Programming Generative And Branding
- Programming Generative And Branding Research
- Programming Generative And Branding Tools
- Programming Generative And Branding Management
- Generative design is the process of defining high-level goals and constraints and using the power of computation to automatically explore a wide design space and identify the best design options. While generative design applied for manufacturing is gaining more and more attention, its use for architectural space planning has been fairly explored.
- Interaction desginer and software consultant who uses programming for generative design of audiovisual and interactive projects. A graduated designer and self taught code specialist whose spatial works include media facades, television studio designs and semi-permanent light installations. Daniel lives and works in Cologne.
In my research I’ve studied the current state and some of the intellectual history of what we’ve come to call generative design. Depending on the definition of the practice, its origin can be arguably pinpointed either in Gerald L. Delon’s 1970 paper A Methodology for Total Hospital Design, or sometime in the 1st Century BC with Vitruvius’s Ten Books on Architecture. Through admittedly less than exhaustive studies, a succinct definition of “generative design” has remained elusive.The first two sentences of the definition on Wikipedia, evidently derived from informal comments by an Autodesk representative, are a serviceable beginning:
Generative design is a form finding process that can mimic nature’s evolutionary approach to design. It can start with design goals and then explore innumerable possible permutations of a solution to find the best option.
This is Hadeel, a multidisciplinary designer & artist focusing on information & generative design. H by Hadeel - Personal Branding. Graphic Design, Branding, Programming. Popsicles - zine. Graphic Design, Art Direction, Editorial Design. Child Mortality Rate - Infographic. Graphic Design, Programming, Art Direction.
However, this definition is diluted by the qualifier “can”, admitting doubt as to whether the definition is comprehensive, or whether it merely describes one approach to a variety of problems that may yield to a broader choice of methodologies. Subsequent sentences of the proposed definition cast further uncertainty on its universal applicability, as a specific implementation is summarized:
By using cloud computing, generative design can cycle through thousands—or even millions—of design choices, test configurations and learn from each iteration what works and what doesn’t. The process can enable designers to generate brand new options, beyond what a human alone could create, to arrive at a most effective design.
In my judgment, this extended definition is too narrow, focusing on certain techniques and technologies to the exclusion of a broader definition applicable in multiple contexts. There is nothing in these ideas that is false or irrelevant to the field, but the blend of concept and implementation leaves unsatisfied the need for a practice definition independent of supporting technologies. Absent a shared definition, we literally won’t know what we’re talking about when we discuss how to employ computation to make better decisions about buildings.
To help move forward related discussions, I propose this definition for comment and possible adoption by the AEC industry, drawn from my research and past work developing prototypes for related commercial software:
Generative design is the automated algorithmic combination of goals and constraints to reveal solutions.
Why “generative”?
Seemingly applied to design after its use in classifying artwork created in conjunction with autonomous information systems, the term retains useful and provocative intellectual connections to studies in emergent behavior and automata dating to the mid-twentieth century. The term is appropriate to distinguish the practice from the more traditional design process, whose every artifact is born in a sequence of human gestures. “Generative” implies at least a measure of autonomous indirection with an attendant possibility of favorable discovery.
Why “design”?
There could be an argument for the addition of “generative construction” to distinguish the sector supported by the approach, but I believe such a distinction beggars the concept of “design”, which should attach to any decisions made prior to attendant actions. Construction schedules are designed; material delivery methodologies are designed; the choice of tooling and its application to materials are design choices. Ultimately, although these are all examples of decisions outside the traditionally understood building “design” process, they remain firmly in the realm of predictive choice. The term “design” is sufficient to summarize decisions that will affect future actions and experiences.
Why “automated algorithmic”?
I questioned whether “automated” required inclusion, but ultimately decided that it is a necessary addition to distinguish the modern endeavor from a mere manual application of building rules as sourced, for example, from architectural directives and guidance authored by Vitruvius, Palladio, and Alexander, three prominent exponents of the approach. Including “automated” fixes the origin of the practice firmly in the post-industrial era, and in combination with “algorithmic” suggests implementation through computational technologies.
Fundamentally, an algorithm is a recorded and repeatable procedure, sometimes cast as one or more mathematical equations, but as easily characterized as a recipe. Following a sufficiently detailed procedure with expected outcomes, one should be able to produce a satisfactory meal. Greek, Roman, and Renaissance architecture is adorned with formal recipes whose purported outcome is firmness, commodity, and delight. Many such “rules” of building are still applied, whether drawn from the behavior of materials or the emotional effect of spatial vocabularies and ornament. The Gothic cathedral is designed to engender a specific response from the inhabitants, while the narrow confines of the prison cell are designed to minimally accommodate biological needs with lesser regard for emotional consequence. Building types and functional programs are algorithms. Whether these algorithms are expressly recorded or implicit in human practice, they remain remarkably consistent and have a profound effect on the experience of the built environment.
Why “combination of goals and constraints”?
Algorithms exist to combine one or more selected goals with unavoidable constraints to arrive at solutions. In the case of a cathedral, any number of goals from encouraging reverence in worshipers to the expression of a community’s capabilities and aspirations may be satisfied; in the case of an effective prison, at minimum the sequestration of a transgressing populace is satisfied. In both examples, budget is an obvious constraint to cite, profoundly influencing the outcomes, whereas schedule may be sufficiently relaxed in the former case to result in literal generations of labor before completion. Siting, environment, available materials and methods, the skills of the labor force, and any number of other limitations and circumstances may constitute constraints on what can be accomplished. All aggregated decisions leading to the singular outcome of the completed building amount retrospectively to an algorithm for creating that building, a procedure more or less successfully satisfying certain goals while more or less successfully respecting relevant constraints.
Why “reveal”?
Several other concepts might have served well here. “Discover”, “propose”, and even “create” were considered, but to capture the metaphor of design exploration while retaining a clear position on which side of the screen creativity continues to reside, it seemed best to select a concept as easily applicable to the microscope and telescope as similar examples of instruments that extend human capabilities.
Why “solutions”?
The plural alludes to the promise of abundance inherent in automation. Whether automation affords an abundance of conserved time, of multiplied product, or of design possibility, the effective application of machinery should lead to increased resources, as evidenced in the explosive growth of food production worldwide after the green revolution, or in the near infinity of manufactured goods on offer in technologically developed regions of the world. Procedures are fundamentally worth automating when automation leads to some form of abundance: in the realm of design, to an abundance of choice, an abundance of understanding through analysis and comparison, and an abundance of recovered time.
Increasingly compressed design and construction schedules have substantially reduced available time for the option exploration and analysis that can result in improved building solutions. Market incentives drive toward arriving at consequential decisions quickly and mitigating undesired consequences as cheaply as possible with the least schedule and budgetary disruption, often leading to unsatisfactory results that may be further imperfectly mitigated. The only method to recover time within other immovable constraints is greater efficiency, often through automation of time-consumptive tasks. In the context of design decisions, the next application of automation should be in augmenting the design team’s knowledge and judgment with systems that produce a multiplicity of options for exploration, analysis, and informed choice. Design choices are, arguably, infinite, with building design choices ultimately constrained by the needs of human biology. The promise of automation is that a substantially larger region of the possible design space can be effectively explored by managed presentation of plausibly acceptable options.
Programming Generative And Branding Research
“Solutions” also implies the existence of one or more problems to solve, for which “design” is employed as an answering methodology. By contrast, the problem of an untied shoelace, absent a novel knotting scheme, need not employ “design” as a solution methodology, but merely the algorithm for constructing a knot whose application is nearly a reflex in the adults of some cultures.
Programming Generative And Branding Tools
Programming Generative And Branding Management
An earlier draft of this definition included the necessity of a “defined problem”, but finally in service to brevity and clarity, “design” and “solutions” were deemed sufficient to absorb the additional burden of implying the existence of problem to be solved. Design decisions are largely dependent on defined problems, and problem definition is a critical, difficult, and inescapably human endeavor inseparable from design. Algorithms will not substitute for the human values and priorities from which definitions of problems are derived. The intertwined activities of problem definition and solution selection rely on human judgment rooted in human value systems. Whether those values are as easily measurable as a targeted return on investment, or as qualitative as an uplifting experience for building inhabitants, designers and their clients will always define problems together and apply their knowledge and judgment to solution selection, in whatever manner solutions are produced. Generative design offers the promise of wider exploration, deeper understanding, and increasingly confident decisions of high quality, as the design space previously unexplored due to restrictions of schedule and budget opens through the application of automation to what remains today a largely manual and laborious process replete with trial and error.
Properly implemented, generative design affords relevant expertise contextually and consistently delivered to every building project, expanding the possibilities to improve the built environment.