A WEEK OF TRASH,
​
Data is everywhere...
Data is everywhere: it categorizes us, informs our decisions, and reveals trends in our world that were previously unseen. In the age of Big Data it is also, in many ways, one of the most valuable resources on Earth. Discussions about personal data being sold at exorbitant prices for targeted advertising purposes dominate the industry. Now more than ever, human beings and their lives are constantly being quantified and categorized, and many times for profit. As people begin to feel more and more like dots on a scatter plot, the movement to bring humanity back into data science is more important than ever. The movement, coined “data humanism” by frontrunner Georgia Lupi, aims to veer away from the pie charts and bar graphs of old and move towards data visualizations that hold within them something uniquely human. The visualizations find themselves at the intersection of data and art, the complexity of the data becoming a somewhat narrative experience for the viewer.
Lupi and fellow award-winning information designer Stephanie Posavec maintained a year-long correspondence in which both parties collected and visualized data about their everyday lives on postcards, eventually publishing their work in a book called Dear Data. On the Dear Data website, the authors write, “Instead of using data just to become more efficient, we argue we can use data to become more humane and to connect with ourselves and others at a deeper level” (Lupi and Posavec). Their visualizations were all hand drawn, imaginative, and contained layers of information which all served to reveal their unique personalities and experiences. Furthermore, their project speaks to the ability for data to achieve more than representation. Their visualizations went beyond the data itself because, in Lupi’s own words, “Numbers are never the point.”("How We Can Find Ourselves in Data")
The Emotion Archive is a project by McKinsey and Company which aimed to research and showcase how people from across the globe emotionally responded to the COVID-19 crisis. Respondents’ feelings were categorized into 16 key emotions, then represented on an interactive map where viewers can read excerpts, see breakdown by country or emotion, and watch the original respondents’ videos. The resulting data visualization is an example of how thoughtful, artistic data visualization is done while keeping humanity at the forefront. This project utilized UX design and movement to give the viewer their own agency in exploring the visualization. It allows for the viewer to make their own meaning and tease out a narrative that is important to them.
The history of data visualization is just as important as its cutting edge. Florence Nightingale changed the landscape of data science in the 1850s when she published a rose chart to advocate for better conditions in war hospitals. Not only did she go against the norm of the time that data was best represented in rows and columns, but she used her visually striking graph as the main support for her argument. (Methot)
Inspired by these projects, I aimed to interrogate different visualization techniques to show how the form that data takes can give it new and layered meaning. I wanted to create something that revealed how innately personal data is, and expanded narratively beyond the face value of the numbers. Specifically, I wanted my resulting visualization to shed light on a topic that held particular importance to me, following Florence Nightingale’s lead using data visualization as a tool for advocacy.
PART ONE:
DATA COLLECTION
Following Dear Data’s precedent, I decided to record something out of my mundane, everyday life. Given my interest in sustainability and fascination with the no-waste movement, I decided to track my trash for a week. My friend then collected her data in a similar manner, to achieve a similar correspondence like Lupi and Posevec.
I collected my data early September, writing down each item I threw away along with the item's weight, and whether it ended up in trash, compost, recycling, or being reused. Given that food waste, yard trimmings, metals, plastics, and paper products make up 77% of the total waste stream in the US, I sorted my data into the following categories: plastics, metals, glass, food, and paper/cardboard. (DeSilver)
Having originally collected my data on a couple of scraps of paper, I thought it would be best to next transfer my data into a spreadsheet software like Excel. As I did, I refined and shaped the data in a way that I felt would be the most helpful. All through the process, I had a hand in how the data turned out. It is for this exact reason that many data scientists argue that there is no truly “raw” data. Even those scraps of paper were categorized and laid out in ways of my choosing. There is not a single point where data is completely objective and separate from human infallibility.
PART TWO:
VISUALIZATION
I began the visualization process by creating designs that you might see in a textbook or at a board meeting. That is to say, I used the default Excel graphs.
Due in part to the sophistication of the program, they were straightforward and simple to create. All it took were some formulas and a few clicks of a button, with Excel doing most of the analysis; recommending charts and graphs based on what its algorithms detected. While time effective and accessible, there is something lost in the experience of creating an excel graph.
As we look at the above graphs, there is a limit to the information that we can glean from them. We can tell that I use more plastic than Claire does, and that she does not compost. We can tell that we both throw things away most often, and we can make other concrete comparisons. But as you look at each, do your eyes start to glaze over? How easy are they to distinguish from each other, let alone any graph you might come across?
More importantly, where is the story?
Because the reality is that the data that I collected has a story within it. But how could I effectively tell that story if my audience can not visually distinguish my visualization from the thousands that they have seen before? And how could I communicate a meaningful narrative about something as personal to myself as my own garbage? A unique, personal, human story deserves a unique, personal, and human visualization.
It was clear that a story was missing from my first attempt at visualization. I was at a point, now, where I could decide what story I wanted to tell. What did I want my viewers to leave understanding or thinking about? As I began sketching out ideas for a visualization that would pay homage to Dear Data, I immediately noticed a difference in how I interacted with the data. The numbers on my computer screen felt nebulous before, but as I manually analyzed the information for the visualization, I felt a deeper understanding of and connection with the data.
The resulting visualization shows a portrait of two college students and their habits.
The first iteration of my postcard included the symbols ordered in rows symbolizing the days of the week. It communicated a sense of chronology and emphasized the daily count breakdown. However, it was difficult to compare the size of the bubbles, and as a result lost the gravity that seeing the large bubble representing a heavy glass bottle might have when compared to the small, red bubble of a food wrapper.
I then created another set of postcards, this time grouping the elements by material and forming an amorphous shape. The proximity of the elements allows the eye to create comparisons of size and color easily. But it lacks the diary-entry feeling of the first visualization.
First iteration of the data"postcard"
PART TWO-POINT-FIVE:
VISUALIZATION, AGAIN
My data "postcard"
Claire's data "postcard"
Looking back on the creation of this postcard (and its many iterations) I feel I missed an opportunity. One of the things that made Dear Data so appealing to me was a sense of visual honesty, which I now believe owes a lot to the visualizations being hand drawn. The artificially perfect lines of Adobe Illustrator add another degree of separation between the data and the human. If I had more time with this project, I would create physical postcards with the same visualization, only drawn by hand.
For my next visualization, I began thinking about the future. My dataset really only accounts for one week
-- but an object’s life, in reality, is much longer than that. Very often, waste is thought of as “out of sight, out of mind” but my old egg carton doesn’t just disappear when I toss it in the recycling bin. How could I show the future of my waste visually? I decided on a map.
My objective was to confront the viewer with the reality of where my trash ended up. The items I reused are still sitting in my cupboard, the countless wrappers I tossed in the trash are sitting in a landfill. By representing these locations graphically, it is my hope that the viewer considers the future of these items, and not just their existence. And I made sure to include that these locations are not stagnant either: they continue on, ad infinitum, changing and evolving as time progresses. To account for that aspect, I made the illustrations into gifs, looping through their cycles of change to simulate the progression of time. My hope is that the viewer begins to think about the context and real-world implications of my data. It is no longer a comparison of weight or even amount of trash, it is all about that trash’s life in the present-tense.
In the visualization’s current state, it only accounts for the data that I recorded personally. The next steps of the project would be to include Claire’s data among my own, to comment on the impact that two people can have on their environment.
CONCLUSIONS
Clean cut, automated Excel graphs have their place, and are by no means a bad visualization tool. They are not, however, equipped to serve the one-size-fits all function that they have historically been used for. They may give the illusion of precision, but they lack the ability to communicate the context and the narrative of a given dataset. As Lupi writes, “Failing to represent these limitations and nuances and blindly putting numbers in a chart is like reviewing a movie by analyzing the chemical properties of the cellulose on which the images were recorded.” Humans are not infallible and our world does not easily fall into neat, ordered rows. In many instances, the “messier” data is, the more complete the resulting image, as it accounts for the contextual messiness of our world.
The fact that these visualizations are unfamiliar to most people and we are not taught to read them in elementary school, is precisely what makes them so effective. We are forced to dig in and invest mental energy to understand the data. The design shocks people with something new and unfamiliar and beautiful to grasp their interest, and the meaning that they do grasp is fuller and deeper than anything Excel might automatically generate.
Although big data may be a source of fear and valid concerns, the future of data visualization is still exciting. With more and more aspects of our daily life relying on visualization to communicate (sleep trackers, smart devices), the opportunities to rework our ideas about data are endless. As reflected upon by Elijah Meeks in his article, “It’s Official: Data Visualization Has Gone Mainstream,” data visualization is becoming less of the exception and more of the rule in our society due to the rapid growth of technology. Growth is just an opportunity to reflect, and as Meeks remarks, “What we need to do next year is reexamine all our preconceived notions about what makes data visualization good and how to achieve it.” My project was an attempt to do just that, albeit on a much smaller scale. Moving forward, embracing data humanism allows designers to focus on how to use data as a tool for advocacy and empathy (in Meek’s words, we can make data visualizations “good”) rather than simply a tool to generate profit.
​
