Freedom Index
Listen to the article
The Metric Is Not the Mission is a ten-part examination of how Big Tech moved from building and expanding the open internet to increasingly shaping it around its own metrics, incentives and assumptions. Across the series, the argument follows the evolution of the platform economy—from the optimism of the early internet to the growing tensions around power, prediction, geopolitics, accountability and the future of digital life.
The series will be published in two parts each week over five weeks, with each installment building on the one before it. At the end of the series, the complete essay will be brought together in a single PDF edition, providing the full argument in one place.
Part V: The Finished Internet
In Part IV, the argument shifted from individual storms to the changing digital climate. Part V examines another consequence of that change: how an internet designed to remain open and unfinished gradually became a collection of increasingly complete, privately governed environments.
There is an observation that appears repeatedly in the history of architecture. Buildings designed to accommodate uncertainty often outlive those designed to achieve perfection. Medieval townhouses become workshops, then apartments, then cafés. Former factories are transformed into galleries. Railway stations become museums. The structures endure because they were never so complete that they resisted reinvention. Their longevity lies not in their efficiency but in their incompleteness.
The same principle has shaped some of humanity’s most enduring systems. Languages survive because they constantly absorb new words. Constitutions persist because they leave room for interpretation. Scientific knowledge advances because every answer generates new questions. Healthy ecosystems are not stable because nothing changes within them; they are stable because change is built into their very nature.
The internet belonged to this family of unfinished systems. This is perhaps the most misunderstood aspect of its design. We often describe the internet as open, decentralized or interoperable, and while each of these descriptions is accurate, they are ultimately consequences of a more fundamental characteristic. The internet was deliberately incomplete. Its architects resisted the temptation to determine in advance how people should use it because they recognized that no one could anticipate the forms of innovation that would emerge decades later.
That decision required an unusual degree of intellectual humility. The protocols that underpin the internet perform remarkably modest tasks. They move packets from one place to another. They identify computers, route traffic and establish common rules through which independently developed systems can communicate. They do not prescribe what applications ought to exist, which services deserve prominence or what business models should prevail. In this respect, the internet resembles less a finished product than a constitutional framework. Its purpose is not to dictate outcomes but to create conditions under which countless different outcomes remain possible.
This architectural restraint proved unusually fertile. Email was not foreseen in its modern form. Neither was the web, online commerce, video streaming, collaborative encyclopedias or social media. Each emerged because the underlying network imposed positively few assumptions upon those building at its edges. Innovation flourished precisely because the center remained intentionally modest.
The major platforms inherited that openness and initially extended it. They built services upon common protocols, benefited from the web’s extraordinary reach and attracted users through products that genuinely expanded what the internet could do. In their early years, they felt less like destinations than gateways. Search engines pointed users toward independent websites. Social networks drove readers to blogs, newspapers, and forums. Video platforms introduced audiences to creators who maintained homes elsewhere on the web. The internet remained larger than any individual service.
At some point, almost imperceptibly, that relationship reversed. The platforms stopped behaving as participants within a broader ecosystem and began organizing the ecosystem around themselves. Their success no longer depended primarily on helping users explore the web; increasingly, it depended on reducing the need to leave the platform at all. Links became interruptions rather than invitations. External websites loaded inside proprietary browsers. News articles were hosted directly on platforms rather than by publishers. Native content received greater visibility than material residing elsewhere. Every additional minute spent beyond the platform gradually came to be understood as value escaping the system.
None of this represented a conspiracy against the open internet. It emerged from incentives that were, from the perspective of individual firms, entirely rational. Every company seeks to improve user experience, reduce abandonment and strengthen customer loyalty. Yet rational decisions at the level of individual organizations do not always produce desirable outcomes for the larger systems within which those organizations operate. Economists have long recognized this tension. Ecologists encounter it constantly. The pursuit of local optimization often degrades the health of the wider environment.
The internet increasingly exhibits precisely this pattern. The more successful the largest platforms became at perfecting their own environments, the less space remained for the kind of distributed experimentation that had characterized the web’s formative decades. Independent blogs gave way to creator platforms. Personal websites yielded to profile pages. Conversations migrated from open forums to proprietary feeds. Software developers increasingly built applications for ecosystems rather than for the web itself. Even innovation began to change its character. Instead of asking what new possibilities the internet might enable, entrepreneurs increasingly asked how new ideas might fit within the rules established by the dominant platforms.
This shift carries consequences that extend well beyond commercial competition. General-purpose technologies derive much of their value from uses that cannot be predicted in advance. Nobody designed the web specifically to produce online education, open-source software, citizen journalism, or remote work. These emerged because the underlying architecture permitted combinations that no planner had anticipated. Economists sometimes describe this as combinatorial innovation: progress generated not by isolated breakthroughs, but by the continual recombination of existing capabilities into unexpected forms. The internet excelled precisely because it lowered the barriers to such experimentation.
Finished systems rarely possess the same generative quality. A finished system is elegant because its boundaries are well defined. Every component has a designated function and interactions are carefully orchestrated. Unexpected behavior is treated as a bug rather than a possibility. This approach works exceptionally well for products. It works less well for civilizations.
The distinction matters because the internet was never merely a product. It became one of the foundational infrastructures of modern society, not unlike roads, electricity or the postal system. Such infrastructures succeed when they enable activities far beyond those imagined by their creators. Their greatest contributions are often indirect. We value roads less because of the asphalt itself than because of the countless journeys they make possible. Likewise, the internet’s greatest achievement has never been any particular product or service but its capacity to generate applications that nobody had the foresight to invent.
There is another, subtler consequence of completeness. Finished systems cultivate a particular kind of confidence. Their designers begin to believe that the most important questions have already been answered. Improvement becomes a matter of optimization rather than imagination. More efficient recommendation engines; more accurate behavioral prediction; more personalized advertising; better artificial intelligence. These are not trivial achievements; on the contrary, they represent extraordinary feats of engineering. Yet they all operate within an existing conceptual framework. They assume that the future consists primarily of making the present work more effectively. History rarely rewards that assumption.
The technologies that reshape societies seldom emerge from the centers of existing power. They arrive from the margins, where experimentation remains inexpensive because failure carries fewer consequences. The personal computer was not born inside the mainframe industry. The web did not emerge from the telecommunications monopolies. Open-source software challenged assumptions held by the largest software companies of its time. Innovation repeatedly enters history through spaces that established institutions have overlooked precisely because those institutions are busy refining what they already possess.
This is the paradox confronting today’s technology giants. No companies have invested more heavily in imagining the future. Their research laboratories employ some of the world’s finest scientists while their capital expenditures rival those of nation states. Their ambitions increasingly encompass artificial intelligence, biotechnology, robotics, and space. Yet for all this extraordinary capacity, they often seem curiously reluctant to imagine a future in which the internet itself might once again become larger than the platforms they have built upon it.
Perhaps that is because every successful institution eventually begins to confuse its own permanence with the permanence of the system that made it possible. The internet has never made that mistake. Its greatest strength has always been its willingness to remain unfinished.
Konstantinos Komaitis, PhD, is a veteran of developing and analysing Internet policy to ensure an open and global Internet.
Read the full article here
Fact Checker
Verify the accuracy of this article using AI-powered analysis and real-time sources.