Rethinking the line: Computational and critical approaches to motion-based drawing

Abstract

This thesis aims to highlight the role of architectural drawing both as a theoretical issue and an experimental form of production, suggesting that drawing cannot be thought in terms of the production of a representative object but rather in terms of a practice occurring in time, a practice which is temporal, situated, and performative. While the representative approach to drawing tends to think the line in terms of a representative object which facilitates the rendering of space legible and the transfer from the drawing to the object, the line becomes an operative object which relates to the conditions in which the drawing is produced, to the behaviors being registered during the drawing, and to the processing which the drawing undergoes to re-produce the object. This means the emphasis is shifted from the object to the drawing to the production of the object, to the production of the trace, and to the reproduction of the object from the trace. In this way, drawing is not thought in terms of a representative object but rather in terms of a practice which is a site for the production of architectural thought at the crossroads of the visible and the conceptual. The historical approach of this thesis's interpretative framework recognizes an evolution of the role of the line in architectural drawing. The line begins as an instrument of representation, then an instrument of meaning, and eventually an instrument of action capable of redefining spatial thinking. The significance of this evolution cannot be entirely explained in terms of progress in drawing media. It is also linked with the ability of drawing to produce thinking beyond its own representational limits and beyond its own context. In this context, drawing is seen as an action situated in an operational field, in which meanings emerge from an articulation of perception, reality, and interpretation, often without being explicitly articulated. The line is therefore seen as an instrument of action, capable of engaging its environment and redefining tensions, boundaries, concentrations, and orientations. In this context, drawing is seen as an operational field in which architectural thinking is produced, and not as a medium for the representation of architecture. This trajectory of theory is expanded to include the paradigm shift of modernism and its implications for drawing. In this context, drawing is seen to have roles beyond those of a purely technical document within a redefined design paradigm and its relation to historical continuity. Instead, drawing becomes a vehicle for a conceptual language. The line no longer merely carries values such as form, function, and clarity. Instead, it participates in their construction. In this context, drawing changes its focus from answering the practical concerns of "How to make" to supporting inquiries such as "What can be thought" and "How can one think". The thesis suggests that this trajectory takes on added force within computer-based drawing environments. This is due to their fundamental redefinition of the status of the line. The use of CAD/BIM and parametric design redefines the line as a construct capable of regeneration, reversal, updating, and derivation from variables. This redefinition of the line does not merely speed up drawing. Instead, it re-constitutes its ontology. The line is no longer merely drawn. Instead, it becomes both product and stimulus for further production. In this context, drawing moves beyond mere hand motion and takes on a form ever-more clearly delineated by algorithms and data flows. In this context, drawing takes on a re-defined and heightened level of processual character. However, it is at this particular point that the present thesis assumes that the conventional/digital drawing dichotomy is limited. Indeed, it is assumed that the crucial role that drawing plays in design thinking is not so much determined by the specific means that are being used, but rather the manner in which drawing is constructed as a cognitive and operative process. In other words, it is not so much that the present research is focused on answering the question of what is being drawn, but rather it is focused on the manner in which action and movement are being explored, encoded, and duplicated. In addition, it is focused on the manner in which traces are being converted into environments that are beneficial for thought. In such a context, it is assumed that the algorithm functions as a new alphabet for the digital era, which directs drawing away from conventional visualization and toward computerization. Indeed, it is assumed that the line is abandoning its specific meaning, which was traditionally assigned to paper, in favor of a new, data-driven, and relational construction, with drawing being constructed as a process that is constantly being expanded via algorithms and computers, rather than being a solitary operation. The original contribution that this thesis attempts to make goes beyond the level of abstract discourses within the theoretical framework. It attempts to empirically materialize the notion of line production through a hybrid approach that incorporates a drawing device, also called the line-maker, and a Grasshopper environment. The Arduino-based drawing machine, while facilitating the drawing output, also becomes a research tool that enables the drawing to be measured, replicated, and varied. Drawing, as a notion, becomes a set of activities that follow a sequence and depend on parameters, motion, and transformations. The physical drawing output that the drawing device produces takes on the characteristics of deviation, dense intersection, change in direction, and rhythmic change that might be seen as error in the context of a traditional approach to technical production. However, this thesis attempts to see these drawing characteristics not as something to be corrected but as a key aspect that enables the drawing to illuminate the performative aspect of drawing. The drawing output becomes an event-based imprint that not only shows what has occurred but also how it has occurred. This move to the digital environment has been conceptualized not only in terms of copying but also in terms of the intentional act of translation. The relationship between the physical and the digital traces does not simply operate in one direction. Rather, the thesis argues that the relationship between the physical lines and the digital lines that represent it is one that functions in both directions. While, on the one hand, the physical lines are sampled and recreated in the digital environment, the digital environment, too, creates the space for reflection and control that can, in turn, impact the subsequent stages of the physical process. Drawing, therefore, ceases to be the product and becomes akin to the function, moving between the physical and the digital environment, and the trace becomes not simply the product that has been created on one surface but the research that takes different forms and is repeatedly recreated and reinterpreted. The production and reproduction of lines in the thesis are represented and organized through two representational levels: 2D and 3D representations. In this context, within the 2D representation, the line is represented as a trajectory in a 2D space. This representation allows for the parametric processing of line pattern and for comparison across disparate production conditions in terms of features such as orientation, density, and heterogeneity. Though this representation of line in a 2D space makes it visually intelligible and measurable, motion drawing is not simply a geometric trace. Therefore, a representation form is required to render this temporality perceivable. This is accomplished through the space-time volume approach. In this context, within the 3D representation, the line is extended beyond its trajectory in the XY space and becomes a spatialization of rhythm and characteristics of temporality. The implementation of cumulative and non-cumulative temporal logics allows for an investigation of temporality within drawing. This investigation occurs within two different and specific reading forms. One form involves an investigation of temporality as an accumulation within drawing, and the second form involves an investigation of temporality as a difference within the line pattern. In this way, the line becomes imbued with a form or trace of form and an investigation of the way in which this form was generated. The drawing becomes a tool for investigation and understanding of space-time. This multi-representative concept will allow for the systematic definition of the relationships between lines and their environment. The thesis will examine the relationships in two different ways: through the agent-space and agent-agent modes. For the agent-space mode, the line will cease to be considered as merely an active representation of the environment. It will, on the other hand, respond to artificial conditions in the environment by interacting through boundary changes, attracting points, and patterns. For the agent-agent modality, the line will be considered as an active agent interacting with other agents. This will imply relationships between lines through control points, orientations, and deformations. This will ensure that changes to one line will influence other lines. For the agent-agent mode, the line will not only be considered as an agent interacting with the environment but will also be considered as an agent changing through relationships with other agents, thereby generating new spatial relationships. This thesis aims to discuss the redefinition of the line as an active agent through the presentation of experimental results, thereby enhancing the theoretical basis for the integration of physical production and digital simulation. The evaluation of the produced drawings is carried out in a way that is consistent with the process-oriented stance of the thesis, using both in-action readings and similarity-based methods. In-action evaluation considers behaviors observed during production, such as rhythm, changes in speed and orientation, the formation of intersections, and local concentrations, through a process-focused perspective and relates them back to the theoretical discussion of the performative character of drawing. Similarity-based evaluation treats drawings as sequential and path-dependent series. Metrics such as Dynamic Time Warping (DTW) and Discrete Frechet Distance (DFD) enable drawings produced under different conditions to be compared, while matrix representations make relations across drawing sets visible. These metrics are used to increase the discussability of similarities and differences and to position the experimental findings within a more systematic comparative framework, rather than to replace theoretical interpretation. This thesis links drawing theory and experiment, reframing the line as trace, data and behavior through motion-based, algorithmic production and evaluation.