Empirical Studies of Bottom-Up Internet Governance

Abstract

The notion of bottom-up governance in the Internet is not new, but the precise underlying mechanisms have received little primary, empirical study. The majority of Internet governance literature is couched in contrasting familiar top-down modes of governance with the design of and subsequent critique of governance institutions such as ICANN or the WSIS processes that created the Internet Governance Forum (IGF). This paper reports on dissertation work collecting and analyzing empirical evidence of how bottom-up governance mechanisms operate in situ. Methodologically, participant-observer ethnographies are supplemented by text mining and social network analysis — the combination facilitates analysis of community-generated artifacts cross-validated against semi-structured interviews. This paper reports on ethnographic studies thus far, drawing on early interviews and private conversations. Scoping the domain, this work evaluates organizational modes at the intersection of critical Internet operations and security. Three categories of non-state organizational modes contribute evidence: network operator groups (NOGs) and RIRs; Internet eXchange Points (IXPs); anti-abuse organizations and communities such as the Messaging, Malware, and Mobile Anti-Abuse Working Group (M3AAWG), Spamhaus, and the Anti-Phishing Working Group (APWG). As of this writing, the anti-abuse studiy is the least developed study and will be addressed comparatively. The author engages as a participant-observer in forums from each category, developing relationships and engaging in semi-structured interviews with participants and organizers. This work attempts to understand how decentralized “close-knit yet loosely organized” communities coordinate localized operational capacity (direct access to private network operations and security incidents) to achieve global operational and decisional capacity sufficient to address problems as they arise, at Internet clockspeed. Ongoing fieldwork provides early insights. Many of these governance arrangements comprise actors pursuing nominally private interests, yet produce substantive collateral public benefits. As such, these arrangements are framed as instances of private authority. Important to this work is that the combination of private interests and the creation of public goods is distinct from both how open these organizational modes are and how transparent the attendant processes are. Various combinations exist within these studies and will be presented. This report will provide a preliminary comparative analysis within and across the three categories of organizational modes described above. Comparative analyses will highlight, among other factors, the variety of vetting and consensus building processes at play, trade-offs between formal and informal rules and norms, mechanisms for evaluating policies, and the density of social networks that facilitate communication within and across differentiated policy and issue communities. Taken together, these factors will contribute to an argument that bottom-up governance (of the Internet) is not simply a varient of multistakeholder-ist or multilateralist governance confounded by a functionalist and/or corporatist flavor. Rather, bottom-up governance will be presented as a broad yet densely connected, pluralistic marketplace of governance arrangements whose continuous engagement in policy experiments allows the community as a whole to keep pace with the development of issues in and on the global Internet. 1. As a report mid-study, attribution is intentionally vague. Respecting existing agreements with research subjects, specific attribution that is not already available publicly will not be addressed in this paper or the presentation. Later publications will provide specific attribution based on those agreements. 2. This language is drawn from Ellickson’s classic study Order Without Law (1991). 3. This work draws on theories of private authority, in particular notions of private regimes and authority derived from networked political configurations rather than state-based authority (direct or delegated). That said, theory will be addressed briefly, but will not be the focus of this work. .