Space Making with Data

We live in a world where petabytes of data are being produced by various satellites and even our smartphones on a daily basis. Such data deluge has forced many of us to contend with a massive, heterogeneous, and dynamic volume of information, extensive portions of which are difficult to understand and analyze in their raw format. Think of an endless stream of letters and numbers intertwined in less than legible ways but in enormous quantities. However, the integration of human judgment and visual representations of data turns such data overload into an opportunity by transforming it into graphical and visual representations, known as data visualizations. Where understanding data in its raw format requires special skillsets, data visualization mechanisms tend to systematize, categorize and simplify info into visual bites that are easily digestible at a glance (See Figure 1). Data visualization allows a broad range of users to understand information concealed within the data by providing mental models of information. An appropriate method of visualization improves users’ cognition in the perception and exploration of data, and helps to improve the efficiency of information retrieval and memorability of it.

However, the history of visualizing data in an easy-to-grasp method dates back to long before the invention of writing and papers: a time during which people used physical forms to record information (See Figure 2). Displaying data through the geometric or physical properties of an artifact leverages perceptual exploration skills, helps with information retrieval and increases the memorability of data when compared to similar designs shown on screen. Such physical artifacts inherit all of the practical and social advantages of everyday objects: they can be explored through touch, carried around, be directly manipulated, or even possessed. They can also address some of the limitations of visual or virtual displays, most importantly the dimensional mismatch between three-dimensional representations shown on two-dimensional displays (e.g., using flat maps vs. curved globes). Such physical models, once paired with a three-dimensional space, can influence spatial perception skills and expand the public’s exploration and understanding of critical and complex data via the inclusion of data in sculptures and architectural installations.

As computational methods continue to develop, architects are taking advantage of the many data streams available to them during the design process. Using computational design tools to generate novel architectural forms provides interesting spatial opportunities and bridges the field of data visualization and architecture. Such combination forms the basis of Data Spatialization.

The Data Spatialization term can be perceived in various methods. For instance, one may use specific data (e.g., a company’s sales data) to generate interesting and eye-catching forms and build a pavilion with it (See Figure 3). Another representation method for such an approach is making forms that visualize the data itself (See Figure 4), which is more relevant to the original definitions and goals of data visualization, i.e., making data more understandable. One last approach to tackle Data Spatialization can be using data to generate architectural forms and design the forms in a way that they can represent the underlying (i.e., the input) data. Such a representation does not require its viewers to employ complicated methodologies for decoding and, thus, understanding the data. As a result, this latter method has very high potential to make an intriguing architectural space in which spatially-visualized data is accurate, dynamic, and easily recognizable by its audience.  

The prime objective of designing such spaces is to keep the input data as legible as possible within the built environment while allowing people to interact with the space. The Anamorphic Data Spatialized Pavilion is an example of such an approach. Architectural spaces similar to this pavilion can define a notion of playful urbanism, where people feel there are discoveries to be made. Such a design approach can reach territories beyond a combination of a simple gimmick and data visualization and can become Data Spatialization. It encourages the audience to discover spatial agencies and orientations and interact with its audience and its environment.

References

1 – Djavaherpour H. GEOPHYS: Design and Fabrication of Geospatial Physicalizations.

2 – Hosseini SV, Djavaherpour H, Alim UR, Taron JM, Samavati FF. Data-spatialized pavilion: introducing a data-driven design method based on principles of catoptric anamorphosis. International society of the arts, mathematics, and architecture, summer. 2019 Jun 17:39-51.

3 – Djavaherpour H, Samavati F, Mahdavi‐Amiri A, Yazdanbakhsh F, Huron S, Levy R, Jansen Y, Oehlberg L. Data to Physicalization: A Survey of the Physical Rendering Process. In Computer Graphics Forum 2021 Jun (Vol. 40, No. 3, pp. 569-598).

Sam Djava-Hehr