Research & Writing
Weekly critical responses from Planetary Machines, a theory seminar on technocriticality in architecture and urban design. Recast here in problem and solution form: the question each reading raised, and the position taken in response.
| Format | Critical writing, weekly response |
|---|---|
| Academic Work | Planetary Machines: A Theory Primer for Technocriticality in Architecture and Urban Design |
| Institution | UCLA, M.S. Architecture and Urban Design |
| Instructor | Guvenc Ozel |
Architecture Theory
Towards Post-Architecture
Response to readings on transarchitecture and cognitive space
Static architecture holds one form regardless of use. A responsive architecture reconfigures through a continuous feedback loop with the people inside it.
Problem
Architecture has always assumed a static container: fixed materials, fixed form, a space people move through but never change. Every adjacent field, robotics, AI, extended reality, has already dissolved the line between digital, physical, and biological systems. Architecture is the discipline still acting like that line holds.
Solution
Reframe architecture as responsive, not static: space that shifts configuration in response to use, not despite it. This isn't speculative. The internet is already proof: an environment that reshapes itself around interaction. Extend that logic into physical space through AR overlays and cognition-linked environments, and the result is transarchitecture: buildings behaving as systems, not fixed objects.
Original response, week 9
Standing at the cusp of the Fourth Industrial Revolution, wherein technology enables humans to inter-permeate into different domains. Boundaries of digital, physical, and biological are no longer retained. The advent of robotics, artificial intelligence, extended reality, big data, the Internet of Things (IoT), and the list that would go on with the passing time has enabled us to achieve new forms and meaning in different fields. Technology has made us go beyond the boundaries of our cognition and understanding and look at different domains from a new perspective. So, why not this new perspective brings about changes in the norms of architecture?
Architecture has traditionally been a static space adorned with different materials to let humans inhabit it. Humans would interact with this space. But what if Architecture goes beyond this physical space, materiality, and static forms? If we question ourselves in this perspective, we can derive a new language and style of architecture transformed by technology like robotics, AI, machine vision, extended reality. The idea here is to create such spaces which could change their morphologies, configurations, or physical qualities when it interacts with humans. This is called transarchitecture.
The concept of transarchitecture may seem to be almost an unreal speculation or a far-fetched scenario, but we are already doing it. The internet a micro-world already existing in this present reality that morphs itself with human interaction seems to be analogous to the concept of transarchitecture. The internet has its ecology just like the architecture, just that it molds itself as per the human interactions. The internet today is confined to a two-dimensional screen but what if it goes beyond those confines. At that stage, there would be the synergy of the physical and virtual world through the means of new media (like augmented and virtual reality). A new architecture would be born. This would be the metaverse, a version of the internet beyond the screens.
Coming back to transforming the physical spaces we are already living in, how this could be transformed beyond the materiality and physical specifications of the building. An interesting experiment illustrated in the reading "Cerebral hut" showcases how cognition can transform spaces points towards the possibility of the architecture which could change with our interactions with it. There are other mediums of transforming our already existing spaces through virtual and augmented reality wherein users could interact with the elements overlaying the real spaces. This could be another example of transarchitecture.
To conclude, as we are moving into a time where there is a synergy of different domains with the help of technology. Architecture is bound to transform and have a new style altogether that could enable us to explore new styles, aesthetics, space, forms, and language.
Design Technology
AI-Images
Response to readings on generative adversarial networks in visual and spatial production
Generator and discriminator refine each other on existing precedent data, producing new spatial or facade options rather than a single deterministic output.
Problem
Spatial layout and facade design are among architecture's most repetitive processes, dependent on a designer's accumulated intuition and precedent knowledge, and slow because of it. Every new option is manually synthesized from what's been seen before.
Solution
Generative adversarial networks learn spatial and facade patterns directly from existing building data, then recombine them into new, plausible configurations, compressing the most iteration-heavy stage of early design. The tradeoff is real: output quality depends on the size and range of training data, and a model trained to perfection can lose the productive imperfection that makes a generated option worth designing with, rather than just realistic to look at.
Original response, week 3
Images have been means of expression and a way of portraying thoughts and concepts in arts and architecture. The medium of making these visuals started from hand drawings to digital tools and now using artificial intelligence. GAN (Generative Adversarial Networks) a class of machine learning is used as a medium to create new images from already existing images.
GAN is a versatile tool that could take input of existing images and identify patterns and give a plausible outcome developed from the already existing images. GAN's have a wide range of applications which start from generating realistic images, translating image to image, translating text to image, increasing the resolution of images, 3D objects generations, and various such image-related manipulations.
This versatility of GAN's makes us think about their possible implementations in architecture. In architecture, developing spatial layout is the most repetitive process which requires intuition previous knowledge, and maybe a lot of time. Using the ability of GAN to learn images, it can be used to identify spatial configurations of already existing layouts of similar buildings and later build new plans using its intuition. Another example of GAN could be developing facades where data of already existing façade can be fed into the algorithms to create one.
GAN when seen as a medium to create art yields interesting outcomes. The visuals generated by GAN algorithms can be surreal and have some closeness reality yet indeterminacy. "Machine Hallucinations" artwork by Refik Anadol is an example of the application of GAN, DCGAN (Deep Convolutional Generative Adversarial Network) where he uses data from digital archives and publicly available resources and fed into machine learning classification models and clustered into thematic categories which are brought together using StyleGAN. The beauty of GAN methods of creating artworks is the ability to create such compositions which may seem to be realistic due to the algorithm's abilities to combine textures and objects which seem to be relatable. It is intriguing to know if the model gets trained to perfection, then these surreal images won't be achieved and the artist would have to tweak the algorithms again to achieve fantastical imageries.
After seeing the possible applications of GAN in art and architecture, there are several considerations that one must investigate. Firstly, the availability of huge data. Since GAN works on a database of images, it will be important to get an adequate and large set of images to train the model to get the required outcomes. Sometimes, as needs generate it may be required to train its models which can be a tedious and time-consuming process. While on the other hand Artificial intelligence, machine learning, and GAN can have versatile implementations as and when the technologies advances in the field of arts and architecture.
Planetary Systems
Backcasting Kardashev One
Response to readings on planetary energy scale and terraforming
Kardashev's scale, built to measure civilizations by energy capture, reframed as a backcasting tool: define the livable endpoint, then work backward to what must change now.
Problem
Human energy consumption sits on an unsustainable trajectory, and the Kardashev scale, a framework built to measure civilizations by their energy capture, treats reaching Earth's ceiling as inevitable without specifying what happens after. Meanwhile, terraforming is already underway through cities, altered terrain, and artificial islands, without any coordinated direction. The ceiling is coming either way. The open question is what shape the aftermath takes, and who has any say in it.
Solution
Treat terraforming as a design practice already in motion, not a future hypothetical. The real design question isn't whether to terraform, it's how to do it toward inhabitable, biodiversity-positive outcomes instead of drifting there as an accumulated side effect. Backcasting, defining the endpoint first and working backward, replaces forecasting from current trends as the operative tool.
Original response, week 1
Energy consumption has been a matter of concern for quite a lot of time. The environmental crisis, climate change, CO2 emissions are the consequences of this incessant consumption of energy. After all these phenomena happening, where our Earth is heading towards? To gauge this, Kardashev's scale has been one of the means to speculate and research.
Kardashev's scale was originally for the study of extraterrestrial civilization and its ability to harvest energy. To determine the types of these civilizations it is necessary to establish some kind of communication with the extra-terrestrial civilizations which is still a work in progress and is far-fetched as it would require highly advanced technologies. Instead, Kardashev's scale leads us to think about what are the consequences of our planet and how we enact upon it. According to the scale, it is inevitable for Earth to reach its maximum energy, at this maximum energy level it is projected to have an apocalyptic ramification. So, the question arises how this would be tackled?
This further leads us to think about the possibilities of terraforming of the Earth, the process of modifying the atmosphere, ecology, topography of the planet. If we understand closely, humans have been terraforming Earth already. Building cities, modifying terrains, making islands, tunnels bridges, etc. all these are processes of terraforming. These activities have altered Earth's natural systems. This in turn has resulted in more damaging effects, a consequence of the capitalist and consumerist approach. Though some of the intergovernmental initiatives for climate change consider techniques like geoengineering, managing biodiversity ecologies, and carbon reservoirs. But according to Kardashev's scale, if the inevitable is inescapable and can only be delayed, it makes us think what if there are forces which humans won't be able to control? Humans could be only a part of this journey or one of the many responsible causes of the inevitable.
What could the future be after the inevitable? It is a question which time would be able to answer but what is possible to do right now is to find those permutations of the future by projecting our current growth. In my opinion, the major factor of this growth is the advancements in technologies. Technologies ranging from our mobile phones to high-tech rockets meant for space exploration. What they can tell us is the possibilities of the future. This could range from harvesting full potential of energy from the sun or the other natural resources, artificially containing or maintain biodiversity, computer machines or artificial intelligence taking up most of the human tasks, space travel and exploration, artificial atmosphere and climate regulation and the list can go. The Earth could be a combination of so many possibilities after the inevitable.
Right now, as far as humans are concerned as a part of this journey, it is our hands greatly where we are taking the planet Earth towards, repairing the damages we caused and seek for better possibilities of the future.