.cfm>.
15 Alexander Belinfante. 2001. "Telephone Penetration by Income by State" (data through 2000). Industry Analysis Division, Common Carrier Bureau, Federal Communications Commission. July. Available online at <http://www.fcc.gov/Bureaus/Common_Carrier/Reports/FCC-State_Link/IAD/pntris00.pdf>.
16 Computer Science and Telecommunications Board, National Research Council. 1996. The Unpredictable Certainty: Information Infrastructure Through 2000. National Academy Press, Washington, D.C.