IMR: MT: A Community Platform for Controlled Experiments on Internet Access Networks

University Of Chicago, Chicago IL

Investigators

Abstract

Internet measurement research has long operated on convenience samples---measurements that are derived from convenience, based on the sets of vantage points that can be obtained through access to a particular testbed, the deployment of an application, or the opportunistic recruitment of participants. Throughout the early decades of the Internet, such an approach was sensible and could often provide important insights about the design of specific protocols or applications, or even some aspect of the Internet writ large. Over the past several years, however, the research community---and society more broadly---has begun to ask a broader set of questions, moving beyond those of technical protocol or application design, to questions of social or policy relevance. The current measurement tools and platforms are ill-equipped to address the concomitant research questions that intersect with broader societal questions. This project aims to develop a platform to change that, delivering the first-ever network measurement platform that can facilitate studying a broader set of research questions than are possible with existing platforms. Perhaps one of the most striking examples of such a set of questions relates to the Internet's digital divide. The challenge is widely acknowledged, by researchers, advocacy organizations, and policymakers alike, as a significant challenge, and providing all Americans with affordable Internet connectivity is a national priority as well. This project is developing a measurement platform that will enable this new family of measurements. Building on early successes in a pilot deployment in Chicago, the project aims to design, develop, and deploy a first-of-its-kind measurement infrastructure that is specifically designed to enable new classes of research questions concerning digital infrastructure across our nation's population (and ultimately, the global population). The infrastructure is open-source, and incrementally deployable and expandable, allowing the project and infrastructure to scale from pilot to national or global deployments as lessons from early deployments are applied to a larger testbed. The potential outcomes of this platform are profound and wide-ranging, catalyzing an entirely new class of Internet measurements and research studies---shifting the computer networking and Internet measurement research areas from questions that focus only on convenience samples (thus limiting our research questions to those of technology and protocols) to samples that concern infrastructure and population (thus allowing networking researchers to ask a broader set of technical questions with importance across disciplines, from economics to policy to social science). This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.