Switching receipts to mobile, absorbing offline procedures into apps, and aggregating data to optimize flows are often accompanied by the rhetoric of ESG. “Use less paper, reduce mobility, operate more efficiently.” All of this sounds convincing enough. Yet the “lightness” of digitalization is, in fact, another expression of a materiality that has grown massive out of sight. And it is precisely this invisibility that so often deceives us.
Let us strip away the assumption we have carelessly believed—“digital = eco-friendly”—and descend through the three layers that uphold the digital society: action (interfaces), infrastructure (data centers, energy generation, cooling water), and networks (terrestrial and submarine cables). Beneath these, we will examine the materiality of devices, the corporate power inscribed in design, the profit structures of the data economy, and the fracture points of institutions. Finally, we will reflect on the small yet resilient practices and the directions of institutional change that remain open to us.
Light Fingertips, Heavy Traces — The Heat of Interfaces
What happens when you send a single photo from your smartphone is far from simple. The photo travels from your device to the nearest cell tower, then to a network hub, and finally to a data center server where it is stored. When the recipient requests it, the server generates a copy and sends it back through the network in reverse. What appears as “instant transmission” on the surface is, in fact, a chain of repeated requests, responses, and duplications.
We experience this as a mere “single click,” yet that click involves roughly half a dozen interactions (interfaces) and multiple acts of copying and scaling. Each time, electricity flows and servers release heat. The notion that “digital is immaterial” is, in reality, closer to saying that materiality has simply been displaced—pushed further away, hidden from view. Our fingertips may touch lightly on the screen, but countless layers of infrastructure quietly carry the weight.
Cooling, Storing, Never Powering Down — The Reality of Data Centers
Data centers never switch off. Our societies have come to depend on them to keep payments, banking, administration, and messaging running without interruption. To maintain this constancy, data centers are built with redundant power supplies and demand vast amounts of cooling water to dissipate the heat emitted by server racks.
Large centers on the outskirts of cities are easy to spot: tall ceilings, massive fans, the constant hum of electronics, racks exhaling hot air, and nearby waterways diverted to serve their needs. In regions where such centers cluster, shortages of household water, noise pollution, and land encroachment become everyday realities. The term “cloud” suggests something intangible, floating in the sky, but its reality is concrete: buildings, electricity, and water. The “cloud” exists only because someone guards those buildings, cools that heat, and channels that water.
The Genealogy of Minerals — The Weight of the Earth Inside a Smartphone
Whereas older mobile phones required only about ten types of minerals, today’s smartphones rely on more than fifty. Many of these are rare metals with few uses outside the electronics industry. Mining them transforms mountains and rivers, darkening soils and skies with pollution.
An even greater challenge is recycling. While devices may be collected, the technology to recycle specific elements often lags behind—or, when the technology exists, collection systems fail to deliver. For many elements, recovery rates remain below 20 percent.
We often say, “All we need is to cut carbon.” But without seeing the entire material flow—from extraction to processing, distribution, use, and disposal—our judgment is blurred. This is why the perspective of MIPS (Material Input per Service), long emphasized by ecological economists, remains so relevant. Holding a single phone in your hand also means carrying the hidden weight of hundreds, even thousands of times more material that was consumed along its lifecycle. Yet we rarely feel that weight.
Engineered Obsolescence — Power Between Updates and Ownership
Digital corporations may “sell” us devices, but the software and operating systems inside them are usually only licensed for use. Once updates are discontinued, perfectly functional hardware can suddenly become “unusable.” When batteries are designed to be irreplaceable or component standards are closed off, reasonable repair and longer lifespans become increasingly difficult.
In response, citizens have begun to demand the Right to Repair. Small neighborhood repair workshops, volunteer-run FabLabs, and initiatives like Fairphone—with its modular design and ethical supply chains—stand as quiet acts of resistance against the command to “discard quickly and buy new.” Repair is not only about reclaiming technology; it is about slowing down the timeline of waste.
How Data Becomes Money — The Economics of “Free” and the Hidden Pathways
The reason platform companies can sustain the massive costs of data centers is simple: data is money. Our interfaces—clicks, scrolls, plays, searches, sign-ups—are repackaged as “big data” and resold in the languages of advertising, recommendation, placement, and bidding.
Take, for example, an online shop that adopts a simplified sign-up API. While it reduces friction for users, it also greatly expands the platform’s access to personal and behavioral data. More information than we ever intended is shared with third parties by design.
Even businesses with short-lived devices and high repair or insurance costs—such as electric scooters—survive because the movement data of Gen Z consumers holds market value. In such cases, what we call “innovation” is often nothing more than the cold logic of data accumulation and sale.
Is Faster Really Greener? — The Law of Rebound
As networks have advanced from 3G to 4G to 5G, we have come to feel that things are becoming “more efficient.” Yet in reality, device replacement, the expansion of base stations, the laying of new cables, and the surge in data consumption all occur simultaneously. The rebound effect—where efficiency gains are offset by increased consumption—has been observed countless times in transport and energy. Digital technologies are no exception. The optimism that “technological progress will offset emissions” is, for now, far from being empirically proven.
What Must Change? — Small Habits, Large Structures
Given this reality, what should we do? I believe we must move on two levels simultaneously.
The first is the level of everyday life. This includes setting email and messenger apps to block automatic image loading by default, clicking only when necessary; skipping the portal homepage to avoid unnecessary reloads and going directly to frequently used sites via bookmarks; regularly deleting redundant emails, large attachments, and duplicate photos to reduce both interface interactions and storage loads. It also means adopting consumer principles such as using devices longer, repairing them, and choosing modular designs or standardized parts. These actions are not merely about “saving”; they are design choices. Even “one less click” can reduce the chain of signals exchanged and copies generated behind the screen.
The second is the level of institutions. This involves requiring data centers to transparently disclose their electricity and water usage, backup power systems, and heat management strategies; establishing criteria to evaluate the social costs of water and land use in local communities; and moving away from narratives of reduction based on easy green certifications, instead prioritizing absolute reductions verified by actual measurements. It also means guaranteeing the Right to Repair through laws and standards, mandating modularity, disassembly, and component compatibility at the manufacturing stage, building public infrastructure for e-waste collection and recycling, and instituting transparent tracking of import–export flows.
Such transformations can be explained through the lens of Doughnut Economics. The social foundation that upholds human dignity can be met with technology, but industries and policies must also inscribe principles of restraint and redistribution, ensuring we do not push beyond the planetary boundaries within which ecosystems can endure.
Conclusion — A New Sensibility of Technological Citizenship
Digital transformation can contribute to solving environmental problems. But it is not, in itself, the solution. Digital is never a “light” world; it is another face of the vast material realm composed of electricity, water, minerals, and land. That is why we must slow down the rush of the word “innovation” and first look at the material foundations and power structures on which it rests.
I call this sensibility technological citizenship: not citizenship confined to the screen in front of us, but one that imagines the resources, infrastructures, labor, and ecologies beyond the screen. It begins with small habits—turning off auto-loading, using bookmarks to skip redundant pages, repairing and extending the life of our devices—but extends to larger shifts: demanding accountability from data centers, ethical design from manufacturers, and new frameworks of regulation and disclosure from governments.
Each time we create one less interface with a fingertip today, a server somewhere produces one less burst of heat. Teaching each other this small awareness, and building institutions that sustain it—that, I believe, is the next sentence that environmental education and civic politics must write in the digital age.
Pitron, G. (2023). How Likes Can Save the Planet (H. Lee, Trans.). Gimmyoung.