How We Ruined The Internet

Abstract: At the end of the 19th century the logician C.S. Peirce coined the term "fallibilism" for the "... the doctrine that our knowledge is never absolute but always swims, as it were, in a continuum of uncertainty and of indeterminacy". In terms of scientific practice, this means we are obliged to reexamine the assumptions, the evidence, and the arguments for conclusions that subsequent experience has cast into doubt. In this paper we examine an assumption that underpinned the development of the Internet architecture, namely that a loosely synchronous point-to-point datagram delivery service could adequately meet the needs of all network applications, including those which deliver content and services to a mass audience at global scale. We examine how the inability of the Networking community to provide a public and affordable mechanism to support such asynchronous point-to-multipoint applications led to the development of private overlay infrastructure, namely CDNs and Cloud networks, whose architecture stands at odds with the Open Data Networking goals of the early Internet advocates. We argue that the contradiction between those initial goals and the monopolistic commercial imperatives of hypergiant overlay infrastructure operators is an important reason for the apparent contradiction posed by the negative impact of their most profitable applications (e.g., social media) and strategies (e.g., targeted advertisement). We propose that, following the prescription of Peirce, we can only resolve this contradiction by reconsidering some of our deeply held assumptions.

• The paper reveals that the original Internet design limits asynchronous point-to-multipoint communication, necessitating overlay networks like CDNs.
• The paper quantifies infrastructure growth with CDN servers rising from 200,000 in 2017 to 350,000 in 2022, underscoring scaling challenges.
• The paper advocates reexamining foundational architectural assumptions to foster a more inclusive and efficient ICT ecosystem.

Critical Examination of the Internet's Architectural Assumptions

The paper "How We Ruined the Internet" by Micah D. Beck and Terry R. Moore conducts an incisive analysis of fundamental assumptions embedded within the original Internet architecture and their implications for modern-day ICT infrastructure. It explores the paradoxical consequences of these assumptions and suggests that re-evaluating them, in line with the philosophical tenet of fallibilism, may be necessary to address the issues faced today.

Key Arguments and Findings

A central thesis of the paper is the assertion that the conception of the Internet as a datagram forwarding "stovepipe" inherently limited its capacity to support asynchronous point-to-multipoint communication, which is critical for many modern applications. This deficiency, the authors argue, necessitated the creation of private overlay networks, such as Content Delivery Networks (CDNs) and cloud services, which ultimately diverged from the open and egalitarian ideals of the early Internet advocates.

The paper introduces the distinction between what is referred to as Internet* and Internet++. Internet*, the original network protocol suite, emphasized unicast datagram delivery, while Internet++ accounts for the complex infrastructure supporting today's applications, including CDNs and cloud networks. This delineation is crucial in understanding the evolution from an open data network paradigm toward a more commercial and controlled Internet ecosystem.

The authors provide quantitative insights into the infrastructure growth necessitated by increasing Internet traffic. The escalation in the number of CDN servers—from 200,000 in 2017 to 350,000 by 2022—illustrates the scaling challenges and the underlying architectural evolution required to accommodate such exponential traffic increases.

Implications and Future Directions

The paper's findings have significant implications for both theoretical research and practical infrastructure development within the field of computer science and networking. It challenges the community to rethink deeply rooted assumptions about network design, particularly the focus on point-to-point, loosely synchronous communication. Furthermore, it sheds light on issues such as network ossification, where the emphasis on high-capacity, low-latency unicast forwarding may inhibit more efficient content distribution solutions.

The proposition of revisiting and potentially reshaping architectural norms may foster a more inclusive and balanced ICT ecosystem, potentially addressing inadequacies in universal service access and combating the commercial dominance of hypergiant tech organizations. In this context, the paper serves as a call to action for the networking community to critically assess the long-term implications of current infrastructural trends and explore alternative pathways that align more closely with the original open data network vision.

Critique and Closing Thoughts

While the paper provides a thorough examination of historical and technological factors leading to current ICT challenges, it raises important yet complex questions about the feasibility and societal impact of reevaluating foundational Internet assumptions. The thorough treatment of architectural constraints, alongside the economic dynamics of private network infrastructure, provides a compelling narrative for rethinking the Internet's future trajectory.

Overall, the paper urges researchers and practitioners to apply a critical lens to the accepted paradigms and consider whether current network architectures serve the broader societal interests envisioned by the Internet's pioneers. By doing so, the field may move towards an Internet that balances innovation, accessibility, and open collaboration, potentially paving the way for more equitable technological advancement.