What’s Behind the Cloud?

what's behind the cloud webpage screenshot

The way we understand the internet today is blurred by the idea that the virtual world is indeed only virtual and omnipresent. And this is a problem: when people believe blindly in something, we delegate a lot of power to the structures that make this  network possible. Behind the cloud is the  result of my research on the boundaries of the internet and the infrastructure of natural resources, human labor and data that lies behind it. In sum, this project addresses and personifies this “Myth of the cloud”, aiming to make this information more accessible and interactive in order to help users question the status quo of the technology that shapes the way we behave today.

Click in the gif below to access and interact with the project.

Below I will describe about the process and references that led me to create this experience.

 It all started with a different question…


For the past couple of years I have been working in the tech industry in different roles, towards the path that led me to the designer and technologist that I am today. I’ve always been passionate about Human Computer Interaction and amazed by the democratic aspect of it. When creating an UX experience, usually for websites or mobile apps, the first thing I think is: “Who am I designing for?”, and do my research to make the best interaction possible accordingly.

But…what about who I am NOT designing for?— what about the people that don’t want to be or that are not online today? Have I ever really researched about them? Isn’t it important that as a designer, and mainly as an optimist person that believes in the potential of the internet as a democratic tool, to really understand the limitations and boundaries of this technology? I thought so, and this idea was stuck in my head.

Consequently, I decided to take this question as the starting point of my research during this semester.


By starting to research through reading a couple of UNICEF and World Economic Forum reports on internet penetration reach and also by talking with fellow students at ITP that are very in depth with the research around data privacy, I started to map the world outside the mainstream web, trying to identify the existing boundaries of user presence in what we understand as the internet today.

This mapping exercise made me realize the two main extremes in these spectrum:


The reality of not being online

Basically 50% of the world is not online. Even though this is very well known, this number is very striking for me.

We act as all the information available online represents all the information on the world. And this is not true. We do map most regions and content creation and access is very well spread around this connected half of the planet. But, besides the information we have access to is mainly shaped by the 5 big platforms/companies players in the market (Amazon, Apple, Google, Facebook, Microsoft), it is also very biased as the other half of the planet doesn’t take part in this virtual world.

Of course internet access and penetration are a result of other gaps, such as economic, education, health, and so on – which is a reflect of these countries and regions being also excluded in these other spheres of the global society. It is not a novelty that we live in an extreme unequal system. As Eleanor Saitta puts

“All technical problems of sufficient scope or impact are actually political problems first”.

What is interesting for me, is that this fact is not highlighted when we think about the computation ideal of big data studies and when we create this myth that we can replicate a perfect model of our planet when analyzing online behavior — we can indeed, but first we have to acknowledge that this accessibility gap also exists online.


The ideal of not being online

On the other extreme of this unequal world, we have the rising awareness in data privacy, and the ideal of changing what the mainstream online is. The idea of a decentralized internet where information is not stored in the main data centers of the companies mentioned above and that also can live free from those platforms. The idea of creating open source decentralized products and networks that go beyond of what we understand of the mainstream World Wide Web, the regulations on data are also extremely important. Those issues and laws are usually raised  in countries where internet access is widespread, mainly in Europe and North America, which again, very much relates to the quote from Eleanor Saitta. As long as we live in this capitalist system, it seems natural that the environment that we create online will also replicate this reality.

Understanding that these extremes, I decided to make some actual maps where I could visualize them. Hoping that they could give me some insight on my research. Therefore, I dived into these reports data, and created the maps below.




The infrastructure of  the internet

Parallel to this research, and as a reflect of understanding that these extremes in the virtual world are a reflect of our real world, I decided to explore why we as creators and users perceive this as separate realities while they are so much of the same. Where does these disconnection between the physical and the virtual begins?

One amazing reference was Kate Crawford’s work, mainly on this article: Anatomy of an AI system.The author tries to map every aspect triggered from the moment an user asks their Amazon Echo to turn on the lights. Everything then, from the mined resources, to the workers, data centers, the data collected from that user too, and the e-waste that the amazon echo material will soon be, are also triggered and unveiled in the text and in the infograph created to illustrate it.



Also, a great reference at this point was the book the New Dark Age, by James Bridle, where he mentions the connections between submarine internet cables and the colonial routes from the colonial period. Which – by going back to my map-dataviz experiments started to play a bit with.



From this part of the research on, I understood that maybe “Who is not online?” wasn’t a good question for my project anymore, but something that related me more towards the search of mapping what was the infrastructure of being online meant.



What’s Behind the cloud?

Why do we relate with this virtual world we created as it is detached from our  physical  reality? Why do we see technology – as James Bindle puts – “…this leveling force, that makes us more equal and allow us to make better, more equitable decisions”? Why do we believe so blindly in the algorithms and see it as something magical?

Internet as a Hyperobject

The idea of computation thinking and the scope that internet took as an Hyperobject, enabled by our systems and by the reinforcement of the way platforms today are designed to be – minimalist, functional and wireless – created this faith in the network that it is this magical far away place where all this stuff we use  just happens and comes back to us. But the reality is that it is super physical, we work creating it, it needs natural resources, energy, data, and it generates chaos, and waste and heat… and so much more.

It is indeed overwhelming to think that in less than two decades we went from the physicality of having to dial to a line…



…to this Myth of the cloud we created today.


The term “cloud computing” was popularized with Amazon when the company released its Elastic Compute Cloud product in 2006.

The main enabling technology for cloud computing is this said ‘virtualization’, when it separates a physical computing device into one or more devices, each of which can be easily used and managed to perform computing tasks. With operating system–level virtualization essentially creating a scalable system of multiple independent computing devices, idle computing resources can be allocated and used more efficiently.

In this sense, the idea of it being virtual is that, instead of stored in a local data base, it would be decentralized in the multiple databases offered by the companies that sell this storage service. It’s still physical, but yes: it works faster and with the technology that enabled bandwidth, 4g and more internet speed, we can rely on it to get and send these files from machine to database, having this seamless virtual feel.

Devices get smaller, more minimalist,  and the processing and storage power can happen elsewhere. In the “cloud”. The companies and systems, are leveraged by reproducing this idea.

Digital Literacy

In the video below, you can check a recording of me interacting with my Google Home.


As a designer and technologist, my google Home’s responses to the questions “Who are you?” and “Who made you?”, bothered me a lot. They are simply not true.

The myth of the cloud does a service on behalf of those that want to keep our world unequal, to make people unaware of environmental impacts and global warming, to keep profiting with our data and to keep reinforcing biases and maintaining the same structures in power in our society. It is a weapon against critical thinking and misinformation.

And, as Google didn’t do in this case, we should not build systems intended to trick and surprise people but that they are actually involved in every step in educating them.

There are indeed several articles, books and documentaries that show the reality of the infrastructure that makes what the internet is today, but, on my references and research, I couldn’t find any  interactive projects that had a different tone to break this idea of the internet as an Hyperobject.


Developing the Project

With my research as the backbone, and using as a base the content studied from the projects/books/articles of Benjamin Bratton, Kate Crawford and James Brindle, I was aiming to create an interactive webpage that turned this more academical information  educating the user in an interactive way using little chunks of information and visuals.

I made some design decisions such as using Voice/Text to Speech, creating a loop and going for this this visual aesthetics for some specific reasons.

The idea of activating it through voice came mainly from the “Anatomy of an AI System” reference and of course, by the video that I showed above, on my interaction with my Google Home. I believe that by the act of asking something from our machine we humanize it in a different way, mainly when the question has its “philosophical” irony in it. The Speech to Text capability comes from a similar standpoint, to subvert the narrative that we are used to hear from machines – very functional, straightforward answers – and adding this irony of the machine revealing its true self.

The loop (and the idea of a loop that doesn’t show any progress bar or idea of user status) comes from the basic acknowledgement that even though I made my research it is still the very tip of the object: the structures behind the cloud are indeed multiple, and complex, and I don’t want to pretend I will be able to map every single aspect that is part of this system. Besides, the idea is to make the user feel as there is more to be discovered, not to overwhelm or actually map everything, but to arise the interest and believe that after going through 2 or 3 loops people will be more skeptical and rethink about the Myth of the Cloud.

This aesthetics look and feel came from the idea that I wanted to create something “brandless”, but that still made users rely on it somehow. I thought about going with some more gamecky or “hacker”/terminal approach, but I though it would lose some of its seriousness. I used this website as a visual reference.

You can check the code on my Github repo here. And enjoy the project using this link.




Mapping the unfindable: the Archipelago of Disconnection

This research has the aim to map what is unfindable on the internet and identify how online presence is distributed. Is internet a privilege?A resource? A need? A curse? What does the internet represents today both for the 50% of the world population that have direct access to it and mainly to the 50% who doesn’t.

Who is not online?

I explored the idea of identifying who we can’t find online and the systems that surround this situation. I talked to experts and people that had done almost anything possible to erase their names from search engines and social media, I read documents and listened to podcasts on the dark web, the right of prisoners to be online, asked around for people to google dead relatives from past generations, and, of course, read a lot of UN and Global Economic Forum reports. With that, I arrived in some specific groups that either are not online at all or have a reduced online presence when compared to the ‘average secular human’ today. These groups are:

  • Data Privacy Advocates;
  • Criminals that don’t use the mainstream web;
  • Dead people from past generations;
  • People that are in a social situation that does not allow them to be online (incarcerated prisoners, Orthodox Jews, Orthodox Mormons);
  • People that are in a social-economic situation and region where there is no access to the internet whatsoever.

By identifying these groups I aimed to create a less blurry image of what internet represents in the world today.

My classification is not complete and I do not have a quantitative stat of how much each specific group represents in numbers.  A lot of times I reached the question of “what’s the difference between not having access to the internet and not being able to be found on the internet”. As I could not answer it, I decided to understand that the intention of not being online, even if idealized, was already somehow part of a movement that I should include in my research.

I became interested in analyzing the paradoxes I saw by these classifications.  There is the obvious fact in my analysis I created a unique group that includes both religious people and incarcerated people – which is very very questionable, but they do usually live in developed regions where you could have internet access but still do not due to either a social imposed restriction or a shared value. And, most importantly, there is the fact that in the spectrum that comprehends the ideal of not being online (Data Privacy Advocates) and the reality of not being online (people living in regions with no internet access) there is such a huge gap of development, resources and education that somehow illustrates the inequality and the extremes of the world we live in.

And, as I saw the UN and World Economic Forum reports, I was stuck with the idea that more than 50% of the world doesn’t have internet access. I live in such a bubble of tech-savvy people, designers, technologists, developers that we never even stop to think about it. I mean, at least I didn’t. And, as I asked people from my school and work here in New York and even friends in Brazil, we have this generalized idea that internet has been fully democratized — for better or worse. But that’s simply not true. Owning a smartphone, having broadband and actually having simple internet access is still a privilege.

By that, I do not mean that the work and fight for Data privacy rights is not relevant today. On the other hand. We are still on the edge of discovering how much our connected devices keep track our information and can directly influence society in a scenario where companies and governments have total access to our personal data. There is a need to regulate and educate internet users towards it and find ways to regain our data privacy. In truth, these paradoxes are a result of the same dynamics of power that feeds itself and this world of extremes.

Still, for the purposes of focusing my research, I decided to explore more about the non-online world.

The Archipelago of Disconnection

By keeping in my mind that in fact more than half of the world doesn’t have access to the internet, I wanted to see through some visualizations what does that represent. Still, I couldn’t any that would give me a satisfying idea of that, as you check below.



Using the data from the ICT Indicators Database  I started exploring some ways to visualize that with the Mercator projection, you can access it by clicking in the GIF below.

As you can note I inverted the default structure of how these maps show data: instead of highlighting the amount of online population I showed the amount of people that lacked internet access. Also, I think that the hovering interaction is more playful and easier to compare. Try doing a mouseover in Australia and then make a mouseover in Madagascar. I think it shows a lot of the paradoxes I talked about earlier.

Still, I felt like exploring this idea of map, by highlighting even more the countries with less than 10% of internet penetration, territories that are largely left out of global digital connectivity.

Therefore I created the interactive map below of the “Archipelago of Disconnection” (you can also click in the GIF to interact with it).



The map highlights an archipelago of land that is mostly disconnected from the internet and thus largely barred from participating in the cultural, educational, political, and economic activities that it affords. This archipelago of disconnection has its centre of gravity in the Sub-Saharan Africa where 18 countries have Internet penetration rates beneath the 10% . Among these Sub-Saharan countries with very low connectivity there are some very large populations – such as the Democratic Republic of the Congo (68 million),  with an overall Internet penetration of a mere 4%.

After these experiments where I looked through the numbers on internet penetrations provided by International Organizations by mapping its data, I realized that I wanted to approach and understand this information both from the data creation perspective (maybe gathering the data flows and understanding the amount of information online generated from these territories) and add a more qualitative and sensitive feel to what I wanted to create as a result of this research.

With that, I remembered of the book Invisible Cities by Italo Calvino and wondered how could I connect this Fantastic realism of the narrative and the Invisible Online Territories from my Archipelago of Disconnection.



Therefore for the next step of my research I will be exploring the data available online on and from some of the cities inside the 18 countries with less than 10% internet penetration rates,  trying to create a new way to visualize and highlight this “hidden information” using Italo Calvino’s Invisible Cities narrative style as a base – maybe  even creating a re-reading of the book.



Learning D3 and the art of making ugly/bad charts

I started using D3.js last semester for my Impossible Maps final  and I want to learn more on how to use this library to create Data Viz projects. Thus, one of my goals during this semester in our Data Art class, is to explore more of it.

Therefore, when the teacher asked us to create three different visualizations for the same dataset for our first assignment I started my D3.js crash course.

In order to understand the basic commands, I started follwing these YouTube tutorials, that I thought are very good.

I made some really ugly and random drawings that reminded me of my early P5.js days.

From there, using the dataset provided, my very little new acquired knowledge of D3, and some examples available online, created 3 basic (and ugly, and probably not very user friendly) charts to visualize it in different ways.

You can check the code in my Data Art/Week1 Github repository and the examples below.

Bubble Chart



Bar Chart



Not Very Understandable Pie Chart



As you can see there is a lot to improve!!! But that was only the beginning 🙂

I’m exciting to keep exploring Data Viz, Data Art and the D3.js library.