By Adriano Belisário 21 September 2018
Based on a review of the basic concepts regarding electromagnetic spectrum and its management, this article explores a few differences between the notions of “free spectrum” and “open spectrum” in order to understand some challenges related to freedom of expression in the 21st century. 
The spectrum, invisible
Today, it is difficult to imagine something more ubiquitous than the electromagnetic spectrum. It is literally everywhere. Not only in the space that surrounds us. The spectrum is not something outside our bodies, but a modulation that passes through us and in which we are literally immersed.
In practice, we know that its use for telecommunications is essential for many of society’s global and local flows. Without communication via the electromagnetic spectrum, we would have no mobile phones, GPS, Wi-Fi, radio, television, remote controls or Bluetooth – to mention just a few of the most well-known technologies. When you answer a call or access the internet on your mobile phone, watch a TV programme, listen to the radio or consult your position via GPS, you are sending information via the spectrum.
However, it is strange that this ubiquitous spectrum continues to be invisible in political discussions. While large corporations believe the electromagnetic spectrum to be a gold mine, closely monitoring and influencing public policies and development of the technological standards that make use of this resource, management of the spectrum continues to lack visibility for the rest of society. Unfortunately, the same applies to civil society organisations who work for the democratisation of communication, freedom of the press and “free internet".
Like other common resources, the electromagnetic spectrum has historically also been subject to gradual curtailment and private ownership. Currently, there are opportunities to change this scenario, but this will largely depend on a new understanding about what the spectrum is and how to use it, as well as significant participation by civil society in debates and decisions on the subject, as this is currently limited to a mere passive public consumer role. In this respect, this article explores the differences between the concepts of "open spectrum" and "free spectrum" in order to address some themes involving regulation of this resource and freedom of expression in the 21st century.
The return to radio
Considering the spectrum as a commons is not a new idea. The precursor of the radio, the wireless telegraph, was already considered as a peer-to-peer technology. In its early stages, the radio spectrum was a predominantly horizontal means of communication based on self-management. Radio was not a loudspeaker for the voice of the few, but a channel for communication between peers, as radio amateurs are well aware even nowadays. In other words, the common perception of radio as a device for passive reception owes less to technical limitations than to sociopolitical orientation.
In fact, some reports show that the transition to the model that we know has not been seamless. In Germany in the 1930s, Bertolt Brecht noted that radio was suffering a form of atrophy of its social function, when it became merely a receiver of the voice of a few transmitters. At the same time, in the United States, Lawrence Lessig pointed out that when the process of spectrum assignment began for commercial use, this met with massive and continued rejection of the new model by the population. With time, however, the eminently commercial approach to the radio spectrum was assimilated and naturalised. Even today, however, radio spectrum occupation through self-management for non-commercial purposes is preserved in practice for various stakeholders such as radio amateurs, free radio movements and some community radios, to mention but a few.
The state monopoly over the electromagnetic spectrum was essentially based on the technical problem of interference, that is, the deterioration of communication that usually occurs when transmitters simultaneously operate on the same frequency in nearby areas. There are various ways of dealing with this. Radio amateurs implemented simple and effective social protocols such as the "listen before talking" rule, today incorporated in technical protocols for wireless devices. The most commonly adopted radio frequency management solution in countries during the 20th century was the use of a monopoly of power to guarantee another monopoly, that of radio transmissions in their own territory.
The spectrum as a territory
Obviously, state monopoly over use of the spectrum guaranteed governments and lawmakers control over an extremely valuable asset in the 20th century. Here, the spectrum is seen as a vast territory to be colonised. During the past century, this ubiquitous “territory” was scanned, allocated and distributed by governments. These artificial borders simultaneously prevented the technical problems related to interference and guaranteed governments an extremely valuable political resource.
Within this paradigm forged throughout the 20th century, the state leased the spectrum to private economic groups offering services to consumers as if it were renting property. Today, there are at least two large groups of players disputing this territory (broadcast and telecom companies) or, in other words, operating within the dynamic of private ownership of the electromagnetic spectrum.
There are fundamental differences between these two groups, not only in terms of their objectives, but also of the type and use of spectrum access. Understanding the alliances and internal conflict dynamics relating to these economic groups is fundamental to the development of strategies for use of the spectrum as a common good.
Concessions were the first way the state found to manage the telecommunications spectrum. Basically, using arbitrary criteria, governments granted licences to business sectors close to them, in a process inseparable from the formation of so-called "public opinion" and the role of the "mass media" during the 20th century.
These private businesses that controlled radio and TV concessions during most of the last century played the leading role in the commercialisation of the electromagnetic spectrum – and expanded their empires by subleasing usage intervals of this resource to advertisers. In many countries, this first group can easily be identified in the local oligopolies made up of a few companies who hold a high concentration of regional TV and radio licences. The situation is now becoming even more serious in countries with debilitated or even non-existent public TVs and radios.
This oligopoly of a few businesses holding radio and TV concessions is the oldest player in disputes involving management of the electromagnetic spectrum. In many countries, their connections to the legislative political system are often explicit, involving local oligarchies of federal and political elites who directly or indirectly control the radio and television transmitters and re-transmitters in their states.
However, at the end of the last century, motivated by the popularisation of mobile telephones, a new group emerged that uses the spectrum for interpersonal communication, and no longer for broadcast transmissions, as on radio and television. In practice, telecoms operators lease the spectrum not to advertisers, but directly to persons and businesses who need it to send and receive information. This business model is based on a form of sub-concession of the spectrum, intended directly for users, controlled by microcredit between "clients" and operators.
In general, especially nowadays, telecoms operators access their slices of the spectrum through another government assignment mechanism: auctions. The criteria are no longer mainly political, but have become economic. Whoever pays, wins. And the astronomical costs of these negotiations have made these auctions into a game for the few.
Apart from their differences, the two models are based on the premise of exclusive use, where people and businesses create a relationship of private ownership with wavebands, like large landholders own their land.
In this case as well, we see major economic concentration within a few private groups. However, although in radio and television broadcasting we have historic experience of free or community TVs and radios, in cellular telecommunications, we have only a few recent records of experiences of independent use, with civil society playing a leading role in managing these services.
This model of exclusive use of the radio spectrum underwent a slight turnaround at the end of the 20th century. In spectrum bands that were considered to be of low value for telecommunications owing to their transmission characteristics, "unlicensed" use was allowed for medical, scientific and industrial purposes. Just a small gap in the regulations allowed people to have self-managed use of the spectrum, but it was enough to start an avalanche of well-known innovations.
This band was known under the label “ISM bands” (industrial, scientific and medical) and served as a basis for widely used technologies such as wireless telephony, Bluetooth, Wi-Fi networks, microwave ovens, hearing aids and many other devices. With variations around the world, ISM bands generally include frequencies around 2.4 GHz and 5 GHz.
We know that it is not necessary to request any authorisation from the government to connect to Wi-Fi or exchange photos with someone via Bluetooth. We also know that there is no risk of you accidentally reading someone else’s emails, just because you share a wireless network with that person. We know that various Wi-Fi networks can share the same frequency and that the devices that transmit on this frequency are "intelligent" enough to distinguish between the signals. In practice, people understand some of the principles and effects behind the new radio transmission technologies, "cognitive radio" or software-defined radio.
However, while Wi-Fi communications only use a shortwave band for small distances, previously limited by hardware and designated by government authorities, software-defined radio and new spectrum transmission technologies have broadened such possibilities as they offer new solutions for the technical problem of interference, which is the basis for the exclusive-use spectrum management model. Although convincing a century ago, the argument in favour of the "exclusive-use licence" as a regulatory model for the spectrum (either through concessions or auctions) is extremely questionable at this time.
Exemption from obtaining an “exclusive-use licence” for the use of certain wavebands is not the same as saying that the state is relaxing its regulations. However, it does involve a new regulatory control model which is no longer based on exclusive use and is moving towards accepting interference as a given factor in radio communications, instead of seeking to avoid it, and thereby creating artificial scarcity. In other words, interference should not be avoided through exclusive use, but controlled by coexistence techniques.
Put another way, the inability of various transmitters to use the same frequency in the same space is not an incontrovertible fact of nature, but the result of adoption of an exclusive-use spectrum model, which seeks to prevent interference in the least efficient and democratic way. In the case of the unlicensed spectrum, regulation has an impact on technical devices such as radio transceivers. When defining the basic operating specifications, the legislation guarantees that any device certified by the regulatory agency can operate on set wavebands without a licence being necessary. Some of these specifications determine the signal emission strength or set standards for digital data modulation.
The economic possibilities and new "business opportunities” created with this new model did not go unnoticed by liberal economists, techno-utopians and a certain "Californian ideology"-style current of thought regarding the spectrum.
In 1995, Eli Noam, a professor of economics and finance at the University of Columbia, predicted: “It won’t be long, historically speaking, before spectrum auctions may become technologically obsolete, economically inefficient, and legally unconstitutional.” A proponent of the open spectrum idea, Noam mocked the sale of ownership rights to the spectrum. “Could the state sell off the right to the colour red [….] [to keep] the different users from colliding into each other? [….] Imagine the state auctioning off, for perfectly good public policy reasons, the right to travel (in order to prevent overpopulation in Los Angeles), to print books (to protect forests)."
He therefore suggests that the "next step" be taken towards an "open market alternative" conceived of by researchers as "open spectrum". In his opinion, the auction policy tends to create an oligopolistic market structure; in other words, it would be more advantageous for the state and for the market to implement spectrum use concessions according to demand, so that prices are controlled algorithmically based on the level of availability of the desired bands.
Similarly, David Reed denounces naturalisation of the metaphor of the spectrum as a territory, where state distribution is necessary to avoid user collision (interference). Criticising this approach, Reed quotes diametrically opposed studies and research projects, demonstrating that the spectrum’s capacity can increase in line with the number of users if cooperative network models are adopted. George Gilder also claims that these radio transmission possibilities are foreseen in Claude Shannon’s information theory developed in the 1940s.
With respect to this new telecommunications regime, which does not interpret the spectrum on the basis of the scarcity paradigm, Reed states:
We don’t know today what the “best” cooperative architecture will be. [...] So any new regime must also encourage innovation in new architectures [...]. I think now is the time to look backwards to the early days of the Internet for inspiration. When the Internet Protocol (now called IPv4) was created, we did not know what the best technology would be for building networks. Today we are using technology that was never contemplated in the late 1970s […] and yet that protocol continues to be the core of the Internet […] We need a regime that allows RF networks to interoperate and cooperate in use of “spectrum” in an open and experimental way, just as the Internet did.
We suggest that the idea of free spectrum is related to the more radical concept of the open spectrum, in the sense that the rights to communication, information and freedom of expression supersede commercial interests. Access to the electromagnetic spectrum is a necessary prerequisite to the achievement of these rights; furthermore, its prohibition can no longer be based on interference as a natural impediment to coexistence, much less on the "tragedy of the commons" argument.  Spectrum sharing techniques based on cognitive radio and the possibilities created by radio-defined software (RDS) dismantle the principle of spectrum scarcity. Free spectrum asserts spectrum abundance and the possibility of coexistence in the spectrum, which could be initiated by defining the basic operating principles that make equitable spectrum occupation a possibility.
As we have seen, the idea of exercising freedom of communication and information via the electromagnetic spectrum is not exactly new. Although the free spectrum does not exist in regulatory terms, this does not mean that it is not being modernised in communication initiatives that freely occupy the spectrum, thus making spectrum exempt (locally, circumstantially) from use and occupation restrictions, concession procedures, as well as state and economic power. These are initiatives that validate Article 19 of the Universal Declaration of Human Rights and that greatly increase the ability to "speak", transmit and receive signals via the spectrum.
Commenting on the advocates of the "open spectrum" and "free spectrum" notions, deemed equivalent concepts, Lawrence Lessig states that although their goals are all different, they are united by "a belief that spectrum should be managed differently". Lessig points to a series of researchers – from Dave Hughes to Paul Baron, through David Reed and George Gilder – who are advocates of the idea that what is more probable than not is that the spectrum, used correctly, would in essence be unlimited.
Although they are indeed similar concepts, it is worth highlighting several possible nuances between the notions of "free spectrum" and "open spectrum", in the same vein as the classic discussion on "free software" and "open-source software". Take, for example, the use of the term “open spectrum” made by Aaron Swartz, which sounds closer to the "free spectrum" concept.
In a short and brilliant text published in 2003, Aaron Swartz used the "open spectrum" idea to suggest the creation of a free radio internet, a large mesh network that connects neighbouring points: "On the Internet, you don't need anyone's permission to talk, you just need an Internet connection. The same is true with this radio Internet, you just start sending your messages to your neighbors, and they pass them on.”
Swartz called on the FCC to stop any process related to privatisation of the spectrum for exclusive use, or in his words, to stop plans to "propertise" spectrum. In his opinion, it is, on the contrary, necessary to motivate research on efficient use and sharing of the electromagnetic spectrum. "We need to define the tools for a cooperative radio Internet. Just as Internet Protocol (IP) brought various networks together into the Internet, we need the same tools that will bring the various spectrum bands into a radio Internet.”
There is an obvious distance between Aaron’s proposal and the automated subconcession of the spectrum, against payment, developed by Eli Noam. Both come under the "open spectrum" umbrella. In other words, while the "open spectrum" concept can be used to extend the power of the few in the name of market freedom, the concept of "free spectrum" provides a direct route to considering the spectrum from the perspective of 20th century freedom of expression.
During the "Spectrum, Society and Communication" debate in 2013, computer scientist Silvio Rhatto emphasised the distinction between the "open spectrum" and "free spectrum" as follows:
Open spectrum as a technical concept makes provision for a cognitive radio, or a radio that is smart enough to find the best transmission frequencies, in order to identify other radios that are transmitting. This is a technical approach, but if you bring the open spectrum principle to the market limit, what we have is this "rational, efficient and automatic use of concessions".
Obviously, giants such as Google, Amazon, Facebook, Apple or Netflix are sympathetic to the idea of relaxing spectrum use rules in order to connect another billion people and increase their consumer base. The open spectrum debate therefore also includes discussion of an unlicensed market as an opportunity for businesses to reach new consumers ("connecting the unconnected").
This is clear: in the end, for the large telecoms and technology companies, opening up the spectrum with no exclusive-use licensing is also a big business opportunity. Instead of paying the state billions in auctions, such corporations could have the spectrum free of charge and thus in the name of a common good, in order to continue selling services to users. It is therefore not unusual to witness the active participation of “liberal” thinkers and large technology corporations in open spectrum debates.
In general, open spectrum advocates are often seeking convergence of all telecommunications into the current internet structures, without a due critical review of the risks, concentrations of power and controls related to the internet, in addition to underestimating the importance of local radios and TVs, seeing them as resources whose fate is to be engulfed by the internet. Concerned less with the emergence of new business models and more with communication as a fundamental right for everyone, free spectrum advocates perceive the spectrum as a common good that must be accessible to everyone, irrespective of economic class. Instead of underestimating local or community free radio/TV experience, the free spectrum movement finds the roots of its practical history herein, as well as in the free software movement. Instead of adopting an uncritical “digital inclusion” discourse, “free spectrum” could be a key notion to mobilise grassroots mobile phone (GSM) stations, digital radio, TVs, community-owned Wi-Fi infrastructure and others uses of the spectrum like high-frequency solutions for rural areas.  Spectrum should be viewed as a common resource available to non-profit initiatives, not only to connect us to the internet as we know it today.
In this vein, hacktivist Daniel Pádua issued a seminal challenge in the early years of Metareciclagem, a Brazilian network fairly active in the first decade of the century and focused on the appropriation of free technology. Pádua suggests the creation of a local wireless network, independent of the internet, “using second-hand Wi-Fi cards, repeater antennas made with Pringles chip cans" and recycled computers running free software to serve as access points in public schools and community associations.
"Creating a ‘garbage backbone’ using Wi-Fi at least connects you with your friends on the other side of the city. Because the point is not the internet but the creation of a local network," commented Pádua.
Like the internet, spectrum-sharing technologies have also arisen from military research. The United States military agency responsible for information technology solutions for the army and high-level government officials, the Defence Information Systems Agency (DISA), maintains a "centre of excellence" for spectrum management, known as the Defense Spectrum Organization (DSO), which is responsible for identifying, monitoring and evaluating emerging spectrum technologies that are seen as either a benefit or a threat to the US Department of Defense’s access to the electromagnetic spectrum.
As part of an agreement with the North Atlantic Treaty Organization (NATO), DISA created the Standard Spectrum Resource Format (SSRF),  a standard to facilitate the interoperability of different frequency technologies. SSRF was developed in collaboration with the Wireless Innovation Forum, an entity made up of government bodies, industrial associations and businesses such as Google, Nokia, Ericsson, Huawei, Qualcomm and others. Although it is not possible to forecast the long-term impact on wireless communications of prioritising military investment, it is quite clear that there is massive asymmetry of knowledge and power when it comes to the spectrum, which cuts civil society and its representatives off from crucial decisions on how the spectrum will be managed in the 21st century.
However, cooperative groups establishing infrastructures based on mesh networks, local networks, free radios, community mobile telephone operators, radio amateurs and other initiatives have already translated these ideas into practice, confronting the privatisation process and the artificial scarcity created by corporate and state oligopolies. This proves that it is not necessary to wait for constitutional consolidation of "free spectrum" proposals for it to become a reality. On the contrary, occupation of the spectrum by independent initiatives is also important in creating these new regulatory frameworks.
Thinking about "free spectrum" is therefore not a return to the early years of radio, when there was complete deregulation of all wavebands. This is not a matter of deregulating, but, in complementarity with the current system, building another regulation that acknowledges and preserves the spectrum as a common good. Instead of turning to the past, it is specifically a matter of looking to the future – which is already present and familiar to most people. Various widely distributed technologies only became possible with the release of certain frequencies for unlicensed use: people are already used to using them.
In the 21st century, freedom of expression is moving – increasingly – towards the right to the use of the electromagnetic spectrum for telecommunications. Controllers of the physical and logical infrastructure through which our information is trafficked concentrate power at an unprecedented scale – and we are still only beginning to understand the political and social impact of this. The spectrum is abundant and cheap. If used efficiently, it has the potential to break this concentration of power trend, in terms of the physical telecommunications infrastructure.
In this regard, it is important to guarantee bands for non-profit use in all bands of the telecommunications spectrum, implementing measures to guarantee that this space will not be dominated by businesses. This would allow for the emergence of more community-based mobile telephone operators, internet service providers and local networks managed by residents’ associations, free and community radios and TVs – or new protocols to come. A more democratic horizon in spectrum regulation would generate not only more innovation, but also a healthy decentralisation. Seeing the spectrum as a commons is a fundamental step in guaranteeing freedom of expression in the 21st century.
 This article draws on a number of the author’s previous individual and joint works, with updates and corrections, including “Communication, surveillance and infrastructure: techno-politics of the electromagnetic spectrum”, “Activism in Landscapes – Culture, Spectrum and Latin America”, and “Espectro Livre como alternativa tecnopolítica à vigilância”. Special thanks to Paulo Lara, Diego Vincentin, Thiago Novaes and Rafael Diniz for their past collaboration with this article.
 The argument in question presupposes that individuals who are acting independently and in accordance with their own interests produce conditions that are unfavourable to the preservation of common goods. The "tragedy of the commons” concept became popular on the basis of the article "The tragedy of the commons", originally published by ecologist Garrett Hardin in 1968.
 See, for example, Rhizomatica's HERMES project. For instance: https://www.rhizomatica.org/hermes/
 OpenSSRF is an implementation of SSRF in Java. SSRF is the North American correspondent of the NATO standard for automated data exchange, known as Spectrum Management Automated Data Exchange Format (SMADEF). SSRF is a mechanism through which the management systems of the United States and other NATO countries will exchange information.