Introduction

The internet is a digital infrastructure at scale. Let’s unpack that for a moment.

It is digital, because it moves bits between devices.

It is at scale, because it moves these bits between billions of devices ranging from websites to mobile phones to robot lawn mowers. Estimates are in the tens of billions of devices. That boils down to multiple devices per world inhabitant. That is impressive technology, but what keeps that together is the infrastructure part of it. That is what defines the relation between the consumers and the providers, and that is what we are exploring here, as I will show you later.

Providers, consumers, automation

Now digital infrastructures at scale - there are a few very essential elements in there. The most important elements are: one, that you have a separation between a provider and multiple consumers; and two, that that service is digital, which means that it’s amenable to automation and therefore scaling. These two together give a really interesting socio-political economic dynamic. At its core, it is about how much autonomy each actor is willing to claim , or let go of, as the case may be.

And that’s what you need to understand if you want your shared service to keep operating - or if you don’t want it to operate, but that’s a different story.

Better understanding digital infrastructures at scale allows us to better navigate the digital world at large.

The same pattern, everywhere

So, what are some of the characteristics of infrastructures? What is it that I am calling a digital infrastructure?

We are taking the internet as an example of a digital infrastructure because most people are familiar with it.

Social media is another example, and one that uses the internet as a component. Similarly digital marketplaces are an example. In the same way, the internet is also built on many other infrastructures: telecommunications, DNS, Number authorities. In the realm of IT management, there are many more examples of shared services. And even though failures in shared services projects are less global headline making, they do impact lots of organizations and their clients.

For example, in my professional life, I have seen many projects aiming to improve collaboration unnecessarily delayed because of a mismatch between expectations and what the technology and power structure could afford. There will be examples of that throughout the book.

Because there are so many examples of digital infrastructures, it makes sense to explore the commonalities. Looking at it through the lens of infrastructures will make it easier to understand use cases, abuse cases, and how to define, create, control, and manage things like the internet.

Value, power, and risk: all up for grabs

How do the examples that I have mentioned deliver value to their users and other stakeholders? What are the mechanisms and principles in there? If we know a bit more about that, we could better engineer that value and the way it is delivered.

What are the power structures around digital infrastructures? Who are the stakeholders that influence them, or are influenced by them? This cuts both ways: it brings value and risk to the table. Stakeholders typically want to increase value and reduce risk to them. But interests may also be conflicting, and resources are never infinite. With more understanding of these power structures, we have another handle on value creation and risk reduction.

What are the risks associated with digital infrastructures, and what are the ways to organize our management of them? Obviously, there is a vast body of knowledge around that, for example in the space of cloud computing, and to an extent also around artificial intelligence. I am not going to treat all of that, instead I want to bring out some core principles that have helped me, my clients, and my students to navigate that.

What this book is not

This is hardly a book on digital technology, but more on the organizational and societal impact of managing this at scale.

For a better understanding of digital technology I highly recommend Brian Kernighan’s “Understanding the Digital World”. And if you want to have fun while learning how to understand software, I’d highly recommend “The Nature of Code” by Daniel Schiffman.

I don’t have final answers to all the questions that I can pose, and I cannot afford to wait until I have them. So what you’re reading is where my thinking stands today.

Less is more

As you will find out in this book, I am putting on some very specific lenses to look at the world. You may find them restrictive; you may find them arbitrary, you may find them opinionated. Yes, this specific way of looking at the effects of digital technology has its limits. However, the essence of understanding is to be able to extract the core elements out of something. Paradoxically, only by restricting the way we look at things can we see more clearly. Cut through the noise. Like Goethe said: “In der Beschränkung zeigt sich erst der Meister”, though he applied that to a different subject.

I have a very limited perspective of the world. There are so many things and concepts that I just fail to remember or understand. And yet, that limitation has served me well. It has allowed me to focus on core principles that you may not see if you take in all the richness and idiosyncrasies. My limited perspective can serve you as a map. And you know that every map is not equal to the terrain it covers. In fact, the purpose of the map is to extract only the most relevant features of the terrain. How good a map is, judgement on the quality of the map is not objective. It is dependent on what you use the map for. A railroad map is close to useless if you navigate by car or boat. That is why you should consider the models and the perspectives that I give you as maps to guide your journey. If they bring you where you want to be, great! If not, get a different map.

How autonomous can you afford to be?

As you will see, the core elements of the models I find useful revolve around interactions between autonomous agents that process information, whether they are providers or consumers of digital services. These agents typically have limited (bounded) capabilities. This is similar to how maps are a means to reduce attention requirements. These agents can be humans, software or AI agents.

As I believe that you are an autonomous agent as well, it is up to you to decide how useful these models and stories are for your practice. You will not find answers to every question that you have around digital infrastructures. Where possible and relevant I will point you to other views, stories and explanations. At the same time, you may find my maps to be also useful for domains that are outside the digital. You should not read this book because I tell you to. You should read this book because you are convinced that it is likely to give you significant insights into a world that matters to you.

Autonomy plays out in provider/consumer relations, it plays out in how you look at the world, in the promises you expect from other actors, and the promises you make. More on promises later. The big question now becomes: how autonomous can you afford to be in today’s digital ecosystem?

That is what this book is about.

The stakes are real, and so is your role

Digital infrastructures, and information technology at large, have moved from a fringe niche technology to becoming an essential part of modern life. Most companies would go belly up in weeks, if not days, if their digital infrastructures would fail. Later we’ll look at big companies such as Maersk and Colonial Pipeline who were close to not surviving large cyberattacks. We’ll also look at how society and elections are influenced through cyber attacks.

What I write about in this book is shaped by the conversations I had with many people. I have worked with IT consultants, risk auditors, and their clients. I have worked with digital business owners, policymakers, IT architects, CIOs, CISOs, and boards of large organizations. I have also worked with researchers, teachers, and other scholars, college students who were one third of my age, and whose parents weren’t even born when I wrote my first computer program.

Giving these people some tools to use for their work would be gratifying.

So if you identify with any of these, or the topics I mentioned, this book is for you.