Mirror, M1rr0r


Data visualization is no longer news. We’re well past talking about the “tsunami of data crashing upon our shores” and the staggering zettabytes and yottabytes of data we’re capable of generating. Everybody knows we’re visual creatures and that visualizations are like an express lane for information to get into our heads. We get it. There is no longer the need for grand declarations of a now obvious fact: we’ve got lots of data at our disposal and visualization is essential to getting the most from it.

We’re graduated from the awareness phase into a full-blown data phenomenon (Data 2.0?), and we’re only now starting to wrap our heads around it.

Making sense of data has become a booming industry of its own. I mean, look at this meta-viz of visualization tools (as of this writing, it boasted 415 tools and books). There’s a growing data lexicon (open data, big data, data science, data journalism, quantified self, smart cities), an elite cadre of data celebrities (Edward Tufte, Hans Rosling, David McCandless, et al), a pantheon of data heroes (Charles Minard, Florence Nightingale, William Playfair, et al), and, naturally, loads of data debates and opinions about the art, science, and very definition of data visualization, scattered like confetti across the online landscape of blogs, news sites, and social media.

Data has also gone mainstream. Fitness trackers dangle proudly from people’s wrists; monitoring personal health performance is now no different from tracking KPI’s on a business intelligence dashboard. News just isn’t news without some visual entry point. We’ve come to expect that data isn’t just a static chart or graph any more — it must dance across the screen or invite us to play with it using fun sliders or scrolly parallax effects. Resumes are gradually ditching the stuffy serif-type-on-ivory-cotton-fiber-paper look and getting hip to the colorful-dashboard-o-data. There’s even a Kickstarter for a series of children’s data visualization books, complete with a Penelope Pie and Barnaby Bar Chart. And more data niches will no doubt keep being filled.

There’s so much we can do with data: mine it, interrogate it, manipulate it, visualize it, toy with it, make decisions with it, and just be dazzled by it. But what we haven’t properly done is examine the data phenomenon itself a little more closely. How do we step outside the data frenzy and critique the products of data visualization as well as the commercial, social, and cultural dynamics surrounding it?

The Medium is the Message

Critical reflection on data visualization isn’t such a new concept. Back in 2011, design writer and lecturer Peter Hall wrote “Bubbles, Lines, and String: How Information Visualization Shapes Society,” a pointed essay about the absence of critical discourse in information visualization (in this piece, information visualization refers to all forms of visual representation of information, from information graphics to charts and graphs). According to Hall, there lacked a deeper examination of the rhetorical meaning of information display, the subtle and not-so-subtle messages conveyed by the “aesthetic of administration” — the graphic equivalent of a suit and tie that conveys a sense of authority (and control) in business and government communications, such as conventional pie and bar charts, not to mention the sterile visual style and stark pictograms characteristic of public warning signs and instructions. He calls for greater scrutiny of the motives, purposes, and means of visualization: What do the acts of data selection and the techniques of representation say about the creator and their own agenda? What meanings do historical and contemporary graphic methods carry, in tandem with the content they represent? And how do we reconcile the roles and functions of different domains of visualization work: scientific, journalistic, and artistic?

(SIDE NOTE: Specifically with regard to information design, it’s important to mention Robin Kinross’ piece, “The Rhetoric of Neutrality,” published in the Autumn 1985 issue of Design Issues and cited in Hall’s essay. He points out that no work of information design can ever be “neutral” or free from some form influence or persuasion, however subtle or unconscious. Typography, color, and even the placement of leader dots in a train timetable can express a point of view about how something should look relative to the cultural and historical context. More on this in a future post.)

The Social Life of Data

What’s particularly fascinating to me is the social dimension of the data phenomenon — how people’s attention has drastically shifted towards data visualization and how public and organizational behavior reflects that. Gatherings around hot topics of the moment are rich ethnographic opportunities to observe how people think about and engage with data visualization, what motivates and appeals to the casually curious and the seasoned experts alike. I’ve written before about my experiences with the data phenomenon (public events and online discourse); over the past eight years of event-going, I’ve seen far more lightweight treatments of data visualization than deep, probing discussions that illuminated what it is and what it can do. Two recent events I attended have broken the mould, in very surprising and wildly different ways:

EVENT 1: Data Visualizations that Bring Data to Life

Without a doubt the most extravagant free event I’ve ever been to, this gathering took place on the 19th floor of Gansevoort Hotel. The venue was stunning: a lounge with two bars, a long balcony, and a stunning view of New York City. Entry into the space was by private elevator, and once inside, attendees created and printed their name badges on the spot. Drinks were free, and there was a smorgasbord of hot appetizers, prepared food, and even a sushi station! As if that weren’t enough, after the talk, there was a gourmet dessert station serving some creative reinterpretation of the Dunkin Donuts munchkin.

Why all the fuss? To begin with, the event was sponsored by Sapient Global Markets, a company specializing in business technology and consulting services for the financial and commodity markets. What was advertised as a networking event-slash-book launch for Visualizing Financial Data, written by Julie Rodriguez and Piotr Kaczmarek (both employees of the company), turned out to be a no-holds-barred marketing extravaganza to establish the company’s data visualization expertise and wine-and-dine some potential new clients. For the predominantly financial industry crowd in attendance (judging from companies identified on people’s snazzy name badges), data visualization is nothing new. What is new is the prospect of doing it better and in a more visually engaging way that provides alternate, richer views of the same data, by learning the advanced graphic design techniques outlined in the book. The learning curve for executing the techniques covered in the book is quite steep, so the more appealing option would be to — surprise, surprise — hire Sapient Global Markets to do all the work. 

It disappointed me to see an audience eager to make better use of data visualization (or at least understand its potential a bit better) get lured into a live equivalent of an advertorial, where the delivery of instructive information (the book) was just a means to commercially-driven ends (data visualization projects for Sapient Global Markets). Maybe I’m being too cynical, but in the ongoing gold rush around data visualization skills and services, companies that blur the line between education and marketing to take advantage of people’s needs and interests for their own benefit cause more harm than good in the long run.

EVENT 2: Visualized Meetup: Is Data Visualization a Social Phenomenon?

Part of a series of bonus events tied to this year’s Visualized conference in New York City, this “meetup,” held at The New School in New York City, remarkably aimed to address some questions I’ve had on my mind for a very long time. The handout for the event listed a few of them: “What is the urgent need to visualize everything? What is the strategic function of datavis? Why is datavis considered cool? Which are the consequences of such visual language?”

Originally described as a panel discussion featuring data visualization professionals Peter Richardson, Ben Rubin, Gaia Scagnetti, Marius Watz, and Chrys Wu, the event was changed (last minute?) to a room-wide roundtable discussion — among about 50 people. After hearing each participant’s brief introduction, some commonalities emerged: most were employed in well-known tech-related companies, were interested in getting “inspired” at the Visualized conference, and were looking to bring some fresh new ideas back to their companies. 

After introductions, there was a brief overview of some select concepts: a run-down of the core functions of data visualization along with philosophical aspects (for instance, the idea that data visualization is an “apparatus,” as described by Italian philosopher Giorgio Agamben and French philosopher Michel Foucault). Despite the strong academic focus of the opening talk, the group discussion that followed pivoted sharply into practical concerns (e.g., the importance of visual literacy, understanding different audience’s needs, and shortcomings of Microsoft products) as well as artistic interests (e.g., the regard of data as a raw material, data art as being absent of intentional meaning or sense-making function). Comments bounced from topic to topic, with some building on one another and others falling flat. All the while, notes were being typed and projected for all to see — a series of thematic snippets following the winding journey of the conversation. Oddly, hashtags suggestions were occasionally announced by one of the event volunteers who was passing the microphone around, for the benefit of any live-tweeters. 

By the end of the discussion, a bounty of thoughts and ideas surfaced. More questions were raised than answered. My takeaway from this event was that, given the tremendous diversity among data visualization practitioners and enthusiasts, and a wide range of skill, knowledge, and expertise levels, there needs to be a more deliberate approach to enabling productive discourse, knowledge sharing, and broader understanding for all stakeholders in data visualization. The open group discussion format was an interesting departure from conventional lectures and one-way presentation — it invited lively participation. However, not every voice was heard and not every idea was seized upon or captured. The synthesis and collective push for understanding that I had been looking for, across the many disparate threads of this perpetually sprawling conversation spanning years, was not to be found that day.

It occurred to me only afterwards that nascent ventures such as this into broad discussion of self-reflective topics in data visualization call for a little ambiguity, a few loose samplings of stakeholders to gauge the sentiments, attitudes, and interests at different places and points in time. This mostly unstructured group discussion served to explore rather than to delineate and define, and in the process, painted an Impressionistic portrait of what mattered to each individual person at that moment. It wasn’t the final word, but many words brought together in a forum of dialogue (one of hopefully more to come). My own interest in connecting the dots and constructing an all-inclusive picture of the universe we all live in is just that — my own interest. It’s probably okay that not everyone is after the same thing, or that not everyone needs to be after anything. Perhaps data visualization needs to have considerable latitude in how it defines itself because the people who engage with it come at it from different perspectives and offer something of their own to the mix.

Putting Seeing Before Critiquing

The biggest distinction we need to make today is between 1) understanding the scope of what is (people, information, events, etc) and 2) applying a critical lens to it (roles, purposes, meanings, consequences, etc). Mixed efforts are happening in parallel: different people see and engage in a slice of the picture, and many judge the general by the specifics. Certainly, there are those trying to bring clarity and offer guidance on different fronts (practice, education, technology, business, government) but as yet, to my knowledge there is no comprehensive conceptual map or framework of the data visualization landscape, or accompanying views into it that broadly help make sense of it.

The data visualization community should embrace its diversity and depth, but it needs to build a stronger collective sense of self-awareness and an ability to reflect on the internal dynamics and external context in which data visualization lives. There’s a whole world of activity, beyond the CSV files and D3 tutorials. There are human stories among the data stories. Some people are just starting out and are looking for guidance, while others are pursuing the next higher level and seeking out unexplored territory. And there’s everyone in between, participating in the data phenomenon.

Putting a mirror in front of data visualization isn’t about producing a perfect, static reflection. It’s about gathering all the data, capturing the whole picture and its dynamics, and using that macro view to understand how the pieces interrelate. Not unlike data visualization practice itself.

Leave a Reply

Your email address will not be published. Required fields are marked *