Digital Crafting Feb 8-10 2010
The first digital crafting workshop event “Parametric Design – Encoded Behaviour” was held at the Royal Academy of Fine Arts, School of Architecture in Copenhagen February 8-10 2010. Organized by Martin Tamke of CITA and London-based Paul Nicholas of Mesne, the three-day workshop brought together seventeen designers from the fields of textile design, engineering, architecture and computation to explore ways of understanding material performance and behaviour.
The aim of the workshop was to create a research environment to observe, measure and then input and relate material behaviour to a digital parametric model. This was seen as a way to begin to understand and control material performance in a component system. The workshop included short group design exercises, material experiments using timber and textiles, grasshopper tutorials, guest lectures by Paul Nicholas, Mette Ramsgard Thomsen and Hauke Jungjohann, group discussions and reflections on the design and production process, and group component assembly of a working 1:1 prototype or “demonstrator”.
How can we understand the relationship between geometry and material property?
In the workshop we focused on a simple parameter, the bending of a linear material, because this allowed us to simulate the behaviour of the wood and encode this information into the digital model. We used Grasshopper as a digital tool, as a visual form of coding, as a platform to begin to understand coding. The relationship between the material and the digital was approached at full scale, a 1:1 material study, through encoding the information gathered through experimentation in the digital model. During the process, we discussed issues of accuracy, because by measuring by hand there is a degree of abstraction and error, and together we questioned if digital design should be focused on accuracy or relationships between elements. In our group discussion we questioned “is there something that we learned from actually observing and bending the material that we could not have learned otherwise?”
Are we bending sticks or are we discussing parametrics?
Using this technique of directly, empirically testing materials, we developed new ways of thinking about something complex (the bending of the sticks is a sine curve, and there are mathematical equations relating to the constant length of the material and how it compresses) without needing to understand the mathematical formula behind it. In this way for designers, material behaviour can become a design parameter. Parametric thinking offers the opportunity to link between material behaviour and computation – we began to ask ourselves, “how can we design the relationship between physical properties not just information in parametric models?” How could this new parametric way of thinking change the way architecture is designed and made?
How can we consider assembly and issues of fabrication?
In conventional design practice, often due to cost and time constraints, thinking about materiality usually comes at the middle to end of a design process. In contrast, during the workshop, material quality and performance were seen as design driver, at the initial stages of the design. We discussed what happens after design: fabrication and assembly. Even with digital design and mass customization, the assembly of components can be time consuming and a design task in its own right. We discussed architectural examples of Sydney Opera House, and the Seattle Library where the contractor was tasked with treating the assembly of the building components as a design exercise. In our group discussions, we talked about how designers can reduce and simplify a design problem to the minimum number of parameters rather than slavishly trying to describe reality. Too many parameters can lead to unpredictable complexity, potentially making the digital model unusable.
Lectures framing further discussions
Three lectures on the final afternoon illustrated design and textile works using a digital approach with material sensibilities but at vastly different scales. These lectures were used to frame discussions about the implications and applications of “encoded behaviour”.
Hauke Jungjohann from Knippers Helbig Engineers showed the Shenzhen Airport project that he worked on at Massimiliano Fuksas´s office, which has a 1.3km long concourse and a double skin façade that used parametric panelization tools to control light, drainage and appearance. Jungjohann illustrated how in this project it was key to develop tools to design basic element variation to create something that can adapt to double curved surfaces. Drawing on ideas from the lecture, we discussed ideas about how a parametric model can simulate material behaviour during the design process and how computational tools could allow more customized and useful designs.
Mette Ramsgard Thomsen from the Centre for IT and Architecture (CITA) showed a new project called “Listener” which showed how these ideas relate at a personal, human scale and can incorporate ideas of responsiveness and adaptation. The project illustrates ideas about computational tools, and what happens when designers are able to design materials. We discussed the way that these ideas could lead to lighter and more intelligent building practices. We also discussed the limits of digital tools, and what happens when the hardware or software being used becomes outdated? How robust and secure are digital models or information?
Paul Nicholas´s lecture illustrated the importance of communication in design projects using several examples of geometric layering including a project which was a series of 10m tall steel sculptures in Melbourne. The design was described and communicated to all collaborators using a single digital model. Building on these ideas and projects, we discussed how moving information from upstream in the design process down to the beginning stages. How can design information be communicated during the design process? When should a digital model be introduced in the design process?