Internet for Us: theory vs. practice in ‘decentralized’ networking

Introduction: material community infrastructures

Mesh networks are community based telecommunications networks made from broadcasting hardware that can be purchased by everyday people, all without relying on corporate actors for their connection. In their messaging, mesh networks aim to provide underserved areas with agency and equitable access to the internet by enabling the self-provision of information infrastructures. The social issues at the heart of the mesh network ethos revolve providing a decentralized alternative to corporate telecommunication services. This research summary investigates tangible mesh networking operations to assess how mesh communities achieve their goals of creating grassroots communications networks. 

In an effort to capture the human voices and stories surrounding mesh infrastructures, this research project contributes to the development of an investigative podcast around a series of recorded interviews with mesh networking organizers. This paper relies on recorded interviews obtained firsthand as a primary method of engagement with interlocutors, and informed consent was obtained from all research participants. This paper summarizes my research findings through interviews with organizers at NYC Mesh, and Professor Greta Byrum, the Co-Director of The New School’s Digital Equity Laboratory. All information presented in the following summary relies wholly on information from mesh network interlocutors, and is cited as such.

Together, interviews with NYC Mesh and Digital Equity Lab aim to understand mesh networks through the eyes of their creators. As such, interview questions were formulated with three primary goals in mind. First, to understand how mesh networks operate. Second, to probe how mesh networks are valued by their creators and users. And finally, to identify key challenges faced by mesh network communities. The following sections are thus scaffolded to answer each of these questions in order, each section providing essential information for the next. In total, we will address the material infrastructure, the ideals and ethos, and some critical challenges in mesh networks.

Mesh infrastructure: a web of distributed information

The key aspect of mesh network infrastructure that one should keep in mind is its decentralized nature. A “traditional” telecommunications network relies on massive cell towers for cellular internet broadcasting, and real networks of fiber-optic cables to deliver hardlined access to the internet. The material demands of traditional internet infrastructure provision thus favor a centralized telecommunications network, where costly broadcast infrastructures are owned and maintained by only a few powerful corporate or state actors. Conversely, mesh networks rely on fairly simple and affordable broadcast devices that both receive and send the network signal. In effect, the material demands of a mesh network distribute the power of infrastructure ownership across the network to its individual users. This section highlights two primary material aspects of the mesh network which provide for this distribution of power: the mesh network’s broadcast devices — or “nodes” — and the mesh network’s encoded traffic coordination operating system, called the “routing protocol”. Together, these two key materialities demonstrate how the mesh network reinforces an ethos of distributed power throughout communities, a topic which is continued in the next section.

First, we look at the broadcasting devices of mesh networks, the “nodes”. As opposed to traditional broadcast infrastructures which require a direct connection between the cell station and the internet device, mesh nodes are network routers which both receive and amplify the signal broadcast to the network. Because mesh nodes both receive and amplify the network signal, the network’s broadcast area can grow based on the physical distance between its connected nodes. The distribution of broadcast power throughout the mesh nodes creates a resilient and interconnected network, which passes the signal forward to other nearby nodes. To provide internet connection to its users, the mesh network requires only one hardline point of connection to the world wide web. In fact, even without a connection to the internet, mesh networks can be used to create “local” sites that are only accessible from a mesh-connected device. MOn top of all that, mesh nodes are relatively low-cost devices. To put this in concrete terms, the NYC Mesh website’s “Installs” page estimates the setup cost for a single household’s connection between “$160 and $290, depending on the equipment used and your ability to pay” (NYC Mesh, 2019). Barring any device malfunctions, mesh network connection is theoretically free after setting up the network’s nodes; NYC Mesh suggests a $20 monthly donation to help support organizational costs of regular network maintenance. 

If the broadcast “nodes” exemplify the physically distributed aspect of mesh network infrastructure, then the “routing protocol” demonstrates the decentralized operating rules of the network. Professor Byrum at Digital Equity Lab describes a mesh routing protocol as the “way a network directs traffic around its nodes … the key fact about a dynamic routing protocol is that it’s automated to find the most efficient route between two nodes … if you lose a node … the dynamic routing protocol is going to find a way around it … it has a kind of self-healing quality” (interview with G. Byrum). To rephrase Professor Byrum’s description, the routing protocol makes sure that information passes through the network taking the most efficient route possible. TIn addition, the routing protocol is dynamic, and allows the network to easily take in new nodes and lose malfunctioning nodes, all without major disruption to the stability of the larger network. The routing protocol thus further ensures the decentralized quality of mesh broadcasting, by dynamically monitoring the network conditions and distributing network traffic evenly across all nodes in the network. 

Together, the nodes and the routing protocol comprise two key materialities of mesh networks. These materialities physically demand that the mesh network distributes power evenly amongst all users within a telecommunications network. In other words, mesh network infrastructures are inscribed to reproduce and materialize an ethos of decentralized community engagement around local needs. However, there are times that this ethos is challenged through the real world demands of providing a mesh network connection. Now that we have reviewed two essential material aspects of mesh network infrastructures, the next section investigates the ideals embedded within the physical infrastructure of mesh networks in greater detail. 

Mesh ethos: decentralized community-building

Brian Hall has become an integral member of NYC Mesh over the course of his five year engagement with the organization, making him one of NYC Mesh’s longest continuous volunteers. When asked what excites him about NYC Mesh, he said that the organization promotes “the idea that people can build infrastructure. [Before mesh networks] It wasn’t really an idea in this country … realizing that communities can get together and build infrastructure” (interview with B. Hall). This sentiment was echoed by other volunteers at the NYC Mesh monthly meeting in November 2019. Internet access is almost a necessity to urban life in the United States, and yet there are still urban families who are not connected to the internet. Mesh networks aim to deliver internet access to people who are unable to connect through dominant forms of telecommunications services. When I first came to the topic of mesh networks, I had presumed that mesh was merely a crafty way of bypassing corporate internet subscription fees. Through speaking with the creators of mesh networks, and studying the networks’ materialities, I have found that mesh networks — in their volunteer-led organizations and decentralized infrastructures — are inscribed to produce an ethos of decentralized community engagement around local needs. This section focuses on this ideal, which is materialized in the provision, use and maintenance of mesh network infrastructures.

The “traditional” telecommunications industry relies on centralized broadcast technologies that are big, expensive, and controlled by large corporate actors. Professor Byrum of Digital Equity Lab suggested that centralized corporate media overpower local broadcasting efforts, and remarked that “local relationships of mutual aid are really important in any disaster or emergency, and that’s at odds with the way our media ecosystem has consolidated over time. What we see in our media landscape over time is the loss of diversity and local autonomy in terms of what’s available in local broadcasting.” (Byrum). As mentioned in the section covering mesh infrastructure, mesh networks create local sites that can only be accessed through mesh-connected devices. This is ideal for a community messaging board, where neighbors can broadcast localized news. Like Professor Byrum emphasized, this is especially important in times of local emergency when it is necessary to filter out all of the noise from the outside world. Whether they are used to advertise for a community block party, or to find missing persons after a hurricane, mesh networks’ local broadcasting abilities promote an ethos of local community engagement. 

Mesh networks advance an ideal of local self-determination, in that they rely on fairly accessible broadcast technologies, and strong community organizing power. As reviewed above, a single mesh network is generative to local information broadcasting. In addition to providing local access to media, the devices used to provide mesh networks reinforce local autonomy over infrastructure provision. Mesh network nodes require occasional maintenance for various reasons, and since the mesh networks I looked at are volunteer-run, a key aspect of their network provision involves building local expertise in maintaining mesh network infrastructures. Zach Giles, a volunteer computer programmer and organizer at NYC Mesh, spoke to me about how the lack of “corporate assurance” in their mesh network provision actually helps reinforce practical skills in their community of volunteers. He said, “part of the goodness of it is that if it breaks, you get to go out and fix it. You get to figure out why [it broke] … There are people in your community that you go contact … So that negative is actually a positive. It’s actually a goal of the community” (interview with Z. Giles). In making organized community action a necessary aspect of their service, NYC Mesh transforms their invisible network connections into a tangible community. This decentralized nature of mesh network provision allows for mesh communities to create resilient networks that expand using sustainable, but sporadic — as volunteer resources fluctuate — means.

Mesh challenges: strength in disorganization

Both organizations engaged in this research — NYC Mesh and Digital Equity Laboratory — rely on grassroots efforts to support their mesh networks. In the two preceding sections, we have reviewed how mesh networks are structured around a distributed model, both in their technical infrastructures, and in their organizational forms. Although mesh networks strive to distribute community power through their decentralized provision of equitable internet access, there are several challenges that pose setbacks to mesh network organizations today.          Summarized, these challenges lie in issues of knowledge transfer and representation, and maintaining operational strength as a volunteer-based organization. This section reviews some of the challenges faced by mesh network communities, as reported by NYC Mesh volunteers, and Greta Byrum from Digital Equity Laboratory. 

As covered above, mesh networks aim to strengthen local ties, and promote the self-determination of communities. “In its ideal form,” Greta Byrum remarked, “a mesh network is a great way to get people learning about internet technology, get local folks involved in actually planning and building and maintaining a network” (Byrum). Of course, as often happens when ideals are enacted in the real world, this process of involving local community members can be difficult to ensure. In the introductory briefing speech that started the night’s event, volunteer Brian Hall highlighted a recent issue that exposed a siloing effect happening within the NYC Mesh organization. The sudden shutdown of a mesh node by the FCC demanded a sudden need for volunteer support. Hall reported that he was one of the only people providing technical support, and said “We've got to decentralize and delegate tasks, especially for when something happens like this FCC thing. We need to spread out tasks so everyone isn’t overloaded” (Hall).

To resolve this issue of an overworked technical force, the NYC Mesh team discussed the possibility of hiring an employee dedicated to providing technical support and servicing the mesh network. However, this suggestion was met with consternation from volunteers who want to see NYC Mesh remain a community network of unpaid supporters. Later, in a different working session in the meeting’s schedule, Zach Giles urged everyone to “remember the mission and the goal” — evidently citing the organization’s principles — “it’s not internet, it’s a community network that happens to have internet … If it’s a wireless ISP [Internet Service Provider], or if it’s any kind of commercial thing, then why bother even doing it?” (Giles). Giles’ comment addresses a core tension within the larger mesh network movement. It is a difficult feat to provide a massive telecommunications network that is both consistent in its service provision, and decentralized in its organizational form. Mesh network users rely on their network’s consistency, so when the network breaks down, expert volunteers are called to commit their labor in a timely manner. 

When it comes to mesh network ethos versus implementation, the issue of knowledge transfer to local stakeholders can challenge mesh network efforts to build self-sufficient mesh communities. Greta Byrum drew on her prior experience with the Detroit Community Technology Project to express the importance of building local expertise in mesh network maintenance. When the Project first started installing mesh networks in Detroit neighborhoods in 2011, communities without technical training were left vulnerable when the technologies they depended on for internet malfunctioned. In response, they built the Digital Stewardship Program which “trains neighborhood leaders in the technical and community organizing skills necessary to design, build, and maintain community wireless networks” (Digital Stewards Program). Byrum said that the “knowledge transfer” aspect of mesh network provision is key to “creating that sense of community ownership in the sense of that ideal of a mesh network” (Byrum). Restated, local expertise is essential to the sustainability of decentralized mesh networks and their communities. With this in mind, we can see that hiring employees would be antithetical to the philosophy of mesh network communities. As such, local representation in a mesh community contributes directly to the ability of mesh networks to persist with their ethos of decentralization intact. 

Beyond the challenge of maintaining organizational strength with decentralized volunteer efforts, the mesh network movement is confronted with the issue of how to measure the broader social impact of mesh networks in their embedded communities. Part of Digital Equity Lab’s role in the mesh network movement is to develop data-tracking measurements of success for mesh networks. In their philosophy — beyond providing an affordable connection to the internet — mesh networks are a world making project that aim to reinforce decentralized community engagement around local issues. When asked to specify the metrics of community success, Professor Byrum said, “the kinds of outcomes that people really want to see tend to be in areas of education, employment, and public health … [and] indicators that relate to community power … So that’s where we’re focusing right now: this kind of in-between area that has to do with agency and a sense of belonging and connectedness, and a sense of what’s possible in the community” (Byrum). According to Byrum, it will take several years to collect data on, and measure the broader social outcomes of mesh networks. Until then, the issue of measuring success will be left up to debate for mesh network volunteers.

Conclusion: 

In the initial research summary laid out above, I have given a basic description of how mesh network infrastructures work, alongside an ethnographic analysis of the people and philosophies that guide mesh network organizations. In the final section, I outlined some of the challenges faced by mesh network communities when their ethos and infrastructures break down. I pointed to research by Digital Equity Lab to show that, moving forward, it will be important to define measurements of success for mesh networks. In total, this preliminary research has highlighted focus areas for continuing research. I have identified unresolved tensions between the ideal and actual practices of mesh network communities. Specifically, between the decentralized ethos of mesh network provision, and the centralized nature of relying on skilled experts for mesh network consistency. Because of this tension, in my continuing research on mesh network communities, I hope to investigate how mesh networks empower their creators and users to assert their agencies over a collective space. In this investigation, I will gather data on who is using mesh networks, and for what. Most important to this continuing research, is the question of whether mesh networks do in fact distribute power and autonomy equitably to all users across the network, or rather, if they merely rearrange and reproduce existing power structures.

This initial research project on mesh networks is part of an endeavor to uncover how mesh networks are deployed in New York City, and across the world. This investigation probes into how mesh networks enact political ideals through their integration with the sociotechnical infrastructures that facilitate modern human life. Continuing research asks: what forms of human relationships do mesh network technologies reward? How might mesh network technologies reproduce and reshape forms of human relationships, and what do those normative forms illustrate about the “intended users” of mesh networks in New York City, and in the world? What is the claim to space being made by the mesh network movement, and how is it different from dominant forms of economic and social organization? This project serves as an installment to this larger investigation on the effects of mesh network technologies in organizing alternative forms of human life in a modern world.

Interviews cited:

Fossum, O. (2019, Nov. 26). Video interview with G. Byrum.

Fossum, O. (2019, Nov. 12). Personal interview with Z. Giles

Fossum, O. (2019, Nov. 12). Personal interview with B. Hall

Works cited:

Appadurai, A., “Introduction: commodities and the politics of value.” In Appadurai, A., Ethnohistory Workshop, & Symposium on the Relationship between Commodities and Culture. (1986). “The social life of things: Commodities in cultural perspective.” Cambridge University Press. 

Akrich, M. "The De-scription of Technical Objects." In Bijker, W. E., & Law, J. (1992). “Shaping technology/building society: Studies in sociotechnical change. Cambridge, Mass: MIT Press.

Çalışkan, K. & Callon, M. (2009) Economization, part 1: shifting attention from the economy towards processes of economization. Economy and Society, 38:3, 369-398, DOI: 10.1080/03085140903020580

Çalışkan, K. & Callon, M. (2010) Economization, part 2: a research programme for the study of markets. Economy and Society, 39:1, 1-32, DOI: 10.1080/03085140903424519

Callon, M. (1998). Actor-Network Theory-The Market Test. The Sociological Review, 46(S), 181-195. doi:10.1111/1467-954x.46.s.10

Callon, M. & Law, J. (1995). Agency and the hybrid collectif. The South Atlantic Quarterly, 94(2), 481508.

Deleuze, G. & Guattari, F. (1994). Nomadology: The War Machine. A.K. Press.

Digital Stewards Program. (n.d.). Retrieved from https://detroitcommunitytech.org/eii/ds.

Foucault, M. (1995). Discipline and punish: the birth of the prison. 2nd Vintage Books ed. New York: Vintage Books.

Gunn, W., Otto, T., & Smith, R. C. (2013). Design anthropology: Theory and practice.

Latour, B. (1996). “On actor-network theory: A few clarifications” Soziale Welt, 47. Jahrg., H. 4, pp. 369-381

Latour, B. (2017). On Actor-Network Theory. A Few Clarifications, Plus More Than a Few Complications. Philosophical Literary Journal Logos, 27(1), 173-197. doi:10.22394/0869-5377-2017-1-173-197

Mumford, L. (1964). Authoritarian and Democratic Technics. Technology and Culture, 5(1), 1. doi:10.2307/3101118

NYC Mesh. (2019, September 24). Retrieved from https://www.nycmesh.net/.

Thatcher, J., O’Sullivan, D., & Mahmoudi, D. (2016). Data colonialism through accumulation by dispossession: New metaphors for daily data. Environment and Planning D: Society and Space, 34(6), 990–1006. https://doi.org/10.1177/0263775816633195 

Winner, L. (1980). Do Artifacts Have Politics? Daedalus Vol. 109 No 1. Modern Technology: Problem or Opportunity? 121-136.