Building with Data: Bridging Architectural Design Practices and Information Visualization

Our work seeks to augment new information visualization research with strategies and workflows from the fields of design and architecture. To this end, this research explores how to adopt tools and methods that can integrate the best of physical and digital modalities to multiple contexts and scales in HCI and data visualization. Designing information visualization systems creates a need for a design approach that addresses and ties together two main threads – 1) how we as humans interact with and make sense of our environment and 2) how we as designers create meaning through geometry, form, and material encodings. While the research community within data visualization has primarily focused on screen-based data visualizations, there is now an opportunity to study how we can create insight with hybrid physical and digital representations of data through the lens of architectural practice. My colleagues and I have conducted this research at the intersection of model building, diagrams, and generative design, applying this knowledge to the design of multifaceted digital environments, from micro to macro scale, in two- and three- dimensional worlds. To develop this research, we first observe and characterize the architectural methods of model making and their potential to facilitate the design process of interactive systems. Next, we describe how physical hand-crafted and digitally fabricated models of different types assist in various stages of the design process. To illustrate how model building could support fluid exploration of multiple data sets, we built a 3D interactive campus model visualizing multiple layers of building-specific data. The system uses physical models as tangible tokens on an interactive touch surface, visualizing energy use and weather data daily over a two-year period. As an extension of our design, we developed a conceptual framework from this project to highlight the potential of physical models for supporting embodied exploration of spatial and non-spatial visualizations through fluid interaction. We then examine the use of diagrams in architecture and develop a conceptual framework based on the concept of data tectonics to organize and structure the design process of physical and immersive data systems. To further study the use of diagrams and generative design for data visualization, I collaborated with researchers at Tableau Software to develop a patented Tableau extension that self-generates and evolves up to thirty different design permutations at a time. The system randomly assigns a pre-specified palette of mark types to a chosen dataset giving designers the option of adding or deleting options that they deem promising. As a final project for this research, we brought the three principles of model making, diagramming, and generative design together to create a large-scale physical and immersive data visualization. In collaboration with the Department of Social Work at the University of Calgary, the project uses diagrams and generative design to prototype a series of three-dimensional encodings visualizing Global Gender Gap statistics from the World Economic Forum. The tent-like forms evoke sheltering structures that can be registered, experienced, and measured with the whole body. For this project, we applied the diagrammatic approach used in parametric design to traditional information visualization design principles and identified workflows that support rapid exploration and fabrication of multiple data design alternatives. There is no doubt that data and digital technologies, including machine learning and AI, will be part of our human fabric in the future, but what that looks like and how it is structured is still up to us. We need artists, and more diversity in general, in order to do this to the best of our potential as humans. In determining which practices encourage the creation of rich data-driven environments, this research underscores the fundamental need of humans to make sense of the world, inspiring designers to develop new spatial constructs that integrate both the art and science of the built environment.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.