Matiss - Architecture’s Deep Pictures
An image is not a building, and a building is not an image. The shaping of the built environment through planning and construction rests on the separation between representation and operation. A building is a spatial outcome of a mobilization of material resources, labor, and time. It is also a product of both visual and alphanumeric assembly instructions that suggest its appearance and properties. Yet, the ongoing enmeshment of computation and material worlds, or as Wigley (2017) put it—the shift from paper to screen space—suggests that buildings and images do not exist too far apart either.[1] In today’s information-rich environments, not only proteins, microchips, wind turbine blades, cars, but also architecture is increasingly often “twinned” with highly detailed digital replicas that are used in simulations, stress tests, maintenance planning, life cycling analysis and other forms of predictive planning.
Since the advent of networked computing and graphics processing in the 1990s, much of architecture’s imagination was captured by the new formal possibilities (e.g. the practices of Zaha Hadid, Greg Lynn and their contemporaries), but also the intersection between digital and physical environments. One example is the “phygital” version of NYSE trading floor designed by Asympotte (2000), another would be MVRDV’s Metacity/Datatown project (1999) where data is a substance from which cities are made.[2][3] Amidst these explorations, arguably a more consequential development has been the widespread adoption of building information modeling (BIM) protocol, which affords the possibility of collapsing all possible information about a building into a single digital entity. Cardoso-Llach (2017) describes such models as “structured images”, which unlike visual images, function as mechanism where every building part is held together by parametrically defined interdependencies.[4]
As a result, most buildings nowadays are built at least twice—first as a structured image, then as an object. Often, they are built many times more, as they co-evolve over time through feedback loops of information exchange between digital and physical environments, surveys, projections, and simulations. All such forms of the building contain information. While the image version of building information is computed via electronic signals, stored in data structures, operated via software platforms, rendered in polygons, and displayed on grids of illuminated squares, its physical version contains information encoded not as data but as the particular arrangement and properties of physical matter. [5]
A building can be an image, and an image can be a building. While BIM represents the most common form of real-time imaging of the built environment, this logic applies to other domains too, such as landscape and city information modeling, as well as the broader forms digital twinning at scales including the planetary, as exemplified by EU’s initiative to build an operational digital replica of the Earth, called Destination Earth.[6] Yet, tethering more material and finer scales to real-time electronic images remains a challenge, if not an impossibility. Both atomic and subatomic structures of matter function differently from computational concepts like pixels, vectors, meshes, textures, and simulation engines, and interpreting data between them involves approximation, noise and other forms of information loss. This distance between material realities and their twinned replicas remain more than a technical obstacle, but perhaps also suggests possibility of other building cultures to emerge.
✺✺✺
- ↑ Wigley, Mark. “Black Screens: The Architect’s Vision in a Digital Age.” When Is the Digital in Architecture?, edited by Andrew Goodhouse and Canadian Centre for Architecture, Sternberg Press, 2017.
- ↑ Asymptote Architecture. NYSE Virtual Trading Floor. Canadian Centre for Architecture, 2015.
- ↑ Maas, Winy and MVRDV (Firm), editors. Metacity Datatown. MVRDV/010 Publishers, 1999.
- ↑ Cardoso Llach, Daniel. “Architecture and the Structured Image: Software Simulations as Infrastructures for Building Production.” The Active Image, edited by Sabine Ammon and Remei Capdevila-Werning, vol. 28, Springer International Publishing, 2017, pp. 23–52.
- ↑ May, John. “Everything Is Already an Image.” Log, no. 40, 2017, pp. 9–26.
- ↑ European Commission. “Destination Earth.” Destination Earth, https://destination-earth.eu/. Accessed 30 Jan. 2025.
Responses:
"You're definitely onto something with linkage between virtual abstractions and physical constructions, though I'd say there's a lot of cases where digital and physical spaces are coevolving. In some of the cases I'm thinking about, the construction site, its physicality, functions as generative material for the development of digital mirror worlds, which then go on to structure the buildings-to-come at the site. It's a push-and-pull dynamic that makes it hard to put one in front of the other." —Kola Heyward-Rotimi
"I can't help but read the text in a way that is not just material-centred [...] does „all possible information about a building“ already contains the ones about the living entities inside? Or is the human component the agents role that would be elaborated in a further step?" —Maja Funke
"I was wondering how this relates to the semiotics of architecture and cities - including the interpretation and construction of architectural discourse + the (un-)readability of architecture and the urban. To what degree, and how, is this different than the idea of the city as semiotic space, which we know from modern cities?" –Søren Pold
