‘Connecting the unconnected’: a critical assessment of US satellite Internet services

Abstract

An estimated 4 billion people worldwide still lack reliable access to the Internet, with the majority concentrated in developing countries. It is within this context that a significant expansion of the satellite Internet industry is taking place, bringing grand visions of social and economic growth through connectivity. Previously dismissed as a limited technology due to high operating costs and latency, as well as the failure of early ventures like Teledesic, satellite Internet has re-emerged in recent years as a viable alternative to the time and energy-intensive practice of laying thousands of miles of fiber optic cable. Utilizing innovative non-synchronous orbit constellations of small, mass-produced satellites that lower production costs and improve latency to better compete with broadband, new Silicon Valley–related companies such as OneWeb, SpaceX, and O3b have promised cheap, high-speed Internet access throughout the world. Drawing upon industry research, site visits, and interviews with technical specialists, development experts, and other actors, this article briefly explores the history of satellite Internet services, analyzes the promises of emerging satellite Internet companies, and argues that without serious commitments from governments and the private sector to follow through on this rhetoric, satellite Internet technology could fail to reach the communities that need it most.

Keywords

bottom billion digital divide Internet satellite Internet satellites Teledesic

The International Telecommunications Union (ITU, 2017) estimates there are 3.5 billion Internet users worldwide, or roughly 47% of the total population. The picture changes considerably when looking at penetration rates by region, however. While Europe and the Americas have connected the majority of their populations (68% and 66%, respectively), Asia and the Pacific (42%), the Middle East (42%), and Africa (18%) have significantly lower rates of Internet connectivity. Since the advent of the Internet, there has been a persistent drive to expand its global reach and user base. This process has been embedded in a ‘connecting the unconnected’ discourse that is predicated on a liberal politics of inclusion. This discourse, which oscillates between a technological altruism and corporate strategy, assumes that universal Internet access is desirable by and for all people in the world, and that the technology, once installed, will inevitably support education, finance, labor, culture, and governance (McLuhan, 1962; Parks and Starosielski, 2015). The mandate to ‘connect the unconnected’ in the age of the Internet has intensified given the high stakes of what it means to be on the wrong side of the ‘digital divide’. Despite these high stakes, very little direct consultation is actually done with publics living in developing regions that are disconnected. Such populations are often spoken for rather than listened to (Srinivasan, 2017) and have been dubbed in industry parlance the ‘O3b’ (other 3 billion), the ‘bottom billion’ (Collier, 2007; Heeks, 2009) or ‘last billion’ (Ben David, 2015). Such industrial discourses imagine people in developing regions as a giant untapped market. The causes and effects of the global digital divide have been addressed by scholars (Ragnedda and Muschert, 2013; Van Dijk, 2005), but fewer have analyzed techniques for bringing unconnected communities into the digital fold.

While the predominant means of connecting people to the Internet continues to be through fiber-optic cables, during the past decade a new crop of satellite Internet providers has emerged from Silicon Valley promising truly global Internet access. Companies such as O3B, SpaceX, and OneWeb have promoted themselves as trying to develop satellite constellations to serve underserved communities around the world. These projects have attracted the attention of scholars, including Fairhurst et al. (2008) and Pultarova (2018), who see satellite technology as potentially being fundamental to the future of the Internet, particularly in places where current infrastructure does not support broadband. In addition, Hansson et al. (2007) and Townsend et al. (2013) have explored how satellite Internet may benefit rural, unconnected communities in Europe. While satellite connectivity has the potential to benefit the unconnected, deeper interrogation of the claims made by these companies – that they will use their new satellites to serve underserved communities – is needed. Their corporate rhetoric often adopts a ‘connecting the unconnected’ discourse, combining altruistic visions of global community and technological integration with telecom privatization and market expansion. In this article, we critically examine emergent satellite Internet companies’ promises to ‘connect the unconnected’.

Perhaps it is no coincidence that this new generation of satellite Internet companies has roots in Silicon Valley, where both disrupting established industries and ‘making the world a better place’ are the dominant corporate mantras. Greg Wyler (2016), founder of OneWeb, has proclaimed that his satellite Internet company will ‘fully bridge the Digital Divide by 2027, making Internet access available and affordable for everyone’. SpaceX Vice President Patricia Cooper (2017) claims her company’s ‘Starlink’ satellite Internet constellation will ‘bring high-speed, reliable, and affordable broadband service to consumers in the U.S. and around the world, including areas underserved or currently unserved by existing networks’. O3b (named after the ‘other 3 billion’ people who lack reliable Internet access) began with a ‘vision to connect the unconnected’, and the company claims that without its efforts, such people ‘will not have the economic relevance, potential and opportunity which we all believe is so important’ (SatMagazine, 2009). Despite the promotional rhetoric of attending to the digitally unserved, the business models for actually delivering such services are less clear. Given this, it is possible that this rhetoric is a cover for the predation that, in Anand Giridharadas’ (2018) view, is often hiding behind elite ‘concern’ and solutions for social problems in the developing world.

This article uses industry research and discourse analysis to assess the bold claims and promises made by satellite Internet companies and asks whether their visions of ubiquitous connectivity are viable from technical and business perspectives. Our analysis draws on a range of technical and industry sources, as well as field research and interviews with eight executives and technical experts from leading satellite companies conducted at the SATELLITE 2018 convention in Washington, DC. We begin with a brief history of satellite Internet services, then describe challenges of orbit, bandwidth, and latency faced by satellite operators, and finally discuss three satellite Internet companies from Silicon Valley that emerged to address those challenges. All the while, we assess whether and how these companies are living up to their promise to ‘connect the unconnected’. Our research suggests that this may be less feasible and likely than these companies claim due to lingering technical challenges and the lack of a business model that can support under-resourced communities.

A brief history of US satellite Internet services

In 1945, science fiction writer Arthur C Clarke first conceptualized satellite communication. He proposed that a geostationary orbit matching the earth’s 24-hour rotation would be the ideal arrangement for global communications as it would ensure that the satellite was never out of range (Clarke, 1945; Pollard, 2018; Tweney, 2011). Clarke’s vision was eventually realized in 1960 with the launch of Echo 1, NASA’s first experimental communications satellite. This was followed in 1962 by Telstar 1, a joint collaboration of the United States, French, and British broadcasting agencies, developed by Bell Telephone Laboratories for AT&T. (Mann, 2012). The launch of Telstar 1, ‘the greatest American device yet developed for global communication’, was a milestone achievement (Schwoch, 2009: 130). Unlike Clarke’s original design Telstar 1 operated in low earth orbit (LEO) at a height of roughly 3600 miles above earth (NASA, 2012). The satellite relayed its first transatlantic broadcast on 24 July 1962, a milestone achievement that brought ‘television journalism into the modern age’ (Mann, 2012).

Syncom 3, developed by NASA and launched in 1964, was the first communication satellite to reach geosynchronous orbit. This satellite served as the basis for future generations of comsats as it relayed radio, television, and other telecommunication signals. As the comsat industry took shape during the 1960s, the Defense Advanced Research Projects Agency (DARPA) project in the United States was beginning to develop Internet infrastructure on the ground. Advancements in Internet technology during the mid-1960s, most significantly the development of packet switching (fragmenting a data message into parts and sending them independently for optimum speed), opened up new possibilities for transmitting data via satellite (Abbate, 1999: 120; Edwards, 1996: 270). In 1973, DARPA successfully used the international comsat Intelsat 1 to connect the University of Hawaii and University College London to the Advanced Research Projects Agency Network (ARPANET) (Abbate, 1999: 121). This experiment enabled the Atlantic Packet Satellite Network (SATNET), jointly sponsored by the United States, the United Kingdom, and Norway, and interlinked a group of universities and research centers in those countries. SATNET could operate independently as a closed network, or be linked to existing terrestrial radio and telephone data networks (Packet Radio Network (PRNET) and ARPANET) and, as such, formed the first international ‘Internet’ (Abbate, 1999: 132).

During the late 1970s, satellite providers launched direct broadcast satellite (DBS) systems in the United States, which allowed consumers to downlink television signals directly from a satellite to home receivers. An October 1979 Federal Communications Commission (FCC) ruling allowed private citizens to operate a home satellite dish without applying for a federal license, and by the 1980s, there were numerous competing US satellite companies, including Microwave General and Channel One, Inc., offering services to consumers (Regulation of Domestic Receive-Only Satellite Earth Stations, 1979; The New York Times, 1979). It was not until the mid-1990s, however, as technology improved and home satellite dishes became more commonplace, that satellite Internet services became viable for home users. In 1996, Hughes Network Systems introduced DirecPC, a one-way satellite Internet system using a Hughes communication satellite (D Rehbehn, 2018, Interview with Dave Rehbehn, Vice President, International Division, Hughes Network Systems). The system required both a satellite receiver and a separate, modem-based Internet service provider (ISP) connection in order to send requests to a server and receive data back via satellite. The service offered up to 400 kbps download speeds, a substantial improvement over dial-up modem speeds (Stewart, 1999). At its peak, the service had roughly 100,000 subscribers, mostly in rural or remote areas where cable was not yet available (Evans, 1996). This amounted to a miniscule percentage of the 100 million estimated Internet users in the United States at the time (US Internet Users, 2018).

Why were consumers so reluctant to embrace satellite Internet, particularly when DBS technology had been adopted for television service? A crucial difference exists between Internet packet switching and television broadcasting. While television signals are broadcasted unidirectionally, Internet connections require a two-way exchange. This need for satellite Internet dishes to transmit and receive data results in a far more complex domestic set up, with corresponding challenges related to cost, capacity, and latency. Compared with cable broadband Internet, satellite connections were challenging to install and expensive to operate – the DirecPC dish and hardware cost US$799 in 1996, in addition to a US$150 installation fee and monthly fees ranging from US$50 to US$150 depending on how much data were used (The Arizona Republic, 1996). Because satellites are launched with a fixed bandwidth capacity (the maximum rate of data transfer) to service a large number of customers, satellite Internet providers must cap or throttle the amount of data available to individual users. In the early days of the Internet, this presented less of an issue, but as streaming video (an hour of high-definition video streamed from Netflix uses roughly 3GB of data, for example) and other data-intensive applications have become more common, satellite’s limitations have become more glaring.

Another key challenge for satellite Internet services is latency. Latency is the delay between a request for data and the start of a transfer. It is an inherent aspect of any Internet connection, but satellite connections create additional layers of delay due to the distance the data must travel between the ground and orbit, which is limited by the speed of light. Most geosynchronous satellite Internet connections have a latency of roughly 500–600 ms (Miller, 2017). The television viewing experience is generally unaffected by satellite latency, because the data are transmitted and decoded at the same predetermined speed, but the same latency has a significant negative impact on Internet browsing and streaming speeds as content is not able to load instantaneously and ‘lags’. Such a delay became particularly noticeable and disruptive with interactive and content-rich Internet services such as live multi-player gaming, Skype, video sharing, livestreaming, and social media, which emerged with broadband networks during the early 2000s. Broadband users came to expect a seamless network experience. Given intense competition among ISPs as the Internet emerged, the latency caused by satellite backhaul use could compromise a company’s market position.

The rapid growth of cable broadband connections (between 2000 and 2010, the number of US adults with home broadband access increased from 1% to 61% according to a 2018 Pew Research Center report) also presented a serious challenge to satellite Internet services. While laying underground and undersea fiber optic cable is expensive and time-consuming, the costs tend to be less than those required to design, build, launch, and maintain a space-based communications system. Furthermore, terrestrial systems have the advantage of being relatively quick to repair, upgrade, and expand based on consumer demand, and can empower regional, national, and local companies and workers. Because of this, in much of the world, fiber optic cables became the preferred infrastructure for Internet traffic. According to Viasat President Rick Baldridge,

the market for satellites had really been the people that had no choice. If you couldn’t get anything else, it was a technology of last resort. It essentially had a ubiquitous coverage but really, not much data. It had been relegated to things like transactions at gas stations. (Hurst, 2018)

Still, some telecommunication experts held great faith in the potential of satellites. Entrepreneur Craig McCaw envisioned a satellite Internet system that could overcome the speed and cost limitations that hindered the technology’s wide-scale adoption. In 1990, he launched a company called Teledesic with the ambitious goal of providing global high-speed Internet access through a constellation of as many as 840 satellites in LEO. The venture aimed to offer ‘a full array of high-speed video, data, and voice services to the vast portions of the planet beyond the reach of today’s ground-based wired and wireless networks’ (Feder, 2000). Historian Martin Collins suggests that ‘Teledesic, with its promise of planetary coverage, heralded the possibility of bringing the benefits of networked personal computing to users everywhere, adding symbolic heft to the Web’s potential for social transformation’ (Collins, 2018: 133). Promoted as worldwide ‘Internet in the sky’, Teledesic attracted considerable hype as well as funding from prominent investors like AT&T and Microsoft’s Bill Gates (Feder, 2000). The company also secured key spectrum licenses from the FCC and ITU (Via Satellite, 2003).

Ultimately, however, Teledesic proved unfeasible. The complexity and expense of constructing the satellite network exceeded the billion dollars in venture capital the company raised. In addition, industry experts believed the project’s market justification was flawed. In a 2003 interview, industry consultant DK Sachdev suggested that while Teledesic’s technology was revolutionary, it tried to ‘force-fit’ a market to the product. He remarked as follows:

It is true that billions of people around the world are without phone service. But what they need is local phone service and not connectivity around the world. And they certainly are not going to pay even 50 cents per minute for a call, notwithstanding the revolutionary technology being offered. (Via Satellite, 2003)

Teledesic suspended its operations in 2003, having never launched a single satellite (Goodwins, 2002).

Following Teledesic’s failure, the idea of a global satellite Internet network was relegated to the sidelines of the telecommunications industry. Companies continued investing billions of dollars in terrestrial Internet infrastructure, running thousands of miles of cables across continents and ocean floors. Satellite Internet settled into a role serving industry and government needs, as well as a diminishing number of rural users still beyond broadband’s reach. One area of substantial growth in satellite Internet services has been the travel industry, particularly commercial air and cruise lines. Two-thirds of all airlines will offer in-flight Internet connectivity via satellite by 2020 (Garcia, 2017). Satellite Internet has also become available on nearly all major cruise lines, although large numbers of users can reduce speeds and increase costs given the limited bandwidth available to accommodate the entire ship (Lyons, 2017).

In a limited number of cases, satellite Internet has been offered in rural areas using community VSAT (very small aperture terminal) systems, which allow multiple users to purchase access through local gateways. Since the early 2000s, companies including Hughes Network Systems and Viasat have been operating these systems in countries such as Mexico, Russia, and India. Hughes Network Systems’ Vinay Patel suggests, ‘Sharing a VSAT’s broadband connection amongst 10, 20 or even 30 users … can reduce the cost of service per subscriber to the affordable range of $10 to $20 per month’ (Patel, 2018). Discussing his company’s community-based VSAT in Mexico, Viasat’s Kevin Cohen explained the following:

You can think of it as a cyber café from space. We put a satellite dish up and an outdoor (Wi-Fi) access point up to widen coverage around a retail location. We sell Internet access to individual people. They can choose bite-size chunks or larger chunks. (Freeman, 2018)

VSAT represents a practical solution to the challenge of satellite Internet affordability in rural, low-income communities. But are these examples scalable, or simply good one-off public relations stories? Questions also remain as to whether or not these systems will maintain their low costs and work with new non-synchronous constellations and swarms, which will require re-designed satellite dishes that swivel to follow satellites as they pass.

Outside of these limited, speciality applications, overall consumer adoption of satellite Internet remains low. According to a 2016 US Census Bureau report, only 6.3% of Americans access the Internet through a satellite connection (Statista, 2016). In Europe, satellite is used by only 1.2% of broadband subscribers (Bell, 2018). While speeds have improved considerably in recent years (a 2013 FCC report found ‘significant advances in satellite Internet performance’, with the new generation of satellites offering ‘performance as much as 100 times superior to the previous generation’), the increased cost, data caps, and other restrictions associated with satellite systems have consistently limited their appeal, particularly where cable-based broadband is widely available (Measuring Broadband America, 2013).

The introduction of satellite communication in the 1960s expanded possibilities for global connectivity, and satellites now form the backbone of numerous vital communication systems. Despite this, consumer Internet has remained largely earth-bound, even though large portions of the world’s population are beyond terrestrial broadband connections. The challenges of speed, cost, and the lack of a profitable business model for connecting low-income, rural markets have largely relegated satellite Internet services to cruise ships and oil rigs. This has remained the status quo until recently. A new crop of satellite Internet companies has promised a revolution in access and affordability.

Satellite Internet’s second wave

Despite limited adoption of satellite Internet by home users, the dream of breaking the chains of the global fiber optic network and connecting the billions of potential customers beyond broadband’s reach continues to attract interest from determined companies. A total of 28 years after the launch of Teledesic, a second wave of satellite Internet companies has emerged, promising an appealing alternative to terrestrial Internet. By utilizing innovative constellations of small, low-cost, mass-produced satellites that significantly improve latency, these companies claim they will deliver cheap, high-speed Internet access almost anywhere in the world. While the field is crowded with industry fixtures like EchoStar and ViaSat, as well as new startups, three companies – O3b, OneWeb, and SpaceX – stand out for their ambition and disruptive industrial approach. These three companies have roots in Silicon Valley, but have also partnered with legacy players in the satellite sector, such as Airbus, Intelsat, and SES, to mitigate financial risks as well as benefit from their decades of experience. To succeed where Teledesic failed, these new companies have endeavored to cost-effectively construct and launch hundreds or thousands of small satellites as well as devise business models that make ‘connecting the unconnected’ profitable.

Founded in 2007, O3b has as of 2019 launched a network of 20 satellites in medium earth orbit (MEO), at an altitude of roughly 5000 miles above earth. As most Internet-providing satellites are in geosynchronous orbit at an altitude of over 22,000 miles, O3b’s arrangement substantially cuts the distance that data must travel and subsequently improves latency times to under 150 ms. O3b’s constellation has been utilized by clients ranging from the governments of Nauru and South Sudan to the US Department of Defense (Clark, 2017). In 2012, O3b was acquired by European satellite operator SES (Figure 1).

Figure 1.

Representation of low, medium and geosynchronous orbits (ESOA: EMEA Satellite Operators Association, n.d.).

O3b does not connect individual users directly to its satellite network. Instead, the company sells a satellite Internet gateway to a local ISP, which in turn sells subscriptions to users, in a process called IP (Internet Protocol) trunking (O3b Networks, 2012). The name comes from the metaphor of a tree with one trunk (the satellite link) and multiple branches (different ground-based clients/users). In South Sudan, for example, O3b has a partnership with the ISP RCS-Communication, which uses O3bTrunk to connect their wireless networks in the South Sudanese capital, Juba, to the Internet (Barton, 2016). Trunking allows the satellite operator and a local ISP to both benefit financially from the provision of Internet service to local customers, and distributes costs among both parties as well.

Like O3b, OneWeb and SpaceX are building constellations below geosynchronous orbit. Both plan to utilize LEO, which at only 750 miles high cuts latency even further to the 25 ms range. OneWeb intends to establish an initial constellation of 648 LEO K_u-band satellites by 2019 (Forrester, 2017). The network will offer global voice and data coverage, utilizing small, self-installed user terminals with solar and battery-power capability. This direct-to-user design contrasts with O3b’s trunking approach. OneWeb launched six satellites in February 2019 as building blocks of the constellation. The company has also partnered with Airbus to manufacture low-cost satellites at mass production facilities in Europe and the United States, with plans to eventually produce up to 15 satellites per week. Production began at a facility in Toulouse, France in June 2017.

SpaceX too is actively developing a range of satellite and space transportation projects. In 2015, the company announced plans to launch a LEO constellation comprised of 4425 small, low-cost satellites, more than double the total number of operational satellites in orbit at the time (Henry, 2017). The constellation ‘will operate in 83 orbital planes at altitudes ranging from 1,110 km to 1,325 km’ (Kharpal, 2017). The proposed LEO constellation will cut latency to as little as 25–35 ms, a speed comparable to terrestrial cable or fiber (Brodkin, 2018). Like OneWeb, SpaceX plans to develop in-house satellite manufacturing facilities – the company already builds and operates its own rocket systems – and will offer direct-to-user service. SpaceX aims to have its initial constellation operational by 2020.

These satellite Internet companies represent a range of approaches and models. O3b has quite literally taken the middle ground with its MEO constellation of twenty satellites, linked to ground-based partners through trunking. This less technically ambitious but economically safer approach has allowed the company to offer services to clients ahead of competitors, whose LEO constellations remain theoretical. If the OneWeb and SpaceX constellations can make the leap from concept to reality, they will dramatically change the global Internet equation. Swarms of low-cost satellites operating in LEO beaming high-speed connectivity directly to users may finally equip satellite operators to compete with, and possibly overtake, cable – or they may suffer the same Icarus-like fate as their predecessor Teledesic. The following section explores some of the technical and economic challenges facing the new satellite Internet companies, and assesses whether they can be resolved to benefit unconnected communities.

The technical and economic challenges of connecting the unconnected

Where satellite Internet once relied on high cost and latency-challenged geosynchronous satellites for connections, new services are based on non-synchronous constellations of low-cost satellites with higher speeds, more extensive services, and enhanced connectivity. These ‘swarms’ of smaller satellites emerging from the new generation of satellite operators have distinct advantages and drawbacks. Although satellites in LEO and MEO offer lower latencies since they are physically closer to the Earth’s surface, they cannot maintain constant contact with ground stations. To address this, companies plan to launch constellations comprised of dozens, hundreds, or even thousands of satellites to share the signal and ensure a satellite is always in range of the ground station. This necessitates more complex, and often more expensive, ground stations that can track and maintain contact with satellites in the constellation. In addition, building more satellites means more satellites must be launched into space – a potentially expensive proposition when launches can account for up to 30% of a satellite’s total costs (De Selding, 2015a). The introduction of thousands of new satellites into the already crowded orbital plane could also present serious challenges with regard to congestion, safety, and debris management.

Conditions in the satellite industry have changed significantly since the Teledesic era. One transformation has been the rise of private launch companies like SpaceX and Blue Origin. These companies have substantially cut clients’ launch costs through innovations including reusable rockets and capsules, streamlined in-house construction, and larger rockets with greater payload capacity (Hull, 2017). Airbus engineer Pierre-François Delval claims, ‘SpaceX changed the satellite industry with its five-ton launcher’ (P.-F. Delval, 2018, Interview with Pierre-François Delval, Head of Telecommunications Advanced Projects, Airbus Space and Defence). Not only do these new launch providers offer a greater range of options and capacity to satellite operators, they do so at a considerably lower price. For example, SpaceX advertises flights on its new Falcon Heavy rocket for US$90 million per launch, while the nearest competitor, United Launch Alliance’s Delta IV Heavy, charges US$300–US$500 million per launch and can only carry half of the weight (Malik, 2018). SES President and CEO Karim Michel Sabbagh explains, ‘As launch costs decrease, the size and complexity of satellites can decrease’ (K. M. Sabbagh, 2018, Interview with Karim Michel Sabbagh, President and CEO of SES Networks).

With decreased satellite size and complexity comes higher volume and lower cost manufacturing. Delval notes that Airbus is rapidly transitioning from a bespoke model to mass satellite production, with plans to build 1–2 small, LEO satellites per day at its new joint production facility in partnership with OneWeb. For comparison, production rates for larger geostationary satellites are usually 8–9 per year. In promotional materials, Airbus claims it will establish an ‘assembly line that will be completely different from classic geostationary satellite assembly labs and organise a procurement supply chain that will be unlike anything the space sector has ever seen before’ (Airbus, 2018).

OneWeb envisions an individual satellite cost of less than US$500,000 based on this mass production model (De Selding, 2016). For comparison, it is estimated that the latest generation of Global Positioning System (GPS) satellites cost US$500 million each to build (Leopold, 2013). The future could bring even further cost reductions for satellites. An Arizona State University research group recently claimed that their experimental ‘FemtoSats’ could cost as little as US$3000 to send into LEO (Hodgkins, 2016). However, these high production rates and low costs are still largely theoretical. While construction has started on the OneWeb–Airbus joint production facility in Exploration Park, Florida, it remains to be seen if the companies will be able to meet their ambitious goal of building several satellites per day.

Although new design and construction techniques have the potential to reduce costs, standard satellites remain expensive and time-consuming to build. Industry analyst Roger Rusch states, ‘large constellations are very inefficient. They’re cheaper, but you need 4,000 of them, so they need to be 1,000 times cheaper’ (Finley, 2015). Other analysts share this skepticism. While a theoretical model may predict low production costs, aerospace analyst William Ostrove believes, ‘it’s really difficult to test the concept without building out the network’, and unforeseen costs may thwart new satellite companies in the same way they did Teledesic (Finley, 2015).

One often overlooked technical challenge is the difficulty of matching satellite technology with receiver technology. A company may be able to build a low-cost, high-speed constellation, but its success also depends on the affordability and reliability of the ground-based infrastructure that consumers use to receive the signal. ‘The challenge with all [newer] constellations is the user terminal. This is the main reason why Teledesic wasn’t able to deploy’, explains OneWeb Systems Engineer Whitney Lohmeyer (W. Lohmeyer, 2018, Interview with Whitney Lohmeyer, Systems Engineer, Oneweb). For the new non-geosynchronous systems, the user terminal and antenna must rotate to pick up the signal from a passing satellite, making the terminal bulkier, and more complex and expensive to manufacture and purchase. As of mid-2018, a high-end maritime user terminal from Intellian, which operates on the Inmarsat constellation, costs as much as US$75,000, while a portable Broadband Global Area Network (BGAN) terminal from Hughes runs between US$2000 and US$3000. The cost of these units, which does not include Internet subscription fees, is well beyond the reach of most consumers, particularly in the developing world.

In publicity material, OneWeb touts ‘small, low-cost user terminals [that] will talk to the satellites in the sky, and emit 3G, LTE, 5G and WiFi to the surrounding areas’. Similarly, SpaceX’s Elon Musk has claimed that the user terminals for his company’s Starlink constellation will cost as little as US$100–US$300 (Boyle, 2018). If this crucial part of the satellite-to-ground equation can be successfully cracked, connecting the unconnected may become more viable. As it stands, however, receiver costs remain prohibitive for unconnected users in much of the world. To facilitate adoption, take-up subsidies must be offered and subscription plans must be scaled to economies in low-income, rural communities. Or, it will be necessary for higher service fees in post-industrial countries to help offset costs for users in the developing world.

Beyond the issue of technical practicality, then, is the question of whether or not these emergent satellite constellations are feasible from a business perspective. While ‘connecting the unconnected’ may be an appealing marketing rhetoric, there is little clear evidence that such a model can deliver profits to companies and investors. As Mark Rigolle, CEO of LeoSat and former CEO of O3b, asked at a 2018 industry forum, ‘is it a realistic business plan to connect people who make a few dollars a day?’ (M. Rigolle, 2018, Interview with Mark Rigolle, CEO of LeoSat).

Depending on the rubric used, the ‘unconnected’ can mean very different things. To the non-profit organization Internet Society (ISOC), the unconnected are the Agariyas ethnic minority in India, who live in the remote salt flats of the Thar desert with little to no communication infrastructure (Singh, 2018). To others in the satellite industry, however, the unconnected can be superyacht owners who want high-speed Internet during transoceanic voyages. The World Economic Fund identifies four factors as inhibiting access to the Internet: infrastructure; affordability; skills, awareness, and cultural acceptance; and local adoption and use (Luxton, 2016). By focusing only on the infrastructure part of this equation – utilizing new and innovative constellations to ensure their signals are available to anyone, anywhere, whether wealthy yacht owners or remote oil pipeline managers – satellite Internet companies will most likely not be able to make much progress in connecting the unconnected.

It is not surprising that US satellite companies face challenges adhering to a vision of supporting unconnected communities given that they are ultimately driven by profit. A 2015 article praising SES’s savvy investment in O3b remarks that while ‘O3b’s initial vision of providing Internet access to the “other 3 billion” people without terrestrial access has not been abandoned … the company does appear to have focused on markets with a more immediate payout’ (De Selding, 2015b). The article continues, ‘O3b by now is used to the jokes: O3b actually means the “other three billionaires” for luxury yachts, the “other three barrels” for offshore energy producers and the “other three battle groups” for military customers’ (De Selding, 2015b). At the same time, however, O3b has successfully brought satellite Internet to multiple developing countries in recent years, including Somalia, the Democratic Republic of the Congo, Papua New Guinea, and East Timor (Clark, 2018).

If satellite companies are able to strike a workable balance between profitable industry and military clients and low-income markets in developing countries, then a viable business model could exist for wide-scale satellite connectivity. The O3b contracts in the developing world mentioned above demonstrate this possibility. However, Somalia still has one of the lowest Internet penetration rates in the world according to the World Bank, despite the introduction of O3b’s connection (Fukui and Hirsi, 2018). Given this reality, it is much more likely that satellite Internet will continue to focus on niche high-value markets more than the developing world.

Conclusion

The voices extolling the virtues of the Internet and calling for greater global connectivity are loud and numerous. Manuel Castells (2014) writes that the Internet ‘is the decisive technology of the Information Age’; a means for communicating from the ‘many to the many’ that ‘increases sociability, civic engagement, and the intensity of family and friendship relationships, in all cultures’. The US State Department argues that ‘connectivity is as critical to economic development as other forms of infrastructure, like roads, ports, and electricity’ (US Interagency Steering Group, 2016). Promotional material from the World Economic Forum positions the Internet as an almost unlimited panacea for the challenges facing the developing world, stating that ‘broadband has the potential to be an enabler for reducing poverty, improving education, promoting gender equality, improving health services, ensuring environmental sustainability and providing a platform for global partnerships for development’ (Barnes, 2015). Following this premise (although these claims are certainly debatable), it would follow that Internet connectivity should be a global priority.

On the surface, it appears that the satellite industry is closer than ever to being able to create competitive satellite Internet services. Satellite manufacturing and launch costs have come down, innovative orbital configurations allow for speeds close to broadband, and advances are being made in ground station technology. However, serious concerns still exist regarding the technical viability and costs of these new constellations. Will these new systems be able to cope with ever-increasing demand placed on global Internet connections by streaming platforms like Netflix, which in 2018 accounted for 15% of all Internet bandwidth? (Spangler, 2018). What measures will be put in place to address the dangerous congestion that thousands of new satellites will add to the already crowded orbital plane? There is a need for further research on these issues.

Even if these technical hurdles can be overcome, our main question remains: who will reap the benefits of these systems? Since their introduction in the 1960s, communication satellites have been articulated with notions of global integration and development (Parks, 2005). While the rhetoric of ‘connecting the unconnected’ has been solidly embraced by the new crop of Silicon Valley satellite contenders, we are still left with more questions than answers as to whether satellite connectivity will ultimately benefit unserved and underserved communities. From our review of the industry, it seems clear the profit margins are not large enough to entice companies to connect the unconnected outside of a few well publicized efforts and community experiments.

If governments and consumers take seriously the benefits of connectivity, then they must hold satellite companies accountable for the promises they make. Future satellite connections for capital cities in the developing world could come with mandatory requirements for rural community VSAT networks, for example. With tech companies returning record profits and aerospace companies not far behind (Google parent company Alphabet reported US$33.7 billion in earnings for Q3 2018; Airbus reported US$17.5 billion) (Ausick, 2018; D’Onfro, 2018), it hardly seems unreasonable to advocate for subsidized connectivity for the same populations that tech and aerospace giants claim to want to support. Reframing Internet access as a basic human right, a view that is growing in popularity, it is also worth considering whether governments and community organizations, not the private sector, should be taking the lead in providing connectivity for their citizens. If we fail to examine the reality behind the lofty rhetoric of satellite Internet and other technological advances that claim to ‘change the world’, we fail to confront structural inequalities that allow companies to continue to profit from a series of broken promises.

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by a grant from the MIT-Skoltech Program.

ORCID iD

Matthew Graydon

References

Abbate

(1999) Inventing the Internet (Inside Technology). Cambridge, MA: MIT Press.

Airbus (2018) OneWeb satellites: connection for people all over the globe. Available at: http://www.airbus.com/space/telecommunications-satellites/oneweb-satellites-connection-for-people-all-over-the-globe.html (accessed 27 June 2018).

Ausick

(2018) Airbus Q3 earnings popped, but challenges remain. In: 247Wallst.com. Available at: https://247wallst.com/aerospace-defense/2018/10/31/airbus-q3-earnings-popped-but-challenges-remain/ (accessed 2 December 2018).

Barnes

(2015) How better connectivity can transform Africa’s economies. Available at: https://www.weforum.org/agenda/2015/06/how-better-connectivity-can-transform-africas-economies/ (accessed 30 June 2018).

Barton

(2016) RCS boosting capacity in South Sudan with O3b Networks. Available at: https://www.developingtelecoms.com/tech/satellite-networks/6746-rcs-boosting-capacity-in-south-sudan-with-o3b-networks.html (accessed 6 July 2018).

Bell

(2018) Changing faces: an overview of Europe’s broadband sector. Available at: https://blog.telegeography.com/changing-faces-an-overview-of-europes-broadband-sector (accessed 26 October 2018).

Ben David

(2015) Connecting the last billion. UCB/EECS-2015-233, 11 December. Berkeley, CA: University of California. Available at: http://www2.eecs.berkeley.edu/Pubs/TechRpts/2015/EECS-2015-233.pdf (accessed 1 December 2018).

Boyle

(2018) SpaceX set to launch first prototype Starlink satellites for global internet. Available at: https://www.geekwire.com/2018/spacex-gets-set-launch-first-prototype-starlink-satellites-global-internet-access/ (accessed 6 July 2018).

Brodkin

(2018) SpaceX hits two milestones in plan for low-latency satellite broadband. Available at: https://arstechnica.com/information-technology/2018/02/spacexs-satellite-broadband-nears-fcc-approval-and-first-test-launch/ (accessed 26 June 2018).

10.

Castells

(2014) The impact of the Internet on society: a global perspective. Available at: https://www.technologyreview.com/s/530566/the-impact-of-the-internet-on-society-a-global-perspective/ (accessed 30 June 2018).

11.

Clark

(2017) SES orders seven satellites from Boeing to help meet growing broadband demand. Available at: https://spaceflightnow.com/2017/09/12/ses-orders-seven-satellites-from-boeing-to-help-meet-growing-broadband-demand/ (accessed 18 October 2018).

12.

Clark

(2018) Four O3b satellites launched to beam Internet to developing world. Available at: https://spaceflightnow.com/2018/03/09/four-o3b-satellites-launched-to-beam-internet-to-developing-world/ (accessed 14 June 2018).

13.

Clarke

(1945) The space-station: its radio applications. In: Logsdon

(ed.) Exploring the Unknown: Selected Documents in the History of the U.S. Civil Space Program. Washington, DC: NASA, pp. 12–15.

14.

Collier

(2007) The Bottom Billion: Why the Poorest Countries Are Failing and What Can Be Done about It. Oxford; New York: Oxford University Press.

15.

Collins

(2018) A Telephone for the World: Iridium, Motorola, and the Making of a Global Age. Baltimore, MD: Johns Hopkins University Press.

16.

Cooper

(2017) Statement of Patricia Cooper Vice President, Satellite Government Affairs Space Exploration Technologies Corp. (SpaceX). United States Senate.

17.

De Selding

(2015a) Eutelsat does the math on reducing future satellite costs. Available at: https://spacenews.com/eutelsat-does-the-math-on-reducing-future-satellite-costs-by-at-least-20-percent/ (accessed 1 December 2018).

18.

De Selding

(2015b) Once-mocked 03b investment now force multiplier for SES. Available at: http://spacenews.com/2014-top-fixed-satellite-service-operators-once-mocked-03b-investment-now-force-multiplier-for-ses/ (accessed 14 June 2018).

19.

De Selding

(2016) One year after kickoff, OneWeb says its 700-satellite constellation is on schedule. Available at: http://spacenews.com/one-year-after-kickoff-oneweb-says-its-700-satellite-constellation-is-on-schedule/ (accessed 29 June 2018).

20.

D’Onfro

(2018) Alphabet Google earnings Q3 2018. Available at: https://www.cnbc.com/2018/10/25/alphabet-google-earnings-q3-2018.html (accessed 2 December 2018).

21.

Edwards

(1996) The Closed World: Computers and the Politics of Discourse in Cold War America (Inside Technology). Cambridge, MA: MIT Press.

22.

ESOA: EMEA Satellite Operators Association (n.d.) Satellite Orbits. Available at: https://www.esoa.net/technology/satellite-orbits.asp (accessed 6 July 2018).

23.

Evans

(1996) DirecPC dishes up a hasty link. The Washington Post, 18 March. Available at: https://www.washingtonpost.com/archive/business/1996/03/18/direcpc-dishes-up-a-hasty-link/982252b6-544b-4720-9c70-38784eeb4d62/?utm_term=.59567cccc915 (accessed 26 November 2018).

24.

Fairhurst

Caviglione

Collini-Nocker

(2008) FIRST: future Internet – a role for satellite technology. In: 2008 IEEE international workshop on satellite and space communications, Toulouse, 1–3 October 2008, pp. 160–164. New York: IEEE.

25.

Feder

(2000) Can Craig Mccaw keep his vision of teledesic from crashing? The New York Times, 4 June. Available at: https://www.nytimes.com/2000/06/04/business/can-craig-mccaw-keep-his-vision-of-teledesic-from-crashing.html (accessed 27 June 2018).

26.

Federal Communications Commission (2013) Measuring broadband America. Federal Communications Commission, February. Available at: https://www.fcc.gov/reports-research/reports/measuring-broadband-america/measuring-broadband-america-february-2013 (accessed 25 June 2018).

27.

Finley

(2015) Internet by satellite is a space race with no winners. Available at: https://www.wired.com/2015/06/elon-musk-space-x-satellite-internet/ (accessed 6 July 2018).

28.

Forrester

(2017) OneWeb plans May 2018 launch. Available at: https://advanced-television.com/2017/10/31/inside-satellite-oneweb-plans-may-2018-launch/ (accessed 31 January 2018).

29.

Freeman

(2018) Carlsbad’s Viasat brings ‘cyber cafe from space’ to rural towns in Mexico to bridge digital divide. San Diego Union Tribune, 12 April. Available at: http://www.sandiegouniontribune.com/business/technology/sd-fi-viasat-mexico-20180412-story.html (accessed 6 July 2018).

30.

Fukui

Hirsi

(2018) Improved internet access connects Somali students to each other and global knowledge. In: World Bank Blogs. Available at: https://blogs.worldbank.org/digital-development/improved-internet-access-connects-somali-students-each-other-and-global-knowledge (accessed 5 June 2019).

31.

Garcia

(2017) Inflight connectivity to hit the skies for two-thirds of airlines by 2020. Available at: https://www.tnooz.com/article/inflight-wifi-sita-onair/ (accessed 29 June 2018).

32.

Giridharadas

(2018) Winners Take All: The Elite Charade of Changing the World. New York: Alfred A. Knopf.

33.

Goodwins

(2002) Teledesic backs away from satellite push. Available at: https://www.zdnet.com/article/teledesic-backs-away-from-satellite-push/ (accessed 18 October 2018).

34.

Hansson

Mihailidis

Larsson

, et al. (2007) The rural wings project: bridging the digital divide with satellite-provided Internet. Phase I: identifying and analysing the learning needs of 31 communities in 10 countries. E-Learning and Digital Media 4(2): 107–115.

35.

Heeks

(2009) The ICT4D 2.0 Manifesto: Where Next for ICTs and International Development? Manchester: Institute for Development Policy and Management, University of Manchester.

36.

Henry

(2017) OneWeb breaks ground on a Florida factory that will build thousands of satellites. Available at: http://spacenews.com/oneweb-breaks-ground-on-a-florida-factory-that-will-build-thousands-of-satellites/ (accessed 6 July 2018).

37.

Hodgkins

(2016) Say hello to FemtoSats – the tiny new satellites that could democratize space. Available at: https://www.digitaltrends.com/cool-tech/femtosat/ (accessed 29 June 2018).

38.

Hull

(2017) Musk’s SpaceX doubles down on method for cheaper rocket launches. Available at: https://www.bloomberg.com/news/articles/2017-12-12/musk-s-spacex-doubles-down-on-method-for-cheaper-rocket-launches (accessed 1 December 2018).

39.

Hurst

(2018) Why satellite Internet is the new space race. PC Magazine, 30 July. Available at: https://www.pcmag.com/article/362695/why-satellite-internet-is-the-new-space-race (accessed 21 October 2018).

40.

International Telecommunications Union (2017) Key 2005-2017 ICT data for the world. International Telecommunications Union, July. Available at: http://www.itu.int/en/ITU-D/Statistics/Documents/statistics/2017/ITU_Key_2005-2017_ICT_data.xls (accessed 1 December 2018).

41.

Kharpal

(2017) SpaceX to start sending internet satellites into space in 2019. Available at: https://www.cnbc.com/2017/05/04/spacex-internet-satellites-elon-musk.html (accessed 18 October 2018).

42.

Leopold

(2013) DARPA seeks to bring satellite costs back down to Earth. Available at: https://defensesystems.com/articles/2013/12/13/darpa-space-access.aspx (accessed 26 November 2018).

43.

Luxton

(2016) 4 billion people still don’t have internet access. Here’s how to connect them. In: World Economic Forum. Available at: https://www.weforum.org/agenda/2016/05/4-billion-people-still-don-t-have-internet-access-here-s-how-to-connect-them/ (accessed 27 June 2018).

44.

Lyons

(2017) Internet at sea: 7 things you need to know. Available at: https://www.cruisecritic.com/articles.cfm?ID=1419 (accessed 29 June 2018).

45.

McLuhan

(1962) The Gutenberg Galaxy: The Making of Typographic Man. Toronto, ON, Canada: University of Toronto Press.

46.

Malik

(2018) SpaceX launched the world’s most powerful rocket. So, what’s next? Available at: https://www.space.com/39632-spacex-falcon-heavy-launch-whats-next.html (accessed 29 June 2018).

47.

Mann

(2012) Telstar 1: the little satellite that created the modern world 50 years ago. Available at: https://www.wired.com/2012/07/50th-anniversary-telstar-1/ (accessed 25 June 2018).

48.

Miller

(2017) Satellite Internet latency: what’s the big deal? In: Exede Satellite Internet. Available at: https://www.exede.com/blog/satellite-internet-latency-whats-big-deal/ (accessed 26 June 2018).

49.

NASA (2012) July 12, 1962: the day information went global. Available at: http://www.nasa.gov/topics/technology/features/telstar.html (accessed 25 June 2018).

50.

O3b Networks (2012) Where fiber fears to trench. Available at: http://www.telecomstalk.com/wp-content/uploads/2012/08/white-paper_where-fiber-fears-to-trench.pdf (accessed 7 July 2018).

51.

Parks

(2005) Cultures in Orbit: Satellites and the Televisual (Console-ing Passions). Durham, NC: Duke University Press.

52.

Parks

Starosielski

(eds) (2015) Signal Traffic: Critical Studies of Media Infrastructures (The Geopolitics of Information). Urbana, IL: University of Illinois Press.

53.

Patel

(2018) Making Internet accessible and affordable for unconnected populations. Available at: https://www.wirelessweek.com/blog/2018/03/making-internet-accessible-and-affordable-unconnected-populations (accessed 6 July 2018).

54.

Pew Research Center (2018) Demographics of Internet and home broadband usage in the United States. Pew Research Center, 5 February. Available at: http://www.pewinternet.org/fact-sheet/internet-broadband/ (accessed 25 October 2018).

55.

Pollard

(2018) The eccentric engineer: Herman Potocnik and his forgotten Space Station. Available at: https://eandt.theiet.org/content/articles/2018/01/the-eccentric-engineer-herman-potocnik-and-his-forgotten-space-station/ (accessed 25 June 2018).

56.

Pultarova

(2018) Mega-constellations: will they bridge the digital divide? Institution of Engineering and Technology, 22 January. Available at: https://eandt.theiet.org/content/articles/2018/01/mega-constellations-will-they-bridge-the-digital-divide/ (accessed 2 December 2018).

57.

Ragnedda

Muschert

(eds) (2013) The Digital Divide: The Internet and Social Inequality in International Perspective (First Issued in Paperback. Routledge Advances in Sociology 73). London; New York: Routledge.

58.

Regulation of Domestic Receive-Only Satellite Earth Stations (1979) 78–374, 74 FCC 2d 205.

59.

SatMagazine (2009) O3B Networks – year in review. Available at: http://www.satmagazine.com/story.php?number=971169404 (accessed 14 June 2018).

60.

Schwoch

(2009) Global TV: New Media and the Cold War, 1946-69. Urbana, IL: University of Illinois Press.

61.

Singh

(2018) Connecting the unconnected: the land of zero connect. In: Internet Society. Available at: https://www.internetsociety.org/blog/2018/05/connecting-the-unconnected-the-land-of-zero-connect/ (accessed 6 July 2018).

62.

Spangler

(2018) Netflix eats up 15% of all Internet downstream traffic worldwide (study). Variety, 2 October. Available at: https://variety.com/2018/digital/news/netflix-15-percent-internet-bandwidth-worldwide-study-1202963207/ (accessed 5 June 2019).

63.

Srinivasan

(2017) Whose Global Village? Rethinking How Technology Shapes Our World. New York: New York University Press.

64.

Statista (2016) U.S. household Internet subscription types 2017. Available at: https://www.statista.com/statistics/185602/broadband-and-dial-up-internet-connection-usage-in-the-us/ (accessed 1 February 2018).

65.

Stewart

(1999) Satellite superguide. Available at: https://www.zdnet.com/article/satellite-superguide/ (accessed 25 June 2018).

66.

The Arizona Republic (1996) Satellite net link like ‘opening up faucet’, user says. The Arizona Republic, 8 September. Available at: https://www.newspapers.com/clip/9934862/arizona_republic_direcpc_part_2/

67.

The New York Times (1979) F.C.C. Lifts rules on antennas in bid to spur satellite service. The New York Times, 19 October. Available at: https://www.nytimes.com/1979/10/19/archives/fcc-lifts-rules-on-antennas-in-bid-to-spur-satellite-service-risks.html (accessed 25 June 2018).

68.

Townsend

Sathiaseelan

Fairhurst

, et al. (2013) Enhanced broadband access as a solution to the social and economic problems of the rural digital divide. Local Economy 28(6): 580–595.

69.

Tweney

(2011) May 25, 1945: Sci-Fi author predicts future by inventing it. Available at: https://www.wired.com/2011/05/0525arthur-c-clarke-proposes-geostationary-satellites/ (accessed 25 June 2018).

70.

US Interagency Steering Group (2016) The global connect initiative: accelerating entrepreneurship and economic opportunity by expanding Internet access globally. U.S. Interagency Steering Group, December. Available at: https://staticshare.america.gov/uploads/2017/01/Final-version-of-the-GCI-Report-2016-01-115PM.pdf (accessed 30 June 2018).

71.

US Internet Users (2018). Available at: http://www.internetlivestats.com/internet-users/us/ (accessed 25 June 2018).

72.

Van Dijk

(2005) The Deepening Divide: Inequality in the Information Society. Thousand Oaks, CA: Sage.

73.

Via Satellite (2003) Teledesic relinquishes license. Available at: https://www.satellitetoday.com/uncategorized/2003/07/14/teledesic-relinquishes-license/ (accessed 27 June 2018).

74.

Wyler

(2016) Note from our chairman. Available at: http://oneweb.world/ (accessed 13 June 2018).