Florian Zeif

Digital Fragments:

Mapping the invisible city

Digital Fragments examines the evolving relationship between data science and architectural knowledge-making. In contemporary architectural practice, many projects are developed for sites that are never physically experienced. Instead, architects increasingly rely on mediated forms of site representation, such as satellite imagery and statistical abstractions. While these representations enable new forms of analysis, they often precede and sometimes replace direct spatial engagement, narrowing the forms of knowledge that inform design decisions.

Rather than rejecting this condition, this thesis takes it as a starting point. If architectural knowledge is increasingly produced through computational means, the central question is no longer whether digital methods should be used, but how they are assembled, navigated, and curated in more meaningful ways. In particular, how these might be reoriented to account for lived, situated, and socially embedded urban conditions, rather than abstracting them.

Therefore this thesis asks:

Can urban data be produced in ways that are more visceral, and socially aware?

Situated within this condition, I position my practice as a hybrid architectural and computational practice, using programming as a means of spatial inquiry. In response to the research question, this project is developed through my own design and coding, aiming to create accessible methods for collecting on-site knowledge through a participatory mapping and analysis tool.

In this process, I developed a series of digital mapping modes within a self-developed web application. By incorporating multiple media formats as imagery, sound, and user-generated labels, participants can conduct site-based research and directly integrate their findings into their own GIS systems.

Through multiple case studies in urban contexts around Vienna, the tool is iteratively developed and tested, allowing its methods, interfaces, and modes of data collection to evolve in response to situated use. Iteration is treated not only as technical refinement, but as a process of architectural learning embedded in practice.

As an integral part of this work, I develop an approach to ethical programming by critically examining the analysis of imagery and situated recordings using contemporary open-source algorithms and datasets. This includes reflecting on how data practices can be de-weaponized, emphasizing transparency and contextual integrity. In doing so, “Digital Fragments: Mapping the invisible city” argues for an architectural practice in which computational tools do not replace spatial knowledge, but are designed to actively mediate, extend and situate the experience.

Universität für angewandte Kunst Wien