Bringing it Together
A Review of a Data Visualization Demonstration at Impakt Festival - Gathered at Theater Kikker on the 3rd of November, Born Digital gave a technical introduction to data visualization as part of Utrecht's Impakt Festival.
Gathered at Theater Kikker on the 3rd of November, Born Digital gave a technical introduction to data visualization as part of Utrecht's Impakt Festival. The center of this demonstration was formed by four tables placed in the middle of the stage crammed with laptops and cables with their operators buzzing around in preparation of the demonstration. However, this was no preparation. The presentation was in full swing. What followed was a quite diverse mingle of different data visualization techniques of which the most used what are called visual programming languages. In these kind of languages logic and mathematics are not reached by typing code, but by linking different logical or mathematical functions and their variables to each other visually in a diagram like fashion. What is striking about these visual programming languages is that not only what is produced is visual, the process of constructing these data visualizations is already visual.
In more serious applications of data visualization the main value of these visualizations lie in the value they add for understanding the object that is researched, however when looking at the more practical day to day use of data visualization the esthetics of the visualization are just as important. This became clear with following the different demonstrations. Decisions made for each object were not primarily led by the goal of a better understanding, but as one of the demonstrators said: “to make it look pretty”. This was the case in the demonstration where demographic information of the city Utrecht was used for creating a visual of these demographics and this esthetic premise can also be found in visualizations used by DJ’s for supporting the music they make. Recognizing this esthetic premise is not to say that making it look pretty is the only goal. Visualizations made during a DJ’s performance can enhance the emotional experience of the audience and the possibility of visualizing tweets can strengthen (or break) the bond between the crowd of that performance. However these applications are not burdened with the task of providing information understood rationally. Projects which set themselves goals like this encounter problems that lie at the core of what is understood as data visualization. Not only do the visualizations created need to be attractive, but they need to be understandable as well. In the case of another project presented during the demonstration – NOSop3 Lab’s experiment in visualizing information provided by all European press agencies and Reuters onto a single urban screen – this problem is the massive amount of information provided.
How are decision made to prioritize, order, visualize, etc. this massive surge of information without becoming a mumbo-jumbo of pictures, videos and text? Another problem is presented by yet another project: e-forecast (http://emotionforecast.com/). This projects goal is to give a forecast of the emotional state of people around the world by counting words such as happy, sad, etc. and through colors show these emotional states on a map. The problem encountered in this project is that by counting words, these words are taken out of context. Mentioning a word such as happy or sad does not necessarily express a person’s emotional state, but the message or sentence as a whole can have many other, even contradicting, meanings.
What can be discerned from these problems is that they cannot be solved by programmers or visual artists alone. In order for the practice of data visualization to develop, to go beyond only making things pretty, it is necessary to bring together a wide array of fields. From new media scholars who study the cultural impact of these visualization, to linguists who learn computers to understand human language and cognitive psychologists or neurologists who help understand the way we process information. Not to mention statisticians,visual designer, user interface experience experts, information modelers and so on.
In the case of data visualization it is not only necessary to ascertain the role of a careful listener, but to actively engage and to show each other – all those different disciplines that are involved with the enterprise of data visualization – where it hurts and together set out the direction for possible solutions. So instead of seating ourselves as a passive audience watching to beautiful visualizations, we should be buzzing around those who have the technical skills to create data visualizations just as they buzz around their beloved laptops.