The papers in this First Monday special issue were originally part of the final symposium of the research project ADAM (Architecture Distribuée & Applications Multimédias, French acronym of Distributed Architectures and Multimedia Applications). This introduction elaborates on the issues at stake in the study of distributed network architectures beyond engineering sciences; it explores their political, social and legal implications and shows, by introducing the papers, the important challenges distributed architectures pose to our societies, economic systems, legal norms and collective behaviors.
Distributed architectures: Today and six years ago
Four areas of (cross-cutting) reflection on distributed architectures
Concluding remarks — Beyond ADAM: Present and (near-)future research on distributed architectures
The interdisciplinary research program ADAM (Distributed Architectures and Multimedia Applications, adam.hypotheses.org), conducted from 2010 to 2014 and funded by the French National Agency for Research (Agence Nationale de la Recherche, ANR), has explored the technical, political, social, socio-cultural and legal implications of distributed network architectures (see a general presentation of the research program in Méadel and Musiani, 2015). While actually covering, in practice, a large variety of technical arrangements and topologies, this term broadly indicates a type of network bearing several features: a network made of multiple computing units, seeking to achieve its objectives by sharing resources and tasks, able to tolerate the failure of individual nodes and thus not subjected to single points of failure, and able to scale flexibly. Beyond this simplified operational definition, the choice, by developers and engineers of Internet-based services, to develop these architectures instead of more centralized models has several implications for the daily use of online services and for the rights of Internet users.
The final symposium of the ADAM project, open to disciplines as varied as science and technology studies, information and communication sciences, economics, law and network engineering, took place at the end of 2014, and addressed these implications in terms of a central issue. With the increasingly evident centralization of the Internet and the surveillance excesses it appears to foster, what are the place and the role of the (re-)decentralization of networks’ technical architectures? What is the place for user autonomy and empowerment at a time when infringements upon privacy and pervasive surveillance practices are often embedded in network architectures? Are distribution and decentralization of network architectures the ways, as Philippe Aigrain (2010) suggested, to “reclaim” Internet services — instruments of ‘technical governance’ able to reconnect with the original organization of cyberspace?
Papers presented at the symposium, of which this special issue presents a “revised and improved” selection, have explored decentralized network architectures in political, social, and legal terms as well as technical. While this exploration has drawn from different methodological and epistemological perspectives, this introduction points out some of the ‘transversal’ dynamics and issues they highlight. Before getting on to that, though, we wish to briefly touch upon the genesis of our project — that has finally led to the symposium — what motivated us to investigate distributed architectures through socio-economic, political and legal lenses, and how we have seen these objects evolve within the frame of the project and beyond.
Distributed architectures: Today and six years ago
The choice to become interested in distributed architectures from legal, economic, social and communicational standpoints, in addition to the technical perspective, was far from being self-evident in 2010, when we first sought funding for our research project. Indeed, the hour of glory of peer-to-peer software was arguably passed, as laws in several countries were targeting it as a technology and streaming was taking its place as the prime means of consuming content. Furthermore, the client-server logic was increasingly taking hold as the privileged strategy to foster the development of the Web, and to answer the needs of billions of Internet users. And yet, discreetly but creatively, myriad projects aiming at developing Internet services on more decentralized and distributed architectures than the Googles and Facebooks of our times were, in parallel, seeing the light.
These network architectures, all too often reduced to the partial vision one could have via the prism of “pirate practices”, seemed to us bearers of much more than the piracy vs. sharing opposition. We considered these projects a very interesting “laboratory” where visions and implementations of alternative Internets were taking place — experiments of “governance by architecture” imbued with different, innovative ways of distributing authority, power, value, sense of community. It is based on this core hypothesis that the ADAM project has unfolded until 2014, unveiling a very diverse ecosystem of practices, technologies, organizations, norms and users. This issue, building on contributions by members of the ADAM project team and by external authors who contributed to the final symposium, is a testament to this diversity.
To our understanding, this initial vision we had of distributed architectures is still accurate, and as of 2016, we can add a few further remarks to it. First, the (re-)distribution and the (re-)decentralization of networks are strictly linked — much more so than a few years ago, and in particular after the Snowden revelations — to the discussions of surveillance and privacy issues, and find themselves frequently associated to discussions about encryption. Also thanks to the role that local and distributed technologies, relying on networks such as TOR, have played in rallying and organizing social movements and grassroots resistance tactics, distributed architectures are increasingly seen as technologies of empowerment and liberation. Yet, they do not escape a powerful double narrative, fueled by previous ones depicting peer-to-peer as an allegedly ‘empowering-yet-illegal’ technology. If on one hand, the discourse on empowerment and better protection of fundamental civil liberties is very strong, several projects seeking to merge decentralization and encryption to improve protection against surveillance show in parallel a desire, more likely a need, to defend themselves from the ‘you’re used by terrorists’–type allegations (Sanger and Perlroth, 2015). This dialectic is taking place in the broader context of discussions about governance by infrastructure and civil liberties (Musiani, et al., 2015), some of them particularly related to encryption (or the breaking of it), such as the Apple vs. FBI case and WhatsApp proposing, since April 2016, encryption by default (Ermoshina, et al., 2016).
Second, we must cite the extraordinary success of the ‘last’ — in chronological terms — distributed architecture: blockchain technologies. The success of the blockchain is perhaps still greater in research projects and anticipations of intermediary-less futures than in fully working applications, and we can recognize, as in previous “expectation cycles” related to distributed architectures and a number of other technologies, both the potential and the hype. Yet, the blockchain and its myriad variants have some specificities that it is and will be very interesting to observe in the coming years; in particular, it seems to be the first decentralized networking technology to be widely accepted — “celebrated” would be more accurate — by national and supra-national institutions, even as its first widespread application, Bitcoin, was born with the explicitly stated goal of making each and every institution obsolete, and its birth, development, functioning are subject to several controversies.
Four areas of (cross-cutting) reflection on distributed architectures
The issue fleshes out several questions and areas of interest for present and future research on distributed architectures, and for their “Internet-reclaiming” potential. This section delves into four of them in detail, highlighting the more specific contributions the different articles make to each.
1) The importance of being historical ...
First, several contributions touch upon the “lessons learned” potential of decentralized and distributed architectures, when seen from a historical perspective. The initial Internet model called for a decentralized and symmetrical organization — in terms of bandwidth usage, but also of contacts, user relations, shared knowledge and machine-to-machine communication, as Harry Halpin and Alexandre Monnin remind us. In the 1990s, the commercial and social explosion of the Internet brings about important changes, exposing the shortcomings — for the network’s usability and its very ability to function — of a model presupposing the active cooperation of all network members. The new preeminent model relies on asymmetrical fluxes and numerous barriers preventing the free circulation of data.
Today — as Dominique Boullier points out in his paper, a mix of academic and practitioner perspectives on distributed architectures whose design and development he was directly involved in — in a world of Internet services where fluxes and data converge towards a few giants, experimentations with distributed architectures are seen as a “return to the origins”. But is it really about the dominance of an organizational principle at different times in history — or is there a co-existence of different levels of resource centralization, hierarchy of powers, and cooperation among Internet users over time? Are we indeed witnessing a “war of the worlds” of which the recent tensions around surveillance are the most recent illustration?
Historical approaches also show us the importance of History for Internet regulation. As Primavera De Filippi and Samer Hassan remind us in their article, and Boullier in his, architectural choices have profoundly political implications (see also DeNardis, 2014): if code is law, and law is code, both are crucially and inextricably transformed by the Internet’s pervasiveness in all aspects of economy, policy and sociability. This Internet includes both distributed, decentralized and centralized models, rather than opposing them. Going back to History allows us to wonder whether we must return to the mythical, original model of a free, open and equal Web (see Halpin and Monnin’s contribution), or learn how to articulate the different realities that are currently forming the Internet(-s?).
2) Heterogeneity of distributed architectures
Most papers in the issue address the issue of (re-)decentralization as a sustainable alternative for the Internet ‘ecology’. The technical features of distributed architectures (direct connections, resistance to failure) and their ability to support the emergence of organizational, social and legal principles (privacy, security, recognition of rights) offer new paths of exploration and preservation of the Internet’s balance — as Mélanie Dulong de Rosnay particularly examines in her contribution. At the same time, the road towards decentralization is far from linear, and distributed organizations do not take shape according to a unique model. The users behind nodes of distributed architectures can assemble in collectives that are very varied in nature, complexity and underlying motivations. This variety may be dependent upon modes of aggregation, visibility devices, types of communication tools and envisaged business models (as well as the difficulty of identifying sustainable ones). The different models enabling the shaping of “collectives” are especially varied, and a crucial feature, as Paris Chrysos shows us in his contribution.
The heterogeneity of distributed architectures raises issues that are especially important when it comes to the interplay of law and economy in these ecosystems, an aspect touched upon by both Federica Giovanella and Panayotis Antoniadis. Network access providers raise economic objections to P2P models, having programmed and organized networking infrastructures with the idea that the most part of users’ online activities were going to consist in downloading data and information from clusters of servers. Finally, developers-turned-entrepreneurs themselves often need to revisit the choice of decentralization, because of unexpected user practices, the impossibility of making distributed technology “easy” for the public, or the seductive simplicity of centralized infrastructures and economic models (see Musiani, forthcoming in 2017).
3) User empowerment
Third, this issue addresses the communicational models at stake in decentralized infrastructures and architectures. Distributed Internet services have transformed and transform today the ways in which actors make sense of their communicational capacities and their responsibilities in information sharing. User empowerment, prompted by several P2P services — increasingly mobile, self-configurable and flexible — open innovative perspectives for infrastructures of communication, their functions and their mediation capacities among actors, a point especially illustrated by Antoniadis’ piece. As we have mentioned above, the issue of user empowerment becomes particularly salient in times when users’ trust (political, economic, technical) in authorities is crucially damaged, as Argyro Karanasiou and De Filippi and Hassan demonstrate.
This prompts further questions, such as: in what ways does this evolution transform data and communication channels? What are the representations of the values subtending these architectures and the relations among their participants, vis-à-vis other Internet services, but also within the spheres of conception, discussion and circulation of these objects? What are the new forms of contribution and what do they enable in terms of pedagogical practices and shared literacies? Finally, in which ways do distributed infrastructures relate to the notion of ‘informational common good’?
The ways in which distributed architectures reconfigure means and tools of authority and control redefines, at the same time, the very entities — people and organizations — concerned by them. As several papers show, in a perspective informed by actor-network theory, the opposition between individual and collective actors is questioned by these architectures. A redefinition and restructuring of communities happens with distributed architectures: what enables the shaping of collectives, what makes them “stick together”, what gathers them and defines their “collective intimacy” becomes more labile and less formal, as Chrysos shows. Distributed architectures can also co-shape the relationship between specific people: as the example of smart contracts demonstrates (see De Filippi and Hassan), the distribution (or decentralization) reaches the very core of the architecture and of code, when it enables the personalization of a contract linking two participants and the nature of their relationship.
The issue of user empowerment, real or expected, and the assessment of its limits, is at the core of this set of papers — and ultimately, it raises the question of the articulation between decentralization and social organization, between decentralization and governance.
4) Law, responsibility and authority
Finally — several articles, in particular those of Mélanie Dulong de Rosnay, Argyro Karanasiou, Federica Giovanella and De Filippi and Hassan are a testament to the importance of this question — distributed and decentralized architectures are used as a prism to observe the reconfigurations of law, of authority and, more generally of democracy, via different forms of distribution. Legal systems are deeply questioned and challenged by the fragmentation of entities caused by distributed architectures. However, at the same time, these could be a “reshaping element for the law”, enabling the re-conceptualization of fundamental notions such as liability, participation, property and privacy, as Dulong de Rosnay shows.
The reconfigurations of legal norms we observe in several articles are not separated from the inscription and embeddedness of norms within the architecture itself; they are rather complementary, and we need to investigate this complementarity, as Karanasiou reminds us. This opens up a number of questions: How do distributed architectures redefine user skills, rights, capacities to control? How can law support user practices and their diversity, instead of countering them? What are the differences if compared to centralized architectures, non-modifiable by users, where data are stored on clusters of servers exclusively controlled by service providers? In terms of authority and control, what are the consequences of introducing encryption, file fragmentation, sharing of disk space in the technical architecture? While “first-generation” P2P networks have affected copyright first and foremost, decentralized Internet-based services prompt us to investigate issues like the redefinition of notions such as creator and distributor, the transformed status of personal data (an issue also raised by Chrysos), the responsibility of technical intermediaries, the ‘embeddedness’ of law into technical devices, the kinds of entities it takes to empower people with knowledge (as Halpin and Monnin remind us, infrastructures and frames of thinking beyond mere data).
Ultimately, distributed architectures show the extent to which the articulation of normative systems and networked technologies is very much a “construction site” — the articles in this issue demonstrate the central role of collective norms that are often informal, and their very ambiguous (but very much present) articulation to legal regulation. As showed by specific points of tension (e.g., those described by Boullier when it comes to the blockchain subtending Bitcoin, its re-intermediations and re-concentrations), the making of distributed architectures is not the exclusive purview of so-called “democratic” arrangements; however, as Giovanella’s and Antoniadis’ papers show, the awareness of the articulation between informal/collective norms and legal regulation seem to be particularly present in those contexts where distributed architectures start to be identified, not only by user communities but by institutions as well, as legitimate and empowering alternatives.
Concluding remarks — Beyond ADAM: Present and (near-)future research on distributed architectures
In 2010, Philippe Aigrain suggested that it was time to “reclaim” Internet services, as instruments of “technical governance” perhaps better connected with the way the Internet once was. This special issue provides a rich portrait of the variety of ways in which this “reclaiming” takes shape, and suggests ways in which social, economic and legal sciences research can approach it as a subject of inquiry.
This endeavour was timely six years ago, when the ADAM project started, and appears even more timely now. Indeed, most authors in this issue are currently coordinators or members of European projects that — mostly within the frame of the CAPS programme  — investigate distributed architectures in terms of value generation, common goods, encrypted communications. The very recently-concluded P2Pvalue , of which Primavera De Filippi, Samer Hassan and Mélanie Dulong de Rosnay were core members and Francesca Musiani a collaborator, mixed technical software development with theoretical and political objectives, aiming to develop a science of commons-based peer production, particularly in terms of defining and assessing value generation in such settings, and promote their sustainability. Started in January 2016, netCommons  seeks to study, support and further promote the emerging trend of community-based networking and communication services, as “a complement, or even a sustainable alternative, to the global Internet’s current dominant model”, and benefits from the coordination of Mélanie Dulong de Rosnay and Panayotis Antoniadis and the participation of Federica Giovanella and Francesca Musiani. NEXTLEAP , also born in early 2016, coordinated by Harry Halpin and Francesca Musiani, seeks to foster an interdisciplinary Internet science of decentralization able to serve as the basis for the development of protocols for secure messaging that are both decentralized and end-to-end encrypted. In the spring of 2016, the first European call specifically dedicated to the topic, “Distributed Architectures for Decentralised Data Governance” , saw the light and the selected projects will begin their activities in early 2017. All these research activities show the liveliness and timeliness of the topic, to which we hope this issue in all its diversity yet coherence will further contribute.
As Philip Agre once stated, architecture is politics, but it is not a substitute for politics (Agre, 2003), an insight whose relevance is more evident by the day. Indeed, questions of intermediation and dis-intermediation, distribution of power, privatization of governance, privacy by design and by architecture, have hardly been more important in recent history than they are in our post-Snowden era, and reach out to the very ways in which Internet governance unfolds in our present times.
About the authors
Francesca Musiani is Associate Research Professor (chargée de recherche), French National Centre for Scientific Research (Centre national de la recherche scientifique, CNRS), Institute for Communication Sciences (ISCC-CNRS/Paris-Sorbonne/UPMC), associate researcher with the Centre for the Sociology of Innovation of MINES ParisTech-PSL, and academic editor for Internet Policy Review (https://policyreview.info). Her current research focuses on Internet governance, and draws upon an interdisciplinary training in information and communication sciences, science and technology studies, and international law. She is currently one of the Principal Investigators for the European H2020 project NEXTLEAP (2016-2018, Next-Generation Techno-Social and Legal Encryption, Access and Privacy).
E-mail: francesca [dot] musiani [at] cnrs [dot] fr
Cécile Méadel is Full Professor at the Université Panthéon-Assas, Institut Français de Presse/CARISM, and associate researcher with the Centre for the Sociology of Innovation of MINES ParisTech-PSL. A sociologist, she has devoted most of her work to the uses of communication technologies. Originally trained as a historian, she has focused on the genealogy of media and particularly the way in which designers, retailers and users collectively negotiate technological and political options that subsequently become irreversible, at least for some time, and can redefine organizations and professional or individual practices. She led the ADAM project from 2010 to 2014.
E-mail: cecile [dot] meadel [at] u-paris2 [dot] fr
Francesca Musiani is supported by the European Union’s Horizon 2020 Framework Programme for Research and Innovation (H2020-ICT-2015, ICT-10-2015) under grant agreement number 688722 — NEXTLEAP. Both she and Cécile Méadel were previously supported by the French National Agency for Research (Agence Nationale de la Recherche, ANR), CONTINT programme, ANR-10-CORD-004 — ADAM — Architectures distribuées et applications multimédias.
1. Collective Awareness Platforms for Sustainability and Social Innovation, seeking to “harness the collaborative power of ICT networks to create collective awareness of sustainability threats and enable collective solutions” — https://ec.europa.eu/programmes/horizon2020/en/h2020-section/collective-awareness-platforms-sustainability-and-social-innovation-caps.
2. https://p2pvalue.eu/, Techno-social platform for sustainable models and value generation in commons-based peer production in the Future Internet.
3. http://netcommons.eu, Network Infrastructure as Commons.
4. https://nextleap.eu/, NEXt generation Techno-social and Legal Encryption Access and Privacy.
Philip E. Agre, 2003. “P2P and the promise of Internet equality,” Communications of the ACM, volume 46, number 2, pp. 39–42.
doi: http://dx.doi.org/10.1145/606272.606298, accessed 10 November 2016.
Philippe Aigrain, 2010. “Declouding freedom: Reclaiming servers, services and data,” In: 2020 FLOSS roadmap (https://flossroadmap.co-ment.com). Third edition, at https://flossroadmap.co-ment.com/text/NUFVxf6wwK2/view/, accessed 10 November 2016.
Laura DeNardis, 2014. The global war for Internet governance. New Haven, Conn.: Yale University Press.
Ksenia Ermoshina, Francesca Musiani and Harry Halpin, 2016. “End-to-end encrypted messaging protocols: An overview,” In: Franco Bagnoli, Anna Satsiou, Ioannis Stavrakakis, Paolo Nesi, Giovanna Pacini, Yanina Welp, Thanassis Tiropanis and Dominic DiFranzo (editors). Internet science: Third international conference, INSCI 2016, Florence, Italy, September 12–14, 2016, proceedings. Lecture Notes in Computer Science, volume 9934. Berlin: Springer-Verlag, pp. 244–254.
doi: http://dx.doi.org/10.1007/978-3-319-45982-0_22, accessed 10 November 2016.
Cécile Méadel and Francesca Musiani, 2015. Abécédaire des architectures distribuées. Paris: Presses des Mines.
Francesca Musiani, Derrick L. Cogburn, Laura DeNardis and Nanette S. Levinson (editors), 2015. The turn to infrastructure in Internet governance. New York: Palgrave Macmillan.
Francesca Musiani, forthcoming in 2017. “Quando il peer-to-peer si ‘ricentralizza’: Vincolo sociotecnico, spinta di mercato o fallimento?” In: Gabriele Balbi and Paolo Magaudda (editors). Fallimenti digitali: Un’archeologia dei nuovi media. Bologna: Unicopli.
David E. Sanger and Nicole Perlroth, 2015. “Encrypted messaging apps face new scrutiny over possible role in Paris attacks,” New York Times (16 November), at http://www.nytimes.com/2015/11/17/world/europe/encrypted-messaging-apps-face-new-scrutiny-over-possible-role-in-paris-attacks.html, accessed 10 November 2016.
Received 6 November 2016; accepted 10 November 2016.
This paper is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
“Reclaiming the Internet” with distributed architectures: An introduction
by Francesca Musiani and Cécile Méadel.
First Monday, Volume 21, Number 12 - 5 December 2016