The Struggle Over Digital Infrastructure

In July 2024, a group of contributors to the Internet governance community penned an Open Letter to the United Nations Secretary-General and Envoy on Technology to raise concerns over the ongoing drafting of the Global Digital Compact, a UN process to “outline shared principles for an open, free and secure digital future for all.” (Disclaimer: I co-signed that letter but here speak only for myself.) The core issue raised in the letter is a perceived shift away from “the enormously successful multistakeholder Internet governance practice that has brought us the Internet of today” and toward a more government-centric approach.

If you’ve been on the Internet at any point in the past decade or so, you might rightly wonder what this “enormously successful” Internet governance practice is and where in the digital space you can find evidence of it. By and large, the Internet seems more akin to a failed state, at best an undergoverned space where fragments of essential infrastructure are provided at the whim of local warlords. And indeed, the Global Digital Justice Forum published a crisp and forceful rejoinder to that open letter, pointing out several shortcomings of today’s Internet governance regime and noting that “the cyberlibertarian vision of yesteryears is at the root of the myriad problems confronting global digital governance today.” It’s hard to disagree.

To understand how it's possible to see the truth in both viewpoints, it's worth taking a step back both in scale and in time to look at the Internet's architecture and at some of its history.

One important property of the Internet is its adhesion to the end-to-end principle, one formulation of which is: “Nothing should be done in the network that can be efficiently done in an end-system.” This may seem somewhat abstract, but we can see the end-to-end principle at work in other infrastructural systems. For instance, if you invent a new type of light bulb or toaster, you don’t need to change the lighting or toasting functions of the electrical grid. You don’t need to substantially upgrade your roads to support electric cars. No one has proposed a next-generation water distribution network so that people may use home fizzy water devices. The end-to-end principle is an important form of modularity between infrastructure and its uses and thus creates the potential for invention, improvements, and happy accidents that go far beyond the imagination of infrastructure providers. From a governance perspective, it can be seen as a subsidiarity principle whereby a central authority — here, the infrastructure — performs only those tasks which cannot be performed at a more local level.

Seen from 2024, this might strike readers as obvious but it wasn’t always so. Before the Internet emerged as the more successful alternative, it was competing for funding and attention with a much more telecoms-centric model. Under the telecoms model, intelligence resides in the network rather than at the edges. This severely constrains the ability of anyone outside of the telecoms operator to create new services and indeed for the longest time in many countries using the phone network for any purpose other than those intended by the operator could lead to steep penalties.

Such a model is one of intermediary capture: when your infrastructure shapes and forecloses what you can do. Crucially, intermediary capture does not only describe the world we had when networking was dominated by telecoms operators, it is also an apt description of the world we now have in which many infrastructure layers of our digital spaces have been captured by tech platforms. In more ways than one, phone companies were the previous generation’s Big Tech.

As governments around the world reckon with the monopolization of our digital sphere and seek to address it in part through industrial policy and massive investment (as supported in Europe by the Draghi report coming hot off the presses), it’s tempting to indiscriminately toss the current system out and direct that cash to telecoms operators who are only too happy to promise next-generation what-have-yous and to reconnect with their tarnished pasts as national champions. What would that look like?

That is where history can help us: we’ve been here before. Writing in 1976, French Internet pioneer Louis Pouzin concluded that “the political significance of the controversy [between Internet and telecoms approaches to networking] is much more fundamental, as it signals initial ambushes in a power struggle between carriers and the computer industry. Everyone knows that in the end, it means IBM vs. Telecommunications, through mercenaries. It may be tempting for some governments to let their carrier monopolize the data processing market, as a way to control IBM. What may happen, is that they fail in checking IBM, but succeed in destroying smaller industries.”

And indeed, that is exactly what came to pass. Pouzin’s own France focused on deploying the Minitel instead of making greater use of his pioneering work that became foundational to the Internet model. The Minitel embodied much that one should want from a dream industrial policy: it was hugely ambitious and egalitarian, putting a free device in every household that wanted one. It was also, for a time, extremely popular. But its limited architecture proved to be its downfall. The device was a dumb terminal entirely designed and controlled by the French national carrier. All services had to be channeled through the carrier’s approval and they took a cut from all interactions. When Internet service started becoming more accessible and the web took off, the Minitel didn’t stand a chance.

We should learn from that experience. Global majority countries (as well as most global minority ones) are entirely right to feel colonized by Google and other platforms, in a very literal sense. The Internet is “the infrastructure of all infrastructure” and having one’s critical infrastructure controlled and exploited by a foreign entity is colonial. But the way forward does not lie with a reversal to the telecoms model. We don't need to Minitelize the Internet to liberate ourselves from those who have captured it. We need not reproduce the mistakes that Louis Pouzin saw coming.

That's where the concerns of the open letter and its rejoinder meet: we need to overthrow the corporate control that throttles our invention and democracy but without reverting to our previous masters. For that, state participation and industrial policy are needed, but they must be deployed in ways that align with and protect the Internet’s core architectural principles, and that apply these principles to the parts of the digital stack that have been captured. Nothing is more powerful in countering hegemonic monopolies than supporting the genius of a vast ecosystem of small and medium players.

The focus of that push must be on infrastructure and its governance. It can often be tempting for industrial policy to pursue innovation investment not just in fundamental R&D (which is always needed) but also much closer to end-user products. But innovations that have to then operate on captured infrastructure never stand a chance since the infrastructure provider can always manipulate the rules of the game. Instead, investing massively in scaling up modular, open, democratically controlled infrastructure is the key to securing innovation from society at large and to unleashing the competitiveness of thousands of smaller projects. Infrastructure is less glamorous — but far more powerful, and to liberate the Internet we need power more than we need glamor.

Keeping such infrastructure open is the challenge presented by the governance aspect of this drive to digital independence. Existing Internet governance instances and standards bodies are woefully under-scaled, remain rooted in market-centric "voluntary standards" when those markets are not competitive, and suffer from way too narrow an understanding of what counts as digital infrastructure. Digital infrastructure comprises all of our digital shared means — including browsers, commerce, secure chat, search, social, and more — and not just relatively low-level protocols with “AI governance” thrown in to sound up to date.

We need to radically strengthen the transnational nature of multistakeholder digital infrastructure governance while organizing it to prevent domination and to support the states’ right to self-determination. We need planetary institutions for various components of digital infrastructure that operate at the scale of the Internet but 1) offer sufficient subsidiarity that states can exert more regional control where meaningful, 2) support enough experimentalism that no one needs to become stuck with a narrow set of Silicon Valley values, and 3) drive a sufficient dispersion of power to enable democracy. Establishing how such institutions should operate and how to fund their work is today’s central challenge in digital policy. Yes, they must maintain “multistakeholder (…) Internet governance” and yes, they must abandon the “cyberlibertarian vision of yesteryears.”

Things are stirring. In Europe, where subsidiarity and experimentalism form a well-trodden territory, the upcoming event “Toward European Digital Independence” should help shape investment in digital infrastructure and governance. We can only hope that political leaders, both there and around the world, will deploy digital industrial policy with an understanding of the Internet’s architecture and history, keeping in mind the tempting errors of the telecoms model.