The United States and Government-Provided Internet Access: Infrastructures of Free Expression and the Role of the State

Enrique Armijo


Government-provided access to high-speed Internet in the United States is on the rise. This is due in large part, of course, to a smartphone-driven explosion in the demand for mobile wireless services. Smartphone ownership increased from 16 percent of Americans in 2009 to 56 percent in 2012 (White House OST and NEC? 2013). Among the young, the numbers are even more pronounced. In 2017, more than nine in ten Americans between the ages of 18 and 29 owned smartphones (Smith 2017). This pervasiveness and the concomitant need for all of those devices and their users to be online, on-demand, at any time, has moved high-speed Internet access in the U.S. from a luxury to an expectation. And governments across the United States are stepping in not only to offer speeds that private cellular-based ISPs do not yet provide, and in places their networks do not yet reach, but also to offer alternatives to private networks and access providers (Cooper 2016).

More than 450 U.S. cities of all sizes have moved to try to meet this demand (Community Networks 2015). These offerings are taking a range of forms. One is a public utility model, i.e., purely government owned-and-operated “municipal broadband” networks. Other cities are adopting public-private partnership models; under one typical model, a private ISP provides free Internet access to the public in a particular public space such as a city park, town hall, or transportation hub in cooperation with a municipality or its administrative subsidiary (Flegenheimer 2013). Similarly, public utility companies have been building their own wholesale-only fiber-and-wireless-based broadband networks and making those networks available on an open access basis to ISPs, which can then use city-owned Internet infrastructure to offer in-home high-speed access to their subscribers for a fee; here, the government keeps the transport layer of the infrastructure for itself and offers the service layer to private ISP companies (Bode 2017).

Where city-wide networks have been cost-prohibitive or otherwise difficult to establish, municipalities have focused their efforts on aggregating smaller service areas within their city limits on their own or with corporate partners, usually via franchise agreements in the case of the latter ( n.d.; MERAKI n.d.; Hinz 2012). Current infrastructure thus makes it possible for a visitor to New York City to enjoy free high-speed Wi-Fi Internet access on their iPad or Android Phone from Battery Park up to Central Park Zoo, across the East River, and back down to Prospect Park in Brooklyn—all without accessing their private carrier network (NYC Parks 2017).1 On the other coast, San Francisco has begun serious consideration of a city-wide high-speed fiber-optic Internet network, including via a “public model” whereby the city would both manage construction of the network and serve as the ISP (City and County of San Francisco 2016). And this revolution is by no means limited to large cities like New York or San Francisco. Cities and counties of all sizes have undertaken or are considering municipal broadband projects or public-private partnerships to expand online access for their citizens (Office of Erie County 2015). Additionally, on the federal level, on March 21, 2017, the U.S. House Energy and Commerce’s Subcommittee on Communications and Technology held a hearing on how to expand mobile broadband in rural America (U.S. House of Representatives Energy & Commerce Committee 2017).2

Underlying, and likely spurred by, this rise is an ongoing debate as to the proper role of governments in providing wired Internet access to their citizens. The entrenched incumbency of existing ISPs has had debilitating effects on start-up efforts in the market for broadband. If Google could not make Google Fiber work, it is difficult to conceive of any potential new entrant devising a business model that would permit it to even survive as a going concern, let alone compete against the incumbents on speed and price (Crawford 2017). Hence advocates call for a more aggressive government role, from the aforementioned building fiber networks and leasing access to ISPs on a wholesale basis to offering an Internet service itself.

Additionally, just like in the private market, the wireless service must at some point connect to wired infrastructure. The development of fiber networks and the capacity for wireless access are thus closely correlated. Greater wired capacity is a necessary predicate to greater wireless capacity. So once governments become involved in offering fiber-to-the-home, customers will demand mobile connectivity from those and other providers on the same terms (Institute for Local Self-Reliance 2007, 21). More ISPs and more networks lead to greater mobile connectivity.

The most important question with respect to this issue is how this “fundamental makeover” of public places from exclusively physical spaces to mixed spaces with both physical and online aspects, including the move of speech spaces from fixed spaces to mobile ones, is “alter[ing| the nature, character, and democratic functions of public places and public expression” (Zick 2007, 1—5).3 The Access to Knowledge framework asks questions related to what Jack Balkin calls the “infrastructure of free expression,” in particular whether “government policies promote the delivery of information and knowledge” (Kaminski and Prakash 2015, 9, quoting

Balkin 2007). Additionally, in the United States context, there is the First Amendment to the Constitution, which bars governments from interfering with speech on the basis of the speech’s content or viewpoint—even when it is doing so for what it claims to be public-interest-related reasons. In all three models of government-provided Internet access set out above, a state actor is playing a significant role with respect to the infrastructure of the network and the traffic to be carried over it. When the traffic being carried is speech, the Access to Knowledge framework is directly implicated. And the record of these governments so far, at least with respect to the conditions they place on the speech traffic that is carried over their networks, is decidedly mixed.

If one’s focus zooms out from the infrastructure and technology involved, Internet access networks are speech spaces—as Stacey Schesser (2006) says, a “new domain for public speech” (6). When the government is involved in their development or operation, these spaces are provided either in name or in fact by the state; to use the traditional parlance of First Amendment jurisprudence, these networks are publicly owned property over which citizen expression travels. There is no question that governments have a critical role to play in helping to bridge the digital divide, stepping in where privately provided service is uneconomical or lower quality due to near-monopoly power of private ISPs, particularly with respect to the “last mile” inefficiencies associated with privately provided Internet access and the high-cost, data-capped business model of mobile access in the United States. However, due to the pace of change, the U.S. runs the risk of accepting government-provided digital speech spaces unreservedly as part of our communications infrastructure without asking the fundamental questions of what law applies, what is permissible, and what the U.S. Constitution bars. Societies must not rush to embrace new technology and, in the process, leave behind all of the longstanding protections from government interference that are well established in every other context. It would be ironic, indeed tragic, if the 21st-century speaker loses the protections that were unquestionably enjoyed by the speaker who was communicating in government-provided speech spaces long before the iPhone was even invented.

Terms of service as speech rules

Comparing examples from two different points in time might illustrate the danger referred to immediately above. In the 1965 U.S. Supreme Court case of Lamont v. Postmaster General, a federal statute empowered the Postmaster General to confiscate foreign-originated mail that he deemed to be “Communist propaganda.” The Lamont petitioner, a pamphleteer who received notice of the Post Office’s confiscation of his copy of the Peking Review, sought to enjoin the statute’s enforcement, arguing that it violated hrs First Amendment right to receive information. Even though the statute provided that an addressee could request mail that the General had confiscated by returning an official reply card, the Supreme Court unanimously agreed with Lamont that the statute violated his First Amendment rights. The Court noted that “the United States may give up the post-office when it sees fit, but while it carries it on the use of the mails is almost as much a part of free speech as the right to use our tongues.”

The principle affirmed in Lamont—that the government must carry the speech of its users without prejudice against the content of that speech— extends back to the founding of the postal service. Though the U.S. Supreme Court strayed from this foundational understanding for a time, finding that the government did have authority to make content-related decisions as to the mail it carried,5 the provenance of the principle is not in dispute. As Ithiel de Sola Pool recounts in his seminal Technologies of Freedom:

In 1836, President Andrew Jackson, concerned at the upsetting effect that the distribution of antislavery propaganda might have in the South, urged Congress to ban such matter from the mails. An ad hoc committee of the Senate chaired by [John] Calhoun, America’s classic defender of slavery, rejected the President’s request because, in hrs view, Congress had no such power: “If it be admitted that Congress has the right to discriminate ... what papers shall or what shall not be transmitted by the mail, it would subject the freedom of the press, on all subjects, political, moral, and religious, completely to its will and pleasure.”

(de Sola Pool 1984, 82)

Compare the statute found unconstitutional in Lamont, as well as the historical basis underlying its invalidation, to the following proscriptions, all of which come directly from municipalities’ terms of service for use of their publicly accessible broadband networks and all of which use blatantly content-based restrictions that would facially violate the United States Constitution’s First Amendment in any other context:

  • • The “Acceptable Use Policy” for the municipal utility-owned and operated Chattanooga, Tennessee fiber-optic network bars users from using the network to “transmit, distribute, or store material ... that is,” in addition to illegal or obscene, “threatening, abusive or hateful,” or that offends “the privacy, publicity or other personal rights of others.” Nor may users of the network “post messages” on third-party blogs “that are excessive and/or intended to annoy or harass others”—“regardless of [the] policies” of the blogs on which the users post (Electric Power Board of Chattanooga 2017) 6
  • • The Terms and Conditions for GOWEX, the private partner offering Internet access for New York City as part of city’s Wireless Corridor Challenge, “bars the transmission of data ... via Hotspots managed by GOWEX ... whose content is threatening, derogatory, obscene, pornographic, or the transmission of any other type of material which constitutes or incites a conduct which may be considered a criminal offense, is prohibited.” GOWEX also “reserves the right to prevent or block access to any user” who violates the content policy (GOWEX n.d.).

• The city of Miami’s terms of use for its Miami Beach Wi-Fi network requires users to waive any claims against the city based on service disruptions:

[Y]our access to the Service is completely at the discretion of the City, and your access to the Service may be blocked, suspended, or terminated at any time, at the sole discretion of the City, without cause or for any reason including, but not limited to, any violation of this Agreement, actions that may lead to liability for the City, disruption of access to other Users or networks, and violation of applicable laws or regulations... ., and may he interrupted, refused, limited, or curtailed at any time.

(City of Miami Beach n.d.)

Similarly, the city of Raleigh, North Carolina states that “under no circumstances shall the City be liable for any damages that arise or result in any way from use of, or inability to use, the service to or access to the Internet or any part thereof”

  • (City of Raleigh n.d.)
  • • The City ofWilson, North Carolina’s Terms of Service for its Greenlight network states that users may not “send[|, post[], or host[] harassing [or] abusive” materials, or may not “engagfe] in any activities or actions intended to withhold or cloak any user’s identity or contact information.” Greenlight also may, in its “sole discretion,” “temporarily or permanently remove content” that it believes violates its Terms of Use. Furthermore, Greenlight may, pending its investigation of a violation, “suspend the account or accounts involved and/or remove or block material that potentially violates this policy” (Greenlight Community Broadband 2017).

As should be clear, these networks, likely along with dozens if not hundreds of others, severely restrict users’ speech on the network in exchange for access. First Amendment doctrine in the United States makes clear that outright bans on protected speech—even indecent speech, let alone “excessive,” “derogatory,” “abusive,” or “hateful” speech—are never sufficiently narrowly tailored to survive constitutional scrutiny.7 It is also black-letter free speech law that implements prior restraints—and there is no question that a network operator’s “rejecting or removing” material because of its content before that material reaches its intended recipient is a prior restraint, as is the case here—are presumed unconstitutional.8 And the right to speak anonymously is well enshrined in the Speech Clause’s protections as well.9 Terms of service such as those used in Chattanooga, Miami, Wilson, and potentially scores of other cities thus violate basic tenets of First Amendment law.

By defining what speech can and cannot be transmitted over the network and by setting out the grounds by which the state can refuse a user access, these terms define the contours of their users’ First Amendment rights (Grimmelmann 2015, 49-75). 0 In Lamont, the Supreme Court declared that once the government decides to provide a mail service, it cannot discriminate in its delivery of that mail because of its objection to the content that users of the service seek to deliver. Yet that is exactly what is happening on dozens, or perhaps hundreds, of municipal broadband networks.

Government-provided internet access as free expression infrastructure

These restrictions on free expression are, to use another principle from the Access to Knowledge framework, a particularly pernicious mix of “old school” and “new school” speech regulation (Kaminski and Prakash 2015, 2, quoting Balkin 2007). They are “old school” because they use the power of the state to censor, based on what the state deems as harmful speech. It is Chattanooga’s determination, not the speaker or listener’s, as to which speech is, to quote from the Terms of Service for their network, “threatening, abusive or hateful,” or that offends “the privacy, publicity or other personal rights of others,” or “excessive and/or intended to annoy or harass others” (Electric Power 2017). They are “new school” because the consequences of the state’s censorship take place ex ante to the speech being restricted rather than ex post.

As is common with state restrictions on information and communications technology (ICT) and unlike most interferences with speech in physical space, the tools for interference with online speech are embedded within the ICT, and the network operator has absolute control. Deleting an en route text or tweet for its failure to comply with a state-imposed term of service can go unnoticed by the speaker in a way that sending a police officer to shut down a protest in the park cannot. This also creates an intractable problem for judicial review, since the speaker may never know that the state has interfered with their speech for a content-based reason. As Seth Kreimer (2006) argues, in such a context “neither speakers nor listeners will know that the message has not been conveyed, and there is no way to determine how dialogue has been deformed” (16-27).

The state as first-order threat to freedom of expression

Given the characteristics of our current networked environment, the focus in Access to Knowledge-related scholarship on the acts and decisions of private intermediaries makes eminent sense. As I have noted in prior work, “through ICT, we have overcome temporal and spatial limitations on communication, but we need constant intermediation in order to do so” (Armijo 2015, 421; Armijo 2014). That trade-off, which was initially realized with the printing press and broadcast but which has become fully consummated with mobile telephony and ubiquitous Internet access, has come at a cost to speakers. And much of that cost is paid in the form of private intermediaries’ control over expression. But First Amendment scholarship should never lose sight of the fact that governments represent the most serious challenges to citizens seeking to communicate online.

Through both direct censorship of user content through design, governance, network management-related decisions (Rosen 2012, 1525), and what Laura DeNardis calls “private industry mediation of delegated censorship” (DeNardis 2015, 16)—i.e., Internet service providers and application and website owners’ enforcement of government content removal requests—the private telecommunications industry plays a critical and often negative role in establishing a freedom of expression-enabling infrastructure online. To the extent online expression is both a mediated and monitored experience, it is certainly so that private companies are doing much, if not most, of the mediating and the monitoring. But it is important to not lose sight of free expression first principles: it is governments, not private parties, which have historically represented the first-order threat to free speech and access to knowledge.

Another comparison across time, this one borrowed from prominent Internet critic Evgeny Morozov, might help to illustrate the point. As noted in Morozov’s The Net Delusion, expression’s move from physical space to online has been a boon for speakers, but it has also made it much easier to be a spy. In the 1970s, it literally took teams of KGB officers to drill holes for bugs, monitor workplaces, establish observation points below and above a single dissident’s apartment, and listen to every conversation in order to collect speech and associations that the government believed to be incriminating. That is a lot of work to surveil one person. Now, however, a single officer of the modern-day secret police can run keyword searches of millions of intercepted emails by thousands of users without leaving his desk (Morozov 2011, 150-51). The overwhelming majority of those email accounts are provided by private companies and the emails themselves mostly travel over private networks, but private actors do not replace government interferences with privacy and association; rather, they facilitate it—so too with interferences with speech.

It is thus certainly the case that because of the mobile economy, the monitoring and mediating of speech has become cheaper, faster, and better, and the privatized nature of Internet infrastructure has enabled this to happen. But privatization can be a brake as well as a throttle. To be able to accomplish its desired results, the government must use its leverage over a private ISP to embed the functionality the state desires into its network, in much the same way the U.S.’s National Security Agency is alleged to have leveraged AT&T and other ISPs’ cooperation as part of its FAIRVIEW program, whereby AT&T installed NSA surveillance equipment in at least 17 of its Internet hubs on U.S. soil (Angwin and Larson 2015; Ball 2013). Despite that notable example, however, companies can and increasingly do, especially post-Snowden, tell the government no (Rozenshtein 2018). If a government is providing an Internet service itself, however, there is no jawboning necessary. The state can design its own communications infrastructure in ways that compel users to waive their rights to be free from content-based interferences with their speech or from warrantless searches. As virtual public speech spaces emerge that are public as well as private, the risks of harm to access to knowledge multiply. And interferences with a virtual speech space can be much more efficient when the government cuts out altogether the middleman private ISP or network owner.11 Accordingly, measures such as network neutrality, which seek to limit the interference private ISPs have over users’ speech, are only a partial solution; meaningful limits must also be placed on governments’ ability to do the same over their own networks.

In addition, the content-based terms of service discussed in the section above, though troubling, are just a small tool in the state’s arsenal for ex ante interferences with mobile ICT-intermediated speech. Preemptive interferences with mobile users’ online speech over private networks, from the blocking of websites and applications to shutting down networks altogether, are now common throughout the world. The civil society group AccessNow, as part of its #KeepItOn initiative, documented 56 Internet shutdowns by national governments in 2016—more than three times than the 15 that it documented the year prior (AccessNow 2017). Additionally, to claim these dangers are real only in countries with traditions of despotic regimes with historical enmity toward free expression in other contexts is dangerously false. The most comprehensive study of government interferences with users’ connections to digital networks found that of the 566 instances of interference involving 101 countries between 1995 and 2010, 51 percent occurred in authoritarian regimes, but 39 percent occurred in democracies (Howard, Agarwal, and Hussain 2011, 222).

And even though government’s attempts to enlist technology companies in the fight against extremist speech online have been longstanding (Yadron 2016), in 2015, the US President Donald J. Trump made hrs views on the issue explicitly known, stating while a candidate that in response to the spread of radical extremism online, “we have to talk [to the executives of private technology companies like Microsoft and Google] about, maybe in certain areas, closing that Internet up in some way” (Statat 2015). Even though ICT at the national level in the United States is highly privatized, it is very likely that the President has the power in the event of a national security emergency to shut down wire communications throughout the United States. The U.S. Communications Act states that the “war powers of President” include the ability to, “if he deems it necessary in the interest of the national security and defense, ... cause the closing of any facility or station for wire communication.”12 So if President Trump had decided that he needed to turn off the Internet in the United States, he would not have needed Bill Gates’s help—or, perhaps, even Congress’s consent.

Again, private mobile technology companies are part of a highly regulated market, so government leverage over them will always exist. But when a government is administering its own networks, there is no need to exercise leverage at all. Even if a nationwide public Wi-Fi network in the United States may not be politically or economically feasible, individual municipalities and states’ entry in the high-speed Internet market represents opportunities for great benefits but also great harms. We should temper our optimism over these developments with a healthy dose of caution, if not distrust.

Preserving an infrastructure of free expression

As argued above, promoting and preserving an infrastructure of freedom of expression remains a project of protection from governments as well as protections granted by them. As has long been the case when it comes to protecting speech more generally, civil society will continue to play an important role. For example, in collaboration with Tor, CitizenLab, and the international digital-rights advocacy group Access Now, The Open Observatory of Network Interference (OONI) has developed ooniprobe, a mobile application with the capability to test whether a network is censoring or blocking a mobile user’s access to a particular URL (Waddell 2017). Internet-wide blackouts, rather than targets of specific URLs, are more difficult to demonstrate, but civil society is developing those tools as well.

In addition, the same private entity gatekeepers that have exercised significant control over online speech have also taken significant steps to protect it. In 2017, America’s largest private ISPs successfully lobbied for legislative repeal of FCC (Federal Communications Commission) rules restricting the companies’ use of their users’ browsing information (Neidig 2017). But other private entities have taken more proactive approaches. Coogle’s research and advocacy arm Jigsaw recently launched a suite of cybersecurity tools called “Protect Your Election,” intended to help protect news sites and human rights groups from DDoS and other security-based attacks against elections-related speech by governments and their hackers (Newman 2017). And private intermediaries such as Twitter and Google commonly resist government requests for their users’ information in the interests of free expression (Otarola 2017) and provide comprehensive information to the public concerning those requests (Twitter 2017; Rozenshtein 2018). As more of our discussions and accessing of information move to mobile platforms and technologies, these protections will play an increasingly important role.


I end with a thought experiment. Imagine a proposal for a nationwide high-speed wireless network in the United States, built by the federal government, with Internet access service provided to users’ mobile phones over the network by local utilities, pursuant to terms of use adopted by those utilities and subject to whatever strings the feds would impose in exchange for network access. Imagine that this network was proposed in 2014, around the time the U.S. Court of Appeals for the D.C. Circuit struck down the FCC’s first attempt at adopting net neutrality rules for private ISPs consistent with the policies of the Obama administration. Would you have been in favor? Would you have found the network consistent with, and promotive of, principles of free expression, innovation, and democratic discourse?

Now imagine the same project being proposed in 2018, with an anti-net neutrality, anti-media, and arguably anti-free speech President Trump in office. If you were in favor in 2014, would you be in support now? Do you still trust the government to provide nondiscriminatory access to the Internet, the way we have always presumed it offers access to other public goods like roads, water, sewers, and sidewalks? Or is mobile Internet infrastructure, where information is the public good on offer, somehow different? Would you consider the mobile device in your bag or pocket a speech engine or a tool the government could use to censor and surveil you?

It is certainly so that creating an infrastructure of free expression involves positive policymaking—affirmative acts by governments intended to fuel the production of knowledge and innovation and the interaction between diverse, and at times conflicting, ideas, cultures, and beliefs. Preserving and promoting Access to Knowledge, however, particularly in the ICT context, must also take into primary account the individual liberty-based principles that protect speakers and listeners from government interferences with speech. This constitutional norm may have been bom with the First Amendment and intended for pamphleteers instead of bloggers. But as speech moves from physical to virtual space and from parks and streets to mobile phones, the United States, as well as rest of the world, needs to recognize and affirm those principles now more than ever.


  • 1 These networks are often actually hodge-podged together, because of the challenges associated with city-wide municipal fiber networks in larger cities attributable to private incumbents’ established infrastructure footprints (Burlington 2016).
  • 2 A Bill discussed at the hearing proposed requiring the Department of Transportation to install needed broadband conduit underneath hard surfaces as part of any federally funded highway construction project.
  • 3 Whether offering free high-speed Internet access is a good economic decision for cities is beyond the scope of this chapter. Some have argued that it is not, particularly since financing the construction of such a network is usually accomplished through debt funding via issuance of municipal bonds (Reason Magazine 2013), while others disagree (National Economic Council 2015).
  • 4 Lamont v. Postmaster General, 381 U.S. 301 (1965).
  • 5 Milwaukee Social Dem. Pub. Co. v. Burleson, 255 U.S. 407 (1921).
  • 6 To be sure, such restrictions on Internet use and access would likely be more permissible in other countries, where governments have a freer rein with respect to barring hate speech or harassment. See Article 19, 1992, striking a balance: hate speech, freedom of expression, and non-discrimination (Sandra Coliver, cd.), at https://www.articlel9.Org/data/fiIes/p dfs/publications/striking-a-balance.pdf. But even in countries that tolerate greater restrictions on hate speech or harassing expression than the United States, the danger remains that those restrictions “become overbroad and are subject to abuse.” Article 19, 2015, hate speech explained: a toolkit , ‘Hate-Speech’-Explained—A-Toolkit-%28201.5-Edition%29.pdf.
  • 7 Sable Commc'ns, of Cal. v. FCC, 492 U.S. 115, 126-128 (1989) (upholding ban on obscene telephone messages but finding ban on indecent messages not narrowly tailored).
  • 8 See Promotions, Ltd. v. Conrad, 420 U.S. 546, 553 (1975) (holding a city’s denial of use of its theater for a production of Hair on the grounds the play was “not in the best interests of the community” was an invalid prior restraint).
  • 9 McIntyre v. Ohio Elections Conini’n, 514 U.S. 334 (1995) (finding that a statute prohibiting anonymous political literature violated the First Amendment, on the ground that anonymity encourages open discourse and protects speakers from retaliation).
  • 10 As Grimmehnann (2015, 42) writes, “owners occupy a privileged position because their control over [Internet] infrastructure gives them unappealable control over the community’s software-based rules,” and “can use their power over the infrastructure layer to make policy at the content layer, for good and for ill."
  • 11 Under a mixed wholesale/retail model, leverage could of course work in a speechpromoting way; if a government builds its own network, it is free to insist on neutrality from any ISP seeking to provide retail service over it. But if many governments themselves feel free and able to discriminate against user traffic on the basis of content on their own networks, it is difficult to imagine that most governments would insist on equal treatment of traffic from their partner private ISPs.
  • 12 47 U.S.C. § 606. See d/jOSENATE COMM. ON HOMELAND SEC. & GOV'T AFFAIRS, PROTECTING CYBERSPACE AS A NATIONAL ASSET ACT OF 2010, S. REP. NO. 111-368, at 10 (2010) (“The Committee understands that Section 706 gives the President the authority to take over wire communications in the United States and, if the President so chooses, shut a network down.”). The Communications Act defines “wire communication” as “transmission of writing, signs, signals, pictures, and sounds of all kinds by aid of wire, cable, or other like connection between the points of origin and reception of transmission, including all instrumentalities, facilities, apparatus, and services (among other things, the receipt, forwarding, and delivery of communications) incidental to such transmission. 47 U.S.C. § 153 (51).


AccessNow. 2017. “#KecpItOn.” Accessed August 11, 2017. kcepiton/ #problem.

Angwin, Julia, and Jeff Larson. 2015. “NSA Spying Relies on AT&T’s ‘Extreme Willingness to Help.” ProPublica. August 15. tts-cxtrcme-willingness-to-help.

Armijo, Enrique. 2014. “Kill Switches, Forum Doctrine, and the First Amendment's Digital Future.” Cardozo Arts and Entertainment Law Journal 32: 411—469.

Armijo, Enrique. 2015. “Government-Provided Internet Access: Terms of Service as Speech Rules.” Fordham Urban Law Journal 41: 1499-1525.

Article 19. 2015. “Hate Speech Explained: A Toolkit.” Accessed August 11, 2017. https:// /‘Hate-Speech ’-Explained—A-Toolkit-% 282015-Edition%29.pdf.

Article 19. 1992. Striking a Balance: Hate Speech, Freedom of Expression, and Non-discrimination (Sandra Coliver, ed.). Accessed August 11, 2017. https://www.articlel9.Org/data/files/p dfs/publications/striking-a-balance.pdf.

Ball, James. 2013. “NSA’s Prism Surveillance Program: How It Works and What It Can Do.” The Guardian, June 8. https://www.theguardian.coin/world/2013/jun/08/nsa-prism-ser ver-collcction-facebook-google.

Bode, Karl. 2017. “Huntsville, Alabama is Suddenly Awash in Broadband Competition, Showing Why Comcast is so Afraid of Municipal Broadband.” TechDirt. February 16. https://www. nd-competition-showing-why-comcast-is-so-afraid-municipal-roadband.shtml?mc_cid=d790 17bcf3&mc_eid=2910c7b52b.

Burlington, Ingrid. 2016. Networks of New York. Brooklyn and London: Melville House.

City and County of San Francisco Board of Supervisors. 2016. “Financial Analysis of Options for Municipal Fiber Optic Network for Citywide Internet Access.” http://sfbos. org/sites/default/files/FileCenter/Documents/55357-FACl .pdf.

City of Miami Beach, n.d. “WiFi Miami Beach—Network Terms and Conditions” (on file with author).

City of Raleigh, North Carolina, n.d. “Downtown Raleigh Free WiFi Access Terms and Conditions” (on file with author).

Community Networks. 2015. “Community Network Map.” Last modified October 2015. Accessed August 11, 2017.

Cooper, Charles. 2016. “Why Municipal Broadband.” The Urbanist, June 2. https://www.

Crawford, Susan. 2017. “Google Fiber Was Doomed From the Start.” Backchannel, March

14. cid= d76683f62f#.rxuus56sz.

de Sola Pool, Ithiel. 1984. Technologies of Freedom. Cambridge and London: Harvard University Press.

DeNardis, Laura. 2015. “The Privatization of Free Expression.” In Access to Knowledge: Global Censorship: Shifting Modes, Persisting Paradigms, edited by Pranesh Prakash, Nagla Rizk & Carlos Alfonso Souza. Yale Law School Access to Knowledge for Development Center, American University at Cairo, & Instituto de Technologia & Sociedade do Rio.

Electric Power Board of Chattanooga. 2017. “Acceptable Use Policy.” Accessed August 11, 2017. pdf

Flegenhcimer, Matt. 2013. “Wi-Fi and Cellphone Service on Subway Trains? M.T.A. Says It May Happen.” New York Times, September 17. 18/nyregion/mta-plans-wi-fi-and-phone-service-on-subway-trains.html ?partner=rss&em c=rss&smid=tw-nytimes&_r=O.

GOWEX. n.d. Terms and Conditions for the New York City Wireless Corridor Challenge (on file with author).

Greenlight Community Broadband. 2017. “Terms of Service.” Accessed August 11, 2017.

Grimmclmann, James. 2015. “The Virtues of Moderation.” Yale Journal of Law & Technology 17 (1): 42-109.

Hinz, Greg. 2012. “City Unveils Plan for Free Wi-Fi, Wider Super-fast Internet.” Crain’s Chicago Business, September 24. BLOGS02/120929936/city-unveils-plan-for-free-wi-fi-wider-super-fast-intemet.

Howard, Philip N.. Sheetal D. Agarwal, and Muzammil M. Hussain. 2011. “When Do States Disconnect Their Digital Networks? Regime Responses to the Political Uses of Social Media.” Communication Review 14 (3): 216-232.

Institute for Local Self-Reliance. 2007. “Localizing the Internet: Five Ways Public Ownership Solves the U.S. Broadband Problem.” files/downloads/report-daggctt.pdf.

Kaminski, Margot, and Pranesh Prakash. 2015. “Introductory Framework.” In Access to Knowledge: Global Censorship: Shifting Modes, Persisting Paradigms, edited by Pranesh Prakash, Nagla Rizk & Carlos Affonso Souza, 9. Yale Law School Access to Knowledge for Development Center, American University at Cairo, & Instituto de Technologia & Sociedade do Rio. n.d. “Wifi." Accessed August 11, 2017.

MERAKI. n.d. “City of Newton.” Accessed August 11, 2017.

Krcimer, Seth F. 2006. “Censorship by Proxy: The First Amendment, Internet Intermediaries, and the Problem of the Weakest Link.” University of Pennsylvania Law Review 155: 11-101.

Morozov, Evgeny. 2011. The Net Delusion: The Dark Side of Internet Freedom. New York: PublicAffairs.

National Economic Council. 2015. Community-Based Broadband Solutions: The Benefits of Competition and Choice for Community Development and High-speed Internet Access, https:// ort_by_executive_oifice_of_ the_president.pdf.

Ncidig, Harper. 2017. “AT&T, Comcast Defend Repeal of Internet Privacy Rules.” The Hill, March 31. 1-of-intemet-privacy-rules.

Newman, Lily Hay. 2017. “A Cybersecurity Arsenal That’ll Help ‘Protect Your Election’.” Wired, March 7. tect-election/?mbid=social_twitter.

NYC Parks. 2017. “Wi-Fi in Parks.” Accessed August 11, 2017.

Office of Erie County Legislator Patrick Burke. 2015. “Municipal Broadband Project.” ort-Erie-County-20150316.pdf. Accessed August 11, 2017.

Otarola, Miguel. 2017. “Police Get Search Warrant for Everyone who Googled Edina Resident’s Name.” StarTribune, March 18. nt-issued-to-edina-police-raises-privacy-concems-of-intemet-users/416442113/.

Reason Magazine Policy Study. 2013. “Lessons in Broadband from Lafayette, Louisiana.” (Policy Study 424, Nov.),

Rosen, Jeffrey. 2012. “The Deciders: The Future of Privacy and Free Speech in the Age of Google and Facebook.” Fordham Law Review 80: 1525—1538.

Rozenshtein, Alan Z. 2018. “Surveillance Intermediaries.” Stanford Law Review 70: 99—189.

Schesser, Stacey D. 2006. “A New Domain for Public Speech: Opening Public Spaces Online.” California Law Review 94: 1791-1825.

Senate Comm, on Homeland Sec. & Gov’t Affairs. 2010. Protecting Cyberspace as a National Asset Act of 2010, S. Rep. No. 111-368, at 10.

Smith, Aaron. 2017. “Record Shares of Americans Now Own Smartphones, Have Home Broadband.” Pew Research Center, January 12. 2017/01/12/evolution-of-technology/.

Statat, Nick. 2015. “Donald Trump Thinks He Can Call Bill Gates to ‘Close Up’ the Internet.” Verge, December 7. http://www.theverge.eom/2015/12/7/9869308/dona Id-trump-close-up-the-internet-bill-gates.

Twitter. 2017. “Transparency Report." Accessed August 11, 2017. https://transparency.

U.S. House of Representatives Energy & Commerce Committee. 2017. “Broadband: Deploying America’s 211st Century Infrastructure.” March 21. hearings-and-votes/hearings/broadband-deploying-americas-21st-century-infrastructure.

Waddell, Kaveh. 2017. “Mapping Countries that Censor the Internet.” The Atlantic, February

9. ing/516051/.

White House Office of Science and Technology Policy (White House OST) and The National Economic Council (NEC).2O13. Four Years of Broadband Growth, https:// broadband_report_fmal.pdf.

Yadron, Danny. 2016. “Revealed: White House Seeks to Enlist Silicon Valley to ‘Disrupt Radicalization’.” The Guardian, January 7. 2016/jan/07/white-house-social-media-terrorism-meeting-facebook-apple-youtube-.

Zick, Timothy. 2007. “Clouds, Cameras, and Computers: The First Amendment and Networked Public Places.” Florida Law Review 59: 1-69.

< Prev   CONTENTS   Source   Next >