We tend to perceive digital technologies in the same way we understand a hammer or a pair of scissors. When we use a Nintendo Switch to play a video game or a smartphone to watch an episode of our favorite series, it feels as if we are interacting with a single device or technology. In reality, however, we are dealing with a great number of technologies at once. The devices can only fulfill all of their valuable functions because they work together with a backbone of software, servers, cables and antennas. In other words, other than ‘traditional technologies’, we can only understand digital technology when we look beyond the device that is directly in front of us.
At FreedomLab Thinktank, we use the framework of the Stack to understand the anatomy of digital systems. The Stack presents digital technology as a layered structure of technological and non-technological components. The framework has originally been developed by philosopher Benjamin Bratton as a means to understand the geopolitical impact of digital technology. We have reinterpreted Bratton’s work and developed it into a tool for analysis and strategic decision-making.
The notion of a vertically layered stack of functional components, actually originates in software development. According to the principle of "separation of concerns", elements are delineated based on their function and internal coherence, which leads to a high degree of modularity, which thus characterises digital systems. That is, the lower layers generally refer to more stable infrastructural technologies, while the higher layers refer to more user and context specific infrastructural combinations that consequently are more fluid and adaptive. We adopted this way of thinking as an appropriate model to think about how digital technologies have reshaped society in the past and how they will do so in the future.
As an analytical tool, the Stack helps us to identify critical developments on individual layers of the stack, such as new approaches in artificial intelligence. Even more so, it sheds light on the interplay between developments on different layers of the Stack; for instance how these can lead to new business models or trigger an acceleration of innovation, as was the case with the breakthrough of the smartphone. Because our interpretation of the Stack includes technological and non-technological components, it also allows us to consider the social and political impact of digital systems, as well as new means of governing the technology itself.
Strategically, the Stack is a practical tool to consider an organization’s – or nation’s – digital strengths and weaknesses, to identify opportunities or to develop new ideas. This may relate to one’s position on one or more layers of the Stack, in a given sector or value chain, but also to an organization’s ability to employ its technological or organizational capacities to make strides in another sector.
We will use the example of the smartphone to walk you through the layers of the stack.
The best way to enter the stack is the interface layer. Human beings use language and semantics the computer cannot understand, so to ‘communicate’ with a digital system, we need to translate human language into computer language. The interface is the layer of technology that handles this task. Currently, the Graphical User Interface (GUI) of the mobile has replaced desktop screens, mouses and T9 keyboard as the dominant interface. But users ultimately approach services through various user interfaces. This interaction may occur along all kinds of different modalities such as vision (screen, VR headset), speech (voice assistant), gestures (3D cameras) and hearing (wireless earphones). On the one hand, user interfaces are a determinant of the information and experience that digital systems can convey, and on the other hand of the type of data that can be gathered about the end user and their environment.
One layer below the interface is the service layer, which simply tells us what the task at hand is the digital system fulfills. Industries such as entertainment, education and shopping have all been ‘disrupted’ by digital technologies and are now largely mediated by ecosystems that operate based on the logic of the Stack. Opening the home page of our smartphone, users can nowadays participate in a wide range of services and tasks within the app economy. A service could take the form of a platform that brings together supply and demand (e.g. the sharing economy) or where communication occurs. Social media, ecommerce and recently also telecommunications services have brought the functions and services of physical places into the digital realm. A large share of the services only relates to the domain of information (such as communication or the streaming of content) but can, naturally, also relate to the physical world (e.g. online shopping or ordering a taxi).
If we move further below, we notice that these services often operate on intelligence. When we scroll through e-commerce websites or use google maps on our smartphone, algorithms personalize our feeds or optimize routes based on real-time traffic data. In this layer, these "smart" algorithms are capable of automatically developing prediction models based on training data. Think of algorithms that are able to recognise objects (machine vision, image recognition) or speech (NLP, voice recognition). These algorithms can then be used within services to scalably offer smart functions to the end user, such as a voice assistant.
Algorithms are trained on data sets and the services a digital stack provides to us depend on the availability and usability of data. Every question we ask to our home speaker is a bit of information that further improves the artificial intelligence behind Alexa or Siri. The availability of data also has become a service in itself: documents or music downloads are on-demand available to us where-ever we are. Data is the oil of the Stack and a digital system is therefore meant to store, send, process and present data. In the data layer, we define the precise nature of the data that acts as the fuel for the intelligence and services layer. For example, on our smartphone, the metadata of our photo albums makes it possible to search them with incredible preciseness. A digital system may work with personal data (behaviour, emotions, features) or more contextual data (weather, location, time) or data from more abstract actors (company data, government data). Furthermore, this layer also looks at the volume, variety, reliability and validity of the gathered data in a system. These features in turn determine the quality of smart algorithms that are trained on these datasets.
Below the data, we find the infrastructure layers. Our smartphone runs on a top of virtual hardware such as cloud software and internet protocols. Much of these technologies are not localized ‘within’ the smartphone but show us that the smartphone is always embedded and integrated into wider computer networks. This layer makes sure, for example, that messages between devices are sent on a reliable way and your newest Netflix favorite streams frictionlessly, even if your connection is fragile from time to time. On this soft infrastructure layer, we find the modular software building blocks that relate to the direct control, connection and virtualisation of hardware (e.g. firmware, network protocols, kernels, operating systems and middleware), the development, management and use of databases, the organisation of the business logic or the way information is eventually presented to the user (presentation layer or front-end). Aside from the fact that this layer provides the virtual semi-finished products for software development, the codified administrative rules of software ecosystems that are built on top of it are also determined here. This also refers to the double meaning of soft infrastructure; although infrastructure is often interpreted as a very technical and uninteresting layer for politics, internet protocols, for example, are to a certain extent social agreements crystallized in digital code. This inevitably has a political side to it.
In addition to the soft infrastructure, the hard infrastructure layer consists out of all the hardware elements we normally define as infrastructure. For the smartphone, if we open it up and dissect it, we find all the different hardware elements such as the screen, sensors, battery, chips, etc. In a more general sense, think of hardware for storage (e.g. hard drives, solid state drives, magnetic tape), computing power (CPU, GPU), transmission (5G antennae, fibre optic cables) and measurement (optic sensors, microphones). Thus, this layer forms the rough computational basis for the possibilities in the rest of the Stack.
If we move to the lowest layer of the stack, we find the resource layer. Our smartphone is built out of basic materials combined with rare earth metals and needs a constant supply of electricity to keep working. Although digital systems can reach virtual heights, ultimately, they remain grounded in a material basis. Every system consists of certain resources, such as standard elements as steal, glass, silicon, gold, but also new materials such as graphene. Furthermore, in this layer we also include the use of energy and space. Although this seems far removed from the reality of apps, smartphones and social networks, this physical base partly determines the economic, social and geopolitical dimension of the Stack. For instance, strategic interests and energy costs play an important role in the location selection for server parks.
Together, the interdependence of these layers interacts with our cities, communities and institutions. Therefore, we added three layers on top of the Stack to emphasize the influence digital technologies have in transforming society and culture.
On top of the user interface, the smart habitat is the layer that points to our digitized environment. For example, the smartphone has become our wallet to pay in stores and functions as our biometric passport to enter a restaurant during times of covid. Hence, our increasingly smart living environment forms an interface between society and the digital Stack that facilitates those services, provides us with information and derives data from us and our activities. This makes our living environment a source of data, but because of digitalisation in various sectors and the addition of robotics, our living environment itself is becoming more dynamic and responsive.
The Stack does not only rearrange cities, it also creates new cultural practices and communities which we call neo-collectives. This holds for both the virtual and physical realm. As digital technology further pervades our lives, the impact of the Stack on social structures is intensifying. The smartphone has fundamentally changed the way we socialize, work, date, cook or shop. It has created new protest groups that originated on social media and developed vibrant internet communities such as those we see nowadays on Reddit or TikTok. Thus, from the Stack, and the digitalisation of daily life, new political and cultural collectives are arising, while these neo-collectives in turn (re)shape the Stack socially. As such, it is analytically useful to consider these collectives an integral part of the Stack.
Last, the smartphone has changed the way we communicate privately and publicly and this raised new questions around privacy, safety and public interests. Therefore, the top-layer of the stack refers to the new forms of governance that have been the result of the digitization of society. We call this layer neo-governance. As a response, we have seen new regulations such as the GDPR and the new set of regulations the EU is rolling-out in the next years. Because of new technology, new institutional structures are arising as well, such as digital forms of participation, decision-making, but also enforcement. This allows for the emergence of new models of governance that also pertain to governing the Stack itself.
As the example of the smartphone makes clear, the influence of the Stack on our cities and society is profound in many ways. Technology is not merely a means to human ends. Yet, this does not imply the model advocates a technological determinism, by making domains such as ethics and politics completely subordinated to the dynamic of the Stack. Although we do acknowledge a certain degree of autonomous behavior from the Stack and its digital systems, we do not believe human beings are completely determined by this dynamic but always influence and mediate the outcome of it. As an analytical tool, the Stack helps us shed a light on the interplay of technologies, that seem to have its own dynamic but also serve the interests of humans and organizations. Instead of mystifying these autonomous processes of technology or reducing technology to a mean, the Stack is a handy tool to illustrate how both interact.
Thus, the Stack resists the temptation to choose the side of human or technology. The reciprocity and co-dependency between technology and humans is what ultimately decides the dynamic of the Stack, making us sometime feel overwhelmingly powerless in the inevitable advent of technological processes and at other times empowered to use the Stack as an instrument to human progress and the good life.
In the following months, we will update this introduction of the Stack with more in-depth analyses. We will show how we utilize the framework to dissect the digital systems of nations, companies and industries. Stay tuned!
This article is an edited version of our report Outlook Digitalisation 2030. A more extensive discussion of the Stack framework can be found in the report.
