 Welcome to our Monday night evening lecture. It's a great pleasure to have David Benjamin with us this evening to speak about his recent work. I'm particularly excited about this evening, not only because Open House is always a warm and energizing marker in the semester, where we have the chance to interact with so many architectural enthusiasts all at once, but also because for those of us who know David in the many capacities he holds at the school. As an incredible teacher who has led the Advanced Four Studios for a number of years now, and who will, starting this spring, expand his directorship to all of the Advanced Studios, bringing his environmental focus to the entire sequence, and as the director of the GSAP Incubator down at the New Museum, which has enabled every year for the past four years, a group of alumni to experiment with new forms of practice. Tonight, tonight we'll cast David Benjamin in a slightly different light, as seen through the work of the living, a practice, his practice, that I believe to be one of the most interesting and pioneering practices that is reshaping the field of architecture today. At a time of increased environmental concerns, the living's work has not only led the way in rendering those concerns central to its position and approach, but more importantly, opened up new possibilities to address those concerns by systematically recasting architectural tools, reframing the discipline's foundations, expanding the ingredients that goes into its practice, and by designing new disciplinary intersections to re-engineer architecture's potential and impact in the world. David's calm and steady yet relentless commitment to turn things on their heads, often moving against the accepted discourses and practices of a certain time, has been inspiringly consistent. At the height of parametricism's former agenda, the living was using computation to enable muscles in the East River to render visible the amount of pollution in the water with the Pier 35 Eco Park Project. At the peak of biomorphism, biomorphism, and the design of buildings that look like jellyfish, David was intersecting biology and building materials to produce a new kind of living brick made of mushrooms and bacteria to enable communities to grow their own building block, as was demonstrated by his winning competition entry Hi-Fi with a team of expert collaborators, students, and many fans that came together to build it in the courtyard of PS1 MoMA in 2014. And in this moment of obsessive compulsive data crunching, where we believe cities will be smarter as a result, David has explored the biases of data recognition and machine learning, using these tools to either reveal the exclusionary designs that these new technologies necessarily produce as with his Twin Mirror Project for the 2017 Seoul Biennale, or at the other end of the spectrum, designing inclusivity into his work as with the Princeton Architecture Lab building that he recently competed, where every detected knot in the wood chosen for the building's walls was identified not for the purpose of elimination, but on the contrary, recognized as an imperfection to be intensified towards character, uniqueness, material sensuality, and beauty. Beyond the radical approach to designing projects, it is the design of his own practice, which David has also innovated in. As the first architectural practice to be acquired by an industry giant such as Autodesk, while still maintaining its research autonomy, David has enabled new ways of thinking about what an architectural practice can be, designing his as a research model that can directly intervene in the building industry and collaborate with industry leaders towards more creative and sustainable designs for the built environment, whether at the scale of a 3D printed airplane part as with his bionic partition project in collaboration with Airbus, at the scale of an office space, or at the scale of a city. David's thinking and design across scales, an obsession we have here at the school, and which has shaped the pedagogies across programs, as you've heard for some of you earlier today, from Kate Orff's Scales of Environment to Anders Haake's Notion of Inter-Scalarity, has also pioneered new territories for the field. His work on embodied energy in particular, assembled through a conference he organized here at the school and the book that ensued, have shed the light on this crucial dimension of architecture, that a brick is not just what a brick likes or wants to be, but rather the infinite resources it has taken to extract its raw materials, the many hours of labor it has taken to assemble it, and the carbon energies it has taken to transport it amongst other things. To render visible, connect, and design all of those scales together, through critical engagement, visualization, making speculation, and the projection and building of alternate possibilities for the future is at the core of what we can do today as architects, faced with the challenges and opportunities of our contemporary condition on this planet together. And it is certainly at the core of what David Benjamin and his practice, the living have not only done, but pioneered for all of us to be inspired by and hopefully follow. Please join me in welcoming David Benjamin. Thanks a lot for that generous introduction. You know, I thought that tonight in the spirit of the school, in the spirit of its experimentation and continuous reinvention as we just discussed with some of our prospective students, always reinventing curriculum, always re-asking questions and asking new ones, I took it upon myself to kind of reinvent my own lecture for tonight with a few new hypotheses and a few new projects. And so I'd like to invite you guys as my kind of community here to tell me your thoughts and ask any clarifying questions. I would very much invite that. First, I'd like to start off by acknowledging the incredible team that I work with. A few of them are here tonight. All of our work at the living is collaborative and these people individually and collectively bring it to life. I believe buildings are living organisms. They breathe and pulse. They inhabit complex ecosystems of species, technologies and culture. And understanding buildings requires understanding these vital signs in these ecosystems. Similarly, buildings are not static solitary objects dug into a single site. They are dynamic systems connecting many different places. They talk with their natural and urban contexts. They join together and cooperate with each other. They involve a longer duration and a wider geography than we typically consider. They actually begin with matter extracted from a landfill, extracted from the earth and end with matter sitting in a landfill. They involve energy, labor and resources from around the globe. And just as buildings might be broader and more multidimensional than the traditional definition, I think that the design ecosystems that architects create for themselves might be broader and more multidimensional. Along these lines, I'd like to describe our design ecosystem at the living, which includes a variety of interconnected forces and involves a hybrid of the familiar and the new, the ancient and the cutting edge, the technical and the social, the tangible and the atmospheric, the practical and the critical. The output of our work involves research, books and buildings. The topic of our work often involves sustainability and climate change, either directly or indirectly. And our approach often involves new possibilities at the intersection of biology, computation and design. But before I describe some of our projects tonight, I would like to outline some of the context for our work with computation and biology. For me, a big part of the story of computation in architecture dates back to 1906 when an Italian economist named Vilfredo Pareto published a thick book called Manual of Political Economy. It was filled with small hand-drawn graphs and with hundreds of equations, the algorithms of this era. One of the concepts advanced in the book was Pareto efficiency. This referred to a society where nobody could be made better off without somebody else being made worse off. As an example, Pareto imagined a society with the fixed amount of resources for producing bread and wine. One option would be to make a large amount of bread and a small amount of wine. Another option would be the other way around. For each option, if production was efficient, then this version of society would become a point on the Pareto frontier, oops, shown there. And since all the points on the Pareto frontier were mathematically equivalent, this frontier could be used to study trade-offs between different options for distributing resources. While the principle of Pareto efficiency was developed to describe economics and the design of societies, in recent years, architects have used a similar framework with aspects of building design, such as the design of a structural frame for a tall building with weight of material on the x-axis instead of bread and horizontal displacement on the y-axis instead of wine. In this sense, whether or not architects realize it, Pareto's theory has provided a foundation for the current obsession with performance and optimization in architectural design. Many architects today, like many economists, are enchanted by efficient and optimal designs because they are so clear and unambiguous. But as Pareto noted a century ago, efficiency is very narrow. An efficient distribution of resources does not necessarily equal a socially desirable distribution of resources. An optimal design does not necessarily equal a good design. At about the same time Pareto was working on efficiency, biologist Darcy Wentworth Thompson was sketching the patterns and body structures of plants and animals for what would become his famous book on growth and form. Thompson set out to explain natural phenomenon in terms of physical laws. He developed an approach that was mathematical and algorithmic, which was unusual for biologists at the time. As you probably know, Thompson developed an ingenious method for graphing the geometric transformation from one organism to another, such as the way the scarce fish becomes the pomecanthus fish when its orthogonal grid becomes a grid of coaxial circles. Thompson's equations were elegant, original, and compelling. And his approach has influenced not only biologists but architects and even features of architectural software. Yet Thompson, like the early Pareto, was so committed to his own novel approach that he neglected important genetic and biochemical evidence. In other words, he had a huge blind spot. Pareto and Thompson are not heroes to me. In fact, both are problematic in their own ways. Pareto in particular advanced theories that may have increased inequality and oppression at the time. So I'm describing these characters not as inspirations, but in order to argue that all important theories and technologies have their blind spots. And my hypothesis is that one way to deal with these blind spots is to cultivate a diversity of approaches. Pareto chose the quantitative over the qualitative. Thompson chose the mathematical over the chemical, but in both cases, it may be more helpful to understand these forces together rather than to choose just one. And when we turn to architecture, when we in this room turn to architecture, we might start by acknowledging that our design approaches and design theories and design technologies all evolve, all involve assumptions and potential blind spots. As an antidote, we might work to cultivate multiple perspectives, multiple tools, and a design ecosystem with sufficient biodiversity. And we might replace the paradigm of the individual genius with a more collaborative, distributed, and open-ended paradigm. And finally, in understanding the past context of Pareto and Thompson and the future that extends well beyond them, I think it is important to note that both computation and biology have changed radically in the past 10 years. While architects and designers have been fascinated by nature and biology for hundreds of years, biology of today is different. It is now possible to grow a cell alone on a glass slide instead of inside an organism. It is now possible to visualize neurons firing inside of a live animal in real time. It is even possible to cut and paste DNA and bring to life creatures that never before existed, such as yeast that produces anti-malaria medicine. And as of just a few years ago with the demonstration of CRISPR and gene drives, it is now possible to redesign or even eliminate an entire species very quickly, essentially molding evolution itself. In addition, it is now possible to apply the latest techniques of computation, such as computer vision and machine learning, to processes such as biological growth. Biological functions involve hundreds of dimensions, but if they can be simulated in a computer model, then they become a more actionable part of the design process. Yet the more you learn about biology, the more you see how complex biology is and how much we still have to learn about it. And in this sense, design with biology may involve design with a black box, design with forces beyond our complete control and design with uncertainty. This might offer an alternative to the framework of efficiency that dominates computation. A biological outlook for design might aim for diversity and robustness of the population rather than perfection of the individual. This new version of biology calls for a new method of design. And in the past few years, we at the living have developed three specific approaches for bio design and three categories for projects. The first is biocomputing. And I define this as using actual living organisms, not the metaphor of biology, but actual living organisms as processors to compute some kind of problem for us, such as reducing carbon emissions. This is a project with the airplane manufacturer Airbus. Airbus is very interested in reducing its carbon emissions and its global carbon footprint. And they're trying to do this in part by reducing the weight of their airplanes. So we set out on this long-term project with Airbus to help design new airplanes. But of course, we started small and in a very practical way, in other words, with this single component of an airplane shown in red, which is called a partition. So it's one of the most boring and forgettable components of an airplane. You've all seen it many times. It holds this fold-down cabinet tendency and it's basically the dividing wall between the seating area and the galley of an airplane. But it's actually a very difficult component to design. And like all components in an airplane cabin, it needs to undergo intensive certification. Part of the certification is what is shown here, a physical test in four tenths of a second where the partition is accelerated to the force of 16 times gravity, the 16G test. And just by looking at this video, you get a sense of the amount of force that this component has to withstand. So while the shape of this component is pretty boring, while it's effective on an airplane is pretty forgettable, it's actually a very challenging component to design in a lightweight way with traditional engineering approaches. First, it can only attach to the fuselage at two points on the top and two points on the bottom. The fuselage is the main structure of an airplane. It can only be one inch thick. And it has to hold the fold down cabin attendance seat hanging off of the side of it, which is basically a very asymmetrical load. And just when you think you might know where you would intuitively sketch some cross bracing structure by making maybe a cross brace, you find out that this small inset shape here has to be removable and therefore it can't participate in the main structure of the component very much. And that's because new regulations in airplanes require that if someone gets injured in the seating area of an airplane, you have to be able to put them on a stretcher and carry them around the corner out to the door. And that requires that you remove a part of this exact component. So in other words, it's a challenging component to design. And it's at this point that we took a kind of outsider approach to the problem and looked at the living organism called phyzerum, otherwise known as slime mold. So slime mold grows in complex networks that are both efficient and redundant. They're efficient in that they use a small amount of material to connect the dots. In this case, the dots are sources of food. And they're redundant in that if you remove one of the lines, the organism as a whole or the system as a whole can typically route around the problem and keep all the dots connected. So what we did here was create a kind of custom algorithm that modeled the logic of slime mold, not the form of slime mold, in order to create different interconnected lattices as cross bracing for the structure of a partition. Then we used the power of computer automation to generate, evaluate, and evolve literally thousands of design options for the partition. And here you see a different view of the data. Each point on this graph is a different design. There are 10,000 different designs here. And we're basically using a process that allows us to explore a very wide design space. It produces some designs that are unfeasible, some designs that are very unusual, many designs that are difficult for a human to design or in some cases even understand. But the idea of this approach, sometimes called generative design, is that we can use the power of computer automation to explore designs, to partner with the computer and uncover new possibilities that wouldn't have been possible otherwise. So here you see the designs color coded according to the type of design that they are. It's kind of virtual DNA. You see them plotted on a Pareto frontier where you wanna have minimum weight and minimum structural displacement. You wanna be at the zero, zero point. And this allows us a kind of interactive interface to see the geometry of different designs and hone in on a range of different ones that might be interesting, including making a trade-off between different designs. So in this project, just like in many aspects of biology and nature, we kind of ran the process and the design approach at multiple scales. So the first idea was to take the outline of the partition and fill it in with these gray lines that you see in the center, this kind of interconnected lattice. But then we could run the process again and make every gray bar in turn out of hundreds of tiny red bars. And we call these micro lattice bars and you see that on the right. So to give you another sense of how this works, once we have the gray lattice, which you see as these bars, instead of using just stock material, we actually recreate each of those bars with a bunch of tiny lattice bars. And like many of our projects were experimental, we're trying to push the limits of some new ideas and some new technologies, but we're also aiming to test it out in the world to really make it and to learn from putting it out there. So in this case, we have 3D printed a full-scale partition with 60,000 micro lattice bars in this new approach that would have been difficult without these kind of approaches. Here you see a kind of close-up of some of the parts coming off the 3D printer. Here's another view of that. And here's some of the components connected together. We've now created what is the largest metal 3D printed airplane component. It's being certified currently to fly in existing A320 planes. The real point of the project and the bottom line is that it's about 45% lighter than the traditional component and also just a little bit stronger. And the real point of this and the impact is that if this is placed into Airbus A320 planes that are flying today, this could lead to the reduction of one million metric tons of carbon dioxide per year. And that would continue every year. Okay, the second approach is what we call biosensing. And by this I mean again using actual living organisms, but in this case to detect some condition of their environment and make it visible, potentially displaying things like environmental quality to the public. So this is a project that began with a kind of public interface to water quality in the East River of New York City. Here you can see a test installation with the Manhattan Bridge in the foreground and the Brooklyn Bridge in the background. This is a real photo, not a rendering. And more specifically, in our test installation, we created this floating network of lights. They were half above water and half below water. Above water we have these dynamic LED lights that can blink and change color. Below water we have a variety of sensors for things like dissolved oxygen, pH, and presence of fish. And in our initial work with this project, which dates back several years, we used electronic circuits, digital sensors, the kind of so-called internet of things as our building block, as our material. But in our recent version of the project, we've been using biosensors. And by that I mean using actual living muscles, the shellfish, to detect water quality. You can see here that muscles, as part of their normal metabolism, open and close their shells a small amount. They're basically eating and breathing dissolved oxygen. And you see them pulsing around here. And the amazing thing is, it turns out that the rate and the amount that muscles open and close their shells is an incredibly sensitive detector of water quality. And that allows us to have this kind of hybrid approach by gluing an inexpensive magnet to one side of the muscle's shell, gluing a $2 Hall Effect sensor to the other side of the shell. And for just a couple of dollars, we have a more sensitive detector, all purpose detector of water quality and water pollution, than a $10,000 dissolved oxygen sensor. So of course, this is good for meeting the project budget and the client always loves to hear something like this or our initially non-profit grant was, our budget was glad to hear this news. But I think in a bigger way, and as we often try to explore bigger ideas through very tangible specific projects, in a bigger way, I think this points to a very interesting and exciting possibility, which is combining artificial intelligence with natural intelligence, combining the best the computer has to offer, this kind of holy grail of AI with a kind of intelligence in biology and in nature that has evolved over sometimes millions of years to perform very sophisticated functions and that has kind of been there all along if only we were to tap into it and partner with it. And so this combination, we think, gives a kind of additive effect or actually even more like a multiplier effect that we think can really enhance the palette of possibilities for design in the future. So here you get a sense of our test installation and the way this network of lights changes colors from a reddish color to a bluish color. And this means that water quality is a little better than last week. That's what the bluish color means. And so there was a tipping point in the data. And based on our initial kind of low budget prototype, we've now been commissioned by the New York City Economic Development Corporation to make a larger, more permanent installation of this kind of network of lights, giving water quality information to the citizens of the city, making visible the invisible conditions of the environment. And that is scheduled to go into construction next year. So the third approach and the third out of three of these approaches that we've been exploring is basically biomanufacturing or biofabricating. And by this I mean using actual living organisms as tiny factories to produce the building blocks of our buildings and cities. This is a project that began with a competition. I'm all mentioned this already. This is for the Young Architects Program at MoMA PS1. And I'm showing this even though it's kind of outdated, there's some aspects of this video that I would revise now if I could, but to give you a sense of the way that we had a kind of hypothesis. We developed an idea for the project and then we articulated it through a kind of animated drawing, which you see here. And then we gradually explored different ways that we might be able to kind of bring this to life, that we could make it real, that we could really test it. So here was the competition video that we created trying to create a hypothesis for how we might approach design in the future, trying to explore a new possibility for a new kind of building block. And this is what kind of allowed us to test out this idea. So the project is called Hi-Fi. And basically the idea was to see if we could design a new building block and a new building prototype with almost no waste, almost no carbon emissions, and almost no embodied energy. And another way to look at the project, and this is an approach that I think is, it's been articulated before for many years and it's seeing a resurgence now, is to kind of design with the Earth's natural carbon cycle, this endless loop of growth and decay and regrowth. And our idea here, our hypothesis, is that maybe we could start with low value raw materials, maybe start with waste instead of plants or precious metals to spend a very small amount of energy converting those raw materials into building blocks instead of what we typically do is spend a huge amount of energy to do that. Then to make a useful structure and at the end of the life of that structure to take all of that stuff, all of that matter and return it back to the Earth and to the carbon cycle. So it's a kind of grand hypothesis, I admit, and a difficult challenge. And the way that we imagined doing this was working with another living organism, in this case mycelium, which is the branching root-like structure of mushrooms. You can see how it grows in these thin white filaments. It kind of fills in space, it develops these interesting patterns. And we were interested not really in the patterns that it created, but in its potential to be combined with other materials, in this case agricultural waste, so not the high value part of agriculture, not corn kernels, which can be used for food, but just the waste of agriculture, corn stover. You can combine those two things together, mycelium and agricultural waste, and in just about five days, which you see here, it grows into a solid object. And the idea was to see if we could kind of harness this natural process of bio-fabrication in order to create building blocks for architecture. So there was one of our early versions of a new kind of architectural brick. Here's the final brick design. And since nobody had really created large scale outdoor architecture out of this material before, we had to do a huge amount of testing, including physical testing. So here we're compressing a single brick, then we also compressed small assemblies of bricks. And it was a very odd experience for us and some of the engineers we were working with to see this material that would compact and compact and compact, but never break. And so it was a kind of research project. We were fascinated by this material, by its potential. We were a little nervous about its potential as well. And we were doing this not only to just learn about the material, not only as research, but really as applied research. And one poignant moment of this project is when our structural engineers at Arup, great collaborators for the project, put our design in their custom structural analysis software, and immediately everyone realized that in any structural analysis software there's no dropdown menu item for mushroom brick. So we had to create a custom material profile and develop the testing, including the physical testing, in order to create that profile and basically determine if the structure was gonna be safe. Here was the first iteration. These red areas are very bad. They basically mean that there was gonna be more than 30 inches of displacement under hurricane force winds. Of course, we never saw hurricane force winds, but this is the new way we have to design structures even in New York City designed for hurricanes. And at this point, we kind of engaged in a pretty interesting design at multiple scales. In other words, we were designing the ingredients of the brick, the different ratios of agricultural waste to mycelium, the different growing times. So in a way, we were almost designing at a molecular scale. We were also designing the shape of the brick, so designing at a kind of material scale. And we were designing a kind of massing the shape of the structure. We were revising all of those simultaneously and in essence trying to get rid of the red so that we could prove to ourselves and to the museum that we could make this structure and make it safe. Another challenge that we faced when designing this project was that the material is not uniform and this is a thread that we've picked up on some future projects. And what I mean is that the material is different on the outside than the inside. On the outside, it has this kind of water resistant protective skin, but if you cut open these bricks, like you see here in this photo, you see that they have a looser, more porous structure inside. So they basically can absorb water if they're cut. And this meant that when we wanted to create our structure with the kind of medium amount of complexity, a complex geometric form, we couldn't achieve that by cutting the bricks on site. And that led us to turning to a kind of problem solving algorithm, which was trying to deal with two challenges at the same time, a fitting problem and a stacking problem. The fitting problem meant that every course of bricks is a different length. And so how do we take units of just a few different sizes and fit them in all these different course lengths? The second is a stacking problem that means that every brick has to sit properly on two bricks beneath it with at least two inches of overlap. And to solve those two problems together, we never could have done that through a single rule of thumb we tried. And only something like a computer model could help us achieve that. At this point, we went on site. We had three weeks to assemble this structure in the courtyard of MoMA PS1. And here we combined into our kind of design ecosystem, a kind of element of knowledge, of intelligence, of labor, of creativity. And what I mean is that we had two different groups working on figuring out how to make this structure. We had Columbia University graduate students in architecture who knew a lot about computer models, about form, about aesthetics. And we had New York City brick masons who knew a lot about stacking things and making them stand up. But neither one alone, neither group alone had ever built something like this before. So we had some pretty interesting discussions and collaborations among these different groups and engaged a kind of ecosystem of creativity, intelligence and problem solving together. And I should note here that everyone who worked on this project was fairly paid. In the end, we created what I think of as a kind of medium scale test of this idea. In other words, it was a 40 foot tall structure with 10,000 bricks, so it was more than one story. It was requiring that we figure out how to make more than a few bricks reliably. And here you see the structure in the context of the glass and steel buildings of Manhattan in the background and the traditional clay brick buildings of MoMA PS1 more in the foreground. The project was certainly a kind of conversation with the natural environment. And it was an experiment in technical performance which I have already described, but more than that. And in addition to that, it was an experiment in a kind of atmospheric performance, in a creative performance. We were interested in what it would be like as a designer and as a human to stand inside a structure made of this new material. What would be the qualities of light and shadow and texture? What would this material feel like, smell like and look like? But of course, everyone who's been to MoMA PS1 in the summer knows that the ultimate test of these projects is their ability to host a party. And this was the first Saturday of the summer where 5,000 people arrived in the courtyard to hear experimental electronic music. And this was thrilling for us, but also extremely terrifying. As we saw these people clawing at our structure, smelling it, walking inside it, walking under it, we were told that the museum would keep people from climbing it, but they definitely did not do that. And we thought, once darkness fell, we were safe. We knew it was gonna work. But really, I mean, I say this and I show this in part to say that this was perfect for us because we were testing our new idea, our new material, our bold idea about the environment and ecosystems out in society, in culture, rather than just in a closed lab on a lab bench and rather than in the corner of a construction site behind closed fences. And this engagement with a kind of dialogue that extended beyond us, that was a little bit beyond our control, was something that was actually a great learning opportunity for us. Finally, at the end of the summer, we took all of the bricks apart a little more carefully than one would typically disassemble a MoMA PS1 structure. We took all the bricks, crumbled them into smaller pieces, combined them with yet other organisms, in other words, worms, bacteria, also food scraps. And in about 60 days, all of the stuff, all of the physical matter of this pavilion returned to soil. And in fact, it was high enough quality soil that we were able to use it, not only for New York City tree planting, but also for community gardens. In other words, it was nontoxic enough to be able to grow food from. So one of the things that we were thinking about by the time this project was wrapping up and after we had learned a little bit from it, was that maybe as architects and designers, we might wanna start thinking about designing to disappear as much as we typically think about designing to appear. Okay, with that as a kind of rough overview of some of our recent research and some of our very specific approaches that we've been developing, I now wanna turn a little bit and talk about books. So it's through writing and publications and books that we can operate at a kind of different register. Books give us and writing gives us a different perspective. They move at a different pace. They allow us to study more deeply, to explore in a more relational way. And we have two books that we've created recently that are kind of interconnected with our work. The first, as Amal mentioned a little bit earlier, is called Embodied Energy, Making Architecture Between Metrics and Narratives. This was published recently by Columbia Books on Architecture in the City and Lars Mueller Publisher. And it includes an amazing range of contributions from architects, engineers, material scientists, historians, and curators about the concept of embodied energy. The book really starts even before Embodied Energy with the idea that buildings account for about a third of global waste, energy consumption, and carbon emissions. Embodied Energy defined as all of the energy required to extract, produce, transport, and assemble materials into buildings has actually been increasing over time. So if you look at the total amount of energy consumed by buildings, it's made up of operational energy and embodied energy. Operational energy you can think of as basically heating, lighting, and cooling. Embodied Energy is basically all of the energy to make the materials, the physical stuff of the building. And you see on the left here in this graph that in the 70s, about 10% of total energy in buildings was embodied energy. Today, that's about 40%. So it's clearly an issue that needs to be reckoned with. On top of that, as one of the contributors, Michelle Addington has noted, by 2050, about a third of global carbon emissions, and by rough association, a third of global energy consumption will come from making, not operating, our buildings and cities. So that's a third of this global thing that we realize is a huge problem is coming from basically embodied energy. So embodied energy is clearly a big deal and it's increasing. And this book and operating on this register allowed us to explore how energy consumption is just one of the many important invisible consequences of making buildings. Another consequence, of course, and this is probably obvious, is the mobilization of labor. As the advocacy group who builds your architecture has emphasized and you see their diagram on the upper left. So clearly these invisible but important effects are important. Clearly embodied energy is something to be reckoned with. But how do we draw it? How do we design with it? Well, we started by creating some data visualizations. Here we're breaking down embodied energy by materials such as aluminum and glass, by building systems such as structure and envelope, and by phase such as extraction, transportation, production. And we're also comparing buildings in some case studies. But we realized very quickly that data was not enough. The quantitative was not enough. We also needed to understand the qualitative and relational aspects of embodied energy. So at this point we moved to drawing material stories. The kind of narrative, the story of a material like concrete or wood or steel from extraction to factory production to transportation to construction. Trying to draw all of that in a single frame to try to understand what it means and how we might intervene, how we might design that. And so here are a couple of those drawings by an incredible architect, Lindsey Wickstrom, who worked on the research of this book and then has created a set of drawings based on them. But more than this, really, the book is a lens through which we might see all kinds of invisible but essential features of architecture, such as embodied water, embodied carbon, and embodied labor. And it may allow architects not only to discover new ways of building, new ways of making stuff, new ways of mobilizing this material, but potentially also new forms of creativity. So the second book is a monograph of the living called Now We See Now, which is about to be published by the Monticelli Press. And this book involves two volumes that describe two sides of the work by the living. The first is projects and the second is research. And you get a sense of some of the flavor of our studio and some of the pages, but the real meat of the book comes in these two sections, volume one, the projects, and volume two, the research. So the book includes projects, a variety of projects, which we can kind of put together and compare in something like a book. And one of the projects shown here is Street Life for the Shenzhen, Hong Kong, Biennale, a few years ago. For this quick and low-budget project, we combined two unique features of these cities of Shenzhen and Hong Kong. The first is eating out on the streets in food stalls that pop up in the evenings called Daipai Dongs. And the second is the manufacturing of electronics in this region, from high-end cell phones to low-end LEDs. And our idea for this kind of quick project was to combine those two things together, to combine street food and electronics by inventing a soup bowl with hidden messages at the bottom. So you can see that we use CNC milling. You see this in the kind of upper left here to carve, not for making form, not for kind of the traditional version of fabrication and architecture, but actually to take a white plastic soup bowl and carve a very thin pocket out of it in order to fit an LED light badge inside. And this allowed us a kind of magic trick of having a soup bowl that looked like a normal soup bowl used at these Daipai Dongs, but it actually had this secret message in the bottom. So at this point we hacked some scrolling LED badges from a factory in Shenzhen and asked the curators to compose short messages that were provocations about the city. You can see a couple of them on the left there. And then the idea was to partner with several of these local Daipai Dongs and use these secret soup bowls to serve unsuspecting customers and put these messages from the curators in the bottom of their noodle soups. So here you see a kind of before and after. This was a project that was at once personal and public. It created a strange micro exchange, a strange kind of dialogue between international architects and local citizens. And maybe most importantly, it brought the Biennale out of the civic hall and the galleries and into the streets of the city. So the second volume of the book is about research. And I'll give an example here as well. The book includes different research documented in a hopefully a reusable way, sometimes in research papers. And the example that I'll show here is our technical paper about Twin Mirror, an installation for the Seoul Biennale. And this paper really takes a critical look at some of the technology behind applications like surveillance, traffic flow, public health, and image detection. Like all technologies, of course, these ones involve assumptions and biases, but the assumptions and biases behind machine learning algorithms like these can be especially troubling. And as writers like Kathy O'Neill and Kate Crawford have pointed out, they can lead to effects like racial discrimination and judicial sentencing, gender discrimination and job listings, and uneven distribution of urban resources. So in case you don't know, machine learning is a kind of computation that derives conclusions without being explicitly programmed. And for this reason, it is opaque, sometimes even to its own inventors, and it tends to overemphasize past data, and it can lead to perpetuating past results and also to things like discrimination. So in this project, we use some of the same machine learning and facial recognition used by companies like Google and Apple, but also by many governments these days. And our idea was really to see if we could expose some of the inner workings of these algorithms to make them more legible to the public and to reveal the hidden assumptions and biases in these processes that at first glance might seem to be neutral and objective. The paper describes how we created two different machine learning models with two different sets of data, and then processed individual faces in two different ways, according to these almost two versions of reality. So we showed these two versions of reality to the public in real time, revealing differences between the two different models, and also revealing, we hoped, the biases of each one, of each model. The results were visible for each person and also for all people as a collective. They were shown both in the gallery, in a gallery setting at the Solbian Alley, and also outside on the building's facade. And here you get a little sense, let's see. And here you can see the project in action. As visitors approached a display, they saw themselves presented through two different versions of AI, two versions of reality. More specifically, we created two different machine learning models based on two sets of data. You can see here the projection of those two different sets of data. The first is a set of data that's become a kind of academic and industry standard for facial recognition. It's called labeled Faces in the Wild, and it consists of about 13,000 images, it's 6,000 people, who are mainly politicians, athletes and celebrities. So this data set is obviously biased because of the kind of people included. The second data set is captured from the people who visit the exhibit. So this data set is more reflective of the local community and soul, but it's also biased because of the types of people that it includes and represents. The exhibit shows these two versions of reality through a kind of twin mirror. Neither version of reality is neutral or objective. Both are biased, and by seeing them together, maybe we can start to get under the hood a little bit of these technologies to apply some judgment to them to become a little suspicious at time to recognize their flaws. In addition to the gallery component, we also showed this project out on the streets, on the facade of the building where they were being displayed inside as well. And this allowed us to kind of speculate about the facades of the future, to have a dialogue with the urban life of soul, and to kind of contribute to and make a speculation about the increasing amount of digital billboards and dynamic LED facades in soul and in other cities. So our research papers, just say this one last thing about them, allow us to kind of document our projects in a way that you might not get from a lecture from a magazine article. And we deliberately try to focus not just on photographs of the completed project, but on open-ended, almost open-source ways in which others can take some of our research findings, some of our hypotheses, some of the techniques, tools, and ideas that we've used and build off of them in their own way. So in that sense, we think of this book not just as a documentation of some of our explorations, but almost as a field guide for others who might wanna take some of this in their own way. And so finally, I wanna talk a little bit about buildings. So in addition to our applied research, in addition to our books, our studio engages in more traditional architectural processes and in designing and constructing buildings. So while we are experimental, while we are exploring the limits of what architecture is and might be, while we believe that architecture does not need to be a single, static, solitary building dug into a fixed site, while we believe that architecture should expand in space and time to address a broader ecosystem with a broader design palette, we are still interested in creating buildings. And I'll quickly show two projects that we designed recently and that also recently completed construction. The first is a kind of automated office design. So this is a new office space for 300 people in Toronto. And the experiment here was to kind of test some of the limits of so-called generative design for architecture. In other words, we started by defining high level goals and constraints and then used the power of computer automation to explore a wide design space of potential options and to see whether we could achieve a kind of hybrid of the human intuitive and the computer and the automated. So this video gives a sense of how our process worked. We started with, as all projects, the brief which was three floors of an existing building in a new discovery district in Toronto. We had a fixed amount of meeting rooms that we had to fit in there. We had other equipment and of course we had people, 300 different people. Then we outlined some measurable goals for the project. Two of the goals involved getting data from people from the future occupants of the space and a series of other goals were more global. So we had some individual goals and some global goals. And here I'll go through the steps of the project and the use of generative design a little more specifically. The first step was generate and that's basically to create a geometric system that allowed us to explore a wide range of potential layout options. So we took this linear space, created this dashed red line as a kind of central spine, created a system where we could create more or less dots which equal neighborhoods, work neighborhoods. We could move those dots to certain offsets and then reconfigure the space and this allowed us to create some neighborhoods which focused outwards and some that focused inwards. Then step two was to create the goals and here we had six different goals. Two of them were based on asking everyone of 300 people what they wanted out of an office. So if we asked for their adjacency preference, in other words, what other people, teams and equipment they wanted to be near and we asked for their work style preference. Basically would they like a kind of quiet and heads down space or an active and social space? Then we measured a couple of global conditions such as buzz or interconnectivity. We wanted there to be places where people could interact with each other, exchange ideas. We wanted there to be things like good daylight. We wanted to avoid distraction. And finally, we wanted there to be good views to the outside. So this was basically a process of translating with the stakeholders in the project things that they wanted out of the project into measurable scores. Once we had done those first two steps we could basically evolve designs. So here we could use computer automation to generate literally thousands of design options. You see them here again as dots. This is a process similar to what I described before with a component for an airplane. But this is a more multi-dimensional data set. Here we have six different objectives. So no single graph can incorporate all of them. And you need to kind of sift through the data in a more kind of creative way than just a bottom line engineering approach. In the end we were able to create a design with all of the required facilities and people and layouts. But that also represented a kind of twist to the typical approach for designing offices. So at this point I wanna note that this process allowed us to have a kind of hybrid of the human and the computer. Ideally to go beyond some of the blind spots and rules of thumb and linear thinking that are typically used in designing an office space and to discover new possibilities and really start imagining a new type of design. But I do wanna note that there are a lot of possibilities for the use of generative design. I'm not advocating for all of them. And I would like to explain a few of the different ways that we use generative design that I think are different than some other uses. First we acknowledge that there are many assumptions in our geometric and scoring algorithms. Hopefully they're good assumptions but we acknowledge that we have assumptions and we have to design those assumptions. There's nothing automatic about them. Then we deliberately engage qualitative features of designs. So not just the quantitative but we engage the qualitative sometimes by incorporating them in the model, sometimes by building in evaluation outside of the model. Then we use the process to structure a discussion and a debate among stakeholders not simply to automatically accept a so-called optimal result. And finally we use this process to manage complexity in a way that allows us to go beyond some of the default rule of thumb one size fits all design approaches especially as they've been played out recently with the design of pretty typical formulaic open office plans. So this was an approach that was again experimental but again we tested it out in a kind of practical way through building it out in the world. The second building and this is nearly the last thing I wanna show is another project that Amal mentioned. This is a project that is really about an open source building. This is a building that explores our vision for living breathing architecture. It's the embodied computation lab at Princeton University and it's a new building for interdisciplinary research on robotics, sensors, and everywhere that computers meet the physical world and become embodied in computation or become embodied computation, examples of computation physicalized in the world. This is a building that's located near the new chemistry building and the athletic facilities. You see it on the bottom right. The main architecture school is the red shape on the top. It's a pretty interesting site with a great history of architectural research. It's where Buckminster Fuller made his first geosphere. It's where the Olgie Brothers did their pioneering work on solar analysis and it's where Jean Latitude, the former dean of Princeton, did some very interesting work on architectural camouflage. The program for the building involved a variety of learning activities but it was more or less a kind of open-ended space for prototyping ideas for architecture at one to one scale. So it involved assembly, testing and experimentation and also exhibition as well as things like loading and preparation. But most importantly, the real program for the project and the real brief required a kind of flexibility to adapt over time to new research and new equipment. And this basically led to our concept for the building which was to create an open-source building. In other words, like open-source software, this building would be released as version one, fully functional but a kind of prototype. And it would also like open-source software be deliberately designed to be changed over time by multiple authors. So what would that mean for a building? For one, we designed a building that was deliberately incomplete. So if you look in the bottom right, you can see that about a third of our structure was an open frame, an incomplete building. Well, two thirds was a kind of typical enclosed volume. And this was basically so that the researchers in the building could build one-to-one prototypes of roofs and walls and test them out and kind of complete the building over time. The building also had a number of swappable modules and systems such as radiant heating on day one that uses as a heat source the waste heat of the building next door. So we're getting heat for zero energy expenditure. But in the near future, this can be swapped in a kind of plug-or-play manner. And the source of the heat is expected to be a geothermal well that the researchers are developing. The building is also rewritable over time. So just as biologists might use a microscope to study organisms, architects will use this building to study buildings and they will add to it and change it over time. And finally, this is a building with a number of different physical sensors and a building that's designed to learn over time. So like living organisms, it continually processes information such as light, heat, humidity, occupation, and it adapts to changing conditions. Here are a few of the sensors that were installed by Forrest Meggers, one of the researchers who's using the building. And there are a variety of different ways in which this building is creating a kind of continuous data stream to be analyzed and experimented on. The building also has several unique sustainability features such as the floor that I described before. It's used for both radiant heating and radiant cooling. Here you can see it in the winter with heat as one of the invisible but clearly important features of the building. And you may have not known this previous to this kind of photo, but the faces, the people are the hottest thing in the building. Most thermal images don't show people. The building also has a structure made of engineered timber rather than steel even though it's a kind of challenging structure to make because it has a five ton gantry crane. And the building finally has a facade made of salvaged New York City scaffolding boards. So it's, the facade is made of a waste material that would otherwise end up in a landfill. So as you probably know, most construction sites in New York City, especially renovations use these scaffolding boards here. You can see some workers passing them all the way up to the top of the scaffolding. But what you may not know is that these boards are typically used for a year and then literally thrown out in the landfill. So these are two inch by 10 inch boards. They're very thick, very good wood, but because of a rule of thumb and because some of the boards, the worst performing ones develop cracks and warps, the basically standard practice is to throw this material away. And our idea was to kind of rescue it from the landfill but also to kind of use it as a new kind of experiment. So one way of describing this is that the final layer of constructing the building, the facade, was the first experiment of operating the building. We started experimenting with these used scaffolding boards as an interesting bio-material that's non-uniform and with a hypothesis that this kind of wood that's aged, that's kind of decaying a little bit, has micro-contours that might offer benefits for performance. It might offer thermal performance by performing as a kind of invisible jacket. So if you look closely in one of those notches in the grains of wood, you can see there's a little space and we believe that that space is going to trap air and act as a kind of immaterial, invisible parka or jacket for the building. These are some experiments performed by forest maggers. Again, one of the researchers. And we also thought that this material, especially when aged in this way, might have some hydro performance benefits in terms of shedding water. But at the same time, we knew that different zones of the facade had different performance requirements and we knew that we had 960 salvage New York City scaffolding boards with different properties and we had to decide how to arrange these different boards on the facade as part of our experiment. So at this point, we turned to computation again to help us manage the complexity and to engage a kind of new experiment and to fine tune the design. So we knew that areas of the wood with the narrow grain performed best in terms of thermal potential and hydropotential. And we knew that the regions around knots in wood tended to have narrow grain. So we set out to identify and treat the knots in the wood in each board through a process, admittedly odd process of both machine learning and CNC sandblasting. So I'll show you a little more about how this works. This was a process that was trying to explore how machine learning might get physicalized in the world in this embodied computation lab. And we started by photographing every single board, all 960 boards, and asking a human to look at small tiles of photographs and click a single button, either not with an N or not with a KN. So humans were just looking at boards and rating them. And this is what a lot of people are doing for all kinds of machine learning applications. It's the same technology that allows Google to know if you're looking at an image of a cat or not. And the really amazing thing and kind of spooky thing was that after training with some samples, the machine, the algorithm got very good at not only looking at a new photo of wood that a human had never seen, but then determining whether it had a knot and not only whether it had a knot, but where the knot is located. So in this view, you see kind of from left to right how the algorithm is honing in on the spot in the photograph where it thinks the knot is. And it's surprisingly good at doing that. At this point, we turned to a version of fabrication that we think is pretty interesting and pretty relevant to biomaterials and non-uniform materials. And that version of fabrication is sandblasting. So unlike CNC milling, which kind of eats away indiscriminately at something like wood, sandblasting eats away just the soft areas of wood and leaves the hard areas of wood. It basically accentuates non-uniform materials like wood. So we developed a process that would take boards from surveying this unusual non-uniform varied material, processing it through a series of things like photographing, algorithms, and then eventually sandblasting just where we thought the most interesting areas of the wood and the best performance benefits would be which were the knots. So the process for each of the boards looked something like this. The custom machine and control software looked something like this. So you see this sandblasting machine that we've kind of hacked and adapted, this big gray thing. Here you see one of the early pieces of wood on the right and the kind of contraption that makes it possible on the left and the two people that make it possible as well. We invented some custom software to help do this. And really more than anything, this allowed us to experiment with a new approach to materials that was both computational and biological. So it allowed us to process units of material individually. Maybe we no longer have to think of bricks or boards as being entirely uniform. It allowed us to think about different regions of the facade and maybe targeting based on different reasons you'd wanna do one thing in one place and one thing in another place. We could use the power of computation to explore a variety of different options. But really I think the real point and the interesting aspect of the project is that it allowed us to think about overcoming a kind of lowest common denominator, one size fits all approach to materials. And that's something that we are continuing to explore in other projects. So the facade itself as we're exploring in this drawing registers the story of a material from resource extraction to initial use as scaffolding to computational analysis of unique features to selective fabrication and finally to reuse as a building facade. And more fundamentally this project and this process has allowed us to develop a new perspective on building materials and on buildings really as a temporary formulation of energy, labor and matter connected to other formulations before and after the life of the building. So the design of the facade itself registers at several different scales. Here you can see some of the incredible topography of a single board exploring the notion that each piece of wood in itself has a story. It comes from somewhere, it grew in a specific way according to specific environmental forces. Here's some of the boards installed on the building offering a kind of unique appearance, an aesthetic and also an experiment in building performance. And in addition, this is an example of what we could describe as embodied computation of the current state of artificial intelligence of machine learning 2017, including all of its false positives and false negatives. We left them in the project as you can see here. And also kind of exploring all of the promise and also perils of this technology. A couple other images of the building. Here's a view of the gantry crane in action. Here you can see two large hangar doors 37 feet wide on either side of the building that open up to adapt to different research scenarios. Here's some of the research equipment for the building including multiple industrial robot arms that collaborate with one another. And in the end, while the building itself is a kind of research project, as I've described, we're also interested in design, in atmosphere, in the experience of space and in what it might feel like to live in a strange new architecture designed for both humans and machines in the context of the natural environment with a vision for a new form of living architecture. And finally, I'd like to describe our most recent project just completed still in process in a way. And I'd like to use this project, which I'll just take a couple minutes to describe, to kind of come full circle. And I'd like to show that the buildings go back to the research. In other words, we think of the design ecosystem as a feedback loop. It works in both directions. It's a cycle with applied research influencing buildings but also buildings influencing applied research. So in one way, this is a project that begins with the physical material from a previous project. You'll see the sandblasted wood here. And with an open question about how the material will perform. And in another way, this project begins with new ideas about ecosystems and public health. So this is a project called Subculture with two good friends and colleagues, Kevin Slavin and Elizabeth Hennaf. And it builds off of some of the amazing work that they and others have done at the intersection of biology and cities. So we started with the hypothesis just that just as we are increasingly aware of the bacteria in our own bodies and the way a gut microbiome contributes to individual health, we might also start paying attention to the bacteria in our cities, not just our bodies but our cities and the way an urban microbiome contributes to our collective public health. In other words, microbes are all around us even though they're invisible. They're in the air, they're on our food, they're in our architecture. And most of them are beneficial to us. So this is the opposite of our fear of microbes for the past 100 years. And it represents a dawning understanding that microbes actually keep us alive. This is also related to a new field called urban metagenomics, the study of all the DNA in a physical environment like a neighborhood. And we believe it as the potential to transform the way we think about and design for life in the city. So in this context, for our installation at Storefront for Art and Architecture, we developed a project with two goals. The first was mapping, prototyping a new method for documenting the microbial environment without the traditional academic or commercial laboratory. And the second was modeling, exploring design scenarios for architecture that promotes microbial life. In other words, we wanted to see what was going on with the microbiome of Storefront in Soho and to compare it to the microbiomes of places like downtown Brooklyn and the Brooklyn Navy Yard. And at the same time, we wanted to explore the implications for design. In the future, will we design architecture with microbes in mind? Could we establish microbial metrics much as we have standards for measuring structural integrity, thermal conductivity and ergonomics? Will we be able to use buildings as sensors for public health threats? And might we be able to design probiotic architecture to improve public health? So first, we developed a bioreceptive material. This is the kind of modeling part of the project that was deliberately designed to catch and host microbes. By bioreceptive, we mean the opposite of sterile architectural materials that have dominated for most of the past 100 years. But rather, our bioreceptive material was meant to be a material calibrated to select for diverse and beneficial microorganisms. And of course, this material in our initial experiment was wood. Wood was once alive as a tree, of course, hosting many other species from beetles and worms to fungi and bacteria. And even in its inert state as a building material, wood is well suited for hosting microbial life. At Storefront's gallery space, the panels of the facade were transformed with wood tiles cut from standard lumber and deliberately eroded through sandblasting at various depths to create diverse microclimates. Each microclimate has distinct grain and knots that create diverse pockets of shade and moisture. And this allows us to see that even the most common and humble building material, which we can imagine to be a four by four post of Douglas Fur from Home Depot, the kind of basic building block of a lot of current American architecture, even the most common building material has a material story to be uncovered that may offer a range of visible and invisible effects for this new bioreceptive area. So after setting up this idea of a bioreceptive material, then the second part of the project was basically metrics and measuring this material. So here you see a series of steps that we engaged by moving Elizabeth Heneff's lab from NYU Tandon School to the Storefront Gallery. And these are steps for basically sampling the DNA that's collected on this bioreceptive material. The first is to swab wood tiles with a Q-tip to collect the surface microbes. Then we place the liquid solution we place the swab in a liquid solution and break down the cell walls. Then we add identifying markers to the end of the DNA molecules using some buffer solutions. And this allows us to create basically a pure DNA solution. So this is basically all of the DNA that's collected from that piece of sandblasted wood. Then we load this pure DNA solution onto what's called a flow cell to be processed by a portable DNA sequencer. So you can see that new technology here. And this is part of what I mean by saying this is a new biology, this is not the Darcy Thompson version of biology anymore. And here you can see in this video shot by Elizabeth with her phone that this DNA sequencer, this portable DNA sequencer just goes to work. So in something like four to 24 hours it will pull strands through the pores of the flow cell converting the electrical charge of passing base pairs into sequence data, the A, C, G and T of DNA. And so the third part of the project is basically three parts. We created this new material, then we tested this new material and then we explored a kind of visualization of this new material. And here you can see a kind of automated process that we created that will basically take the DNA sequencing data from three different sites and automatically draw the output as an early and rough map to an uncharted territory. So the map is raw and it is at a coarse resolution like our understanding of urban metagenomics itself at the moment. But it already allows us to see things like the presence of heavy metals in the Brooklyn Navy Yard via microbes there that perform the function of breaking down heavy metals and also to see the presence of rice DNA at storefront and to start asking questions about that and interrogating that deeper. So in the end we're using a gallery installation as an active living experiment. We're exploring a new architectural envelope and a new way of seeing some of the important matter and organisms that are all around us and keep us alive. We're experimenting with reframing buildings as stewards for the urban microbiome. And the initial evidence indicates that our collective future, all of us, multi-species included, is a lot more collective than any of us can probably see or imagine. Thank you. Thank you, David, for this kind of mind-boggling tour of possibilities. I think your work and your teaching gives a pretty sort of clear sense of what we mean today when we mean, what we mean by not only expanded practice but a research practice and a kind of feedback loop between research and building and practicing. And although at times when we speak about an expanded architectural practice, there's a sense of, or there's been a sense of kind of explosion where everything can be architecture and it doesn't matter whether someone is writing, researching, building, designing an installation or an exhibition, that all of that can be considered architecture. I think what your talk tonight and I think where your work is going to is almost sort of expanding to re-enter architecture in new ways. There's a kind of real kind of re-convergence of the work in kind of redefining very fundamentally the different elements of architecture, and of course when one thinks about biology and you mention it, there's a very long legacy of architecture and architects engaging with the biological from Frank Lloyd Wright's kind of organist designs that were meant to come in a reaction to a kind of industrialized idea or a reality of the world of manufacturing of this kind of repetitive aspect of efficiency that came with modernism. Whether it's the metabolists in Japan who thought about the city as a tree and kind of worked with these metaphors. You mentioned Fuller, of course, who looked at or we can think of Christopher Alexander and thinking about patterns or even Greg Lynn who started here at the school. But in fact, I think across all of these approaches was a sense of metaphor, like the biological, a kind of, and yet architecture remains static. It was just sort of mimicking and even when it was pretty complex as we got to with some of the kind of organist, let's say forms of a Greg Lynn or it remained nevertheless metaphorical. But I was thinking then not so much about let's say the Greg Lynn trajectory but the Frank Gehry trajectory who took these explorations but used computing to actually change manufacturing and production and suddenly forms that could have been imagined but not built were able to be built, right? And that became really part of the practice and produced a whole new language, architectural language built in the world. And I see your work now more in that light of, it is of course, as you said, no longer just about the metaphorical or the kind of sort of pattern learning from but rather literally the living, integrating, living, breathing, ingredients, materials within the buildings themselves. So kind of moving away from, it gives maintenance a whole other dimension and now there is a sense of engaging and working with aging or breathing or growing. All the things that we've tried to keep away from architecture, moist, mold now are kind of swallowed into and becoming part of it. And so that's kind of one aspect which I think is quite different from the way we understood architecture and biology before which is not the kind of formal aspect. But then the other aspect, nevertheless, which I think for me is the loop is what you start to do. A lot of the projects are exploring certain parts but then at the end with the laboratory and now the installation at storefront, you made a case for what you said initially which is optimization is not always good design and suddenly you're coming with your judgment as architect, as designer in terms of bringing, let's say design sensibility and a formal sensibility to performance. The two things are brought together rather than an either or scenario. So which I think that's a new door and I kept thinking with the image of storefront at the end, Stephen Hall who also teaches here but he's not here tonight. But it was such a radical project when it was first built with ideas about construction and materiality and now kind of becoming this wooden, living, breathing organism. His facade was intended to be a kind of mechanical facade of transformation and now you've replaced the mechanical with the biological transformation that your walls are producing. So the shift from the mechanical to the biological is also something that you're kind of pushing forward. So anyway, all that to say maybe first question which is in what part do you think, it's very clear to me the part that biology is adding but do you think that computation is also, I mean obviously, but the kind of iterative, for me that's more of the, similar to the past and the biological is really new and that the partition for Airbus is you're using computation to sort of iterate and find the most performance, the highest performance but then using, so that's more like what we used to do in the biological and the breathing and the living and the decay is a kind of new aspect. So I was trying to kind of maybe draw the line between what is still within the world of computation as a way to enhance performance and then the world where biology form and performance are coming together in new ways in the work. So, it's not really a question, it's a long day. Yeah, I mean there's a lot in there but I think they're both relevant but I mean it's, so a couple thoughts. One is it's true that some of our projects have a much clearer lineage and should just be considered in a certain trajectory of say computation, some maybe in biology as well and I think that's a good point about Stephen Hall's design for storefront. That was, no, I don't know. But I'm sure that he imagined bringing that space to life by a kind of dynamic version of architecture. So in some ways that's a lineage as well. I mean he's not the only person who's done that but it was a good crystallization of that design intention. So and of course the bionic partition working on an airplane component and trying to make a lighter weight component, that has a clear trajectory and lineage. What we've tried to do in a couple of ways, sometimes minor and sometimes major, is kind of poke around for new opportunities. So I think by combining a kind of biological algorithm with this optimization process of efficiency, there's the start of something new that's possible. It's one of the things that we discussed with Airbus engineers, which was fascinating, was that they couldn't look at it and understand it. So what does it mean when our designs can get so complex that we can't understand them with all of our human tools but we have to look at them in a different way like through data. But I think you're right that the more fundamental way that we're exploring new possibilities is through using actual living organisms. I stated that several times, you're probably sick of hearing me say that but I think that's pretty different. So designing for other species but also with other species. And I think that's still relatively unusual. It's always been possible but it's possible now in new ways. There's, I think, in a way, much more of an acceleration of biological technologies than computational technologies in the past 10 years. But I think the hybrid could also be interesting. And the last thing I'm reflecting on based on your comments and question, I've never thought of it this way before so you'll have to tell me if this is a good concept, is that I think we're taking, we're a little bit practical as I've discussed with you before. You know, we're strangely like a little bit radical but a little bit practical. Over time. And I think what we're doing for the most part is trying to change the system from the inside. You know, so there's kind of models and politics for that kind of thing. And so that involves sometimes some compromises and it involves sometimes more constraints than we would want. And sometimes having to push and not get what we want because we're working with clients and other people. But I think that's work that's important for somebody to do. Somebody should work to change from the inside. And so that's how I feel like, you know, a lot of these technologies I think are a little dangerous, are a little suspect, should be criticized. We should interrogate them and be skeptical of them. But at the same time, we need someone out there who is getting under the hood and trying to use them, you know, in interesting ways and in beneficial ways and in humanitarian ways. And so that's I think a role that we've, you know, started to play, to know enough about the conversation about optimization or machine learning or biotechnologies, much less gene drives, which we will talk about in another session. You know, so that we can be there in the discussion because it's happening without us anyway, or it will happen without us. So we need to be part of that discussion. I mean, that's a position and that's one that I think we're taking and of course, you know, there can be criticism on either side of it, not being practical enough or being too practical or making too many concessions. But that's a kind of line where we're trying to tread. Well, a few things though about that. One is I think the way I hear this, for me, resonates a lot with some of the conversation with students or maybe this generation, not to put it necessarily only in generational terms, but where there is an interest to kind of be very real and focused and intervening in the world and finding ways to act at whatever scale. I mean, certainly with the incubator, we're seeing that quite a bit with the members experimenting on even designing an app that can change one thing. So I think that level of engagement, you know, versus a sort of kind of staying within a kind of disconnected, let's say, maybe more autonomous kind of position. But I also think that it's interesting when you say from the inside, it's literally from the inside. Like you're literally getting on the, you know, in terms of like what makes the wall section? And for a long time, we didn't care of what is inside the wall section as long as the wall looked a certain way. So you're kind of entering that space which architecture or architects maybe left behind for some time and re-entering and rediscovering the making literally of that wall space. And again, I kept thinking about, you know, in some way some of the explorations that took place in this school, like with people like Greg Lynn, et cetera, maybe did not engage enough in the building. In fact, it stayed kind of very theoretical explorations that were very formal and limited to a kind of academic environment and sort of hit the wall and the construction industry and, you know, ways of making things where, you know, still things were layered. You know, and there was no way to create this homogeneous, smooth space that everybody was talking about. So this kind of literally getting dirty and understanding those layers and constructing them from the inside makes, you know, it's quite kind of radical, I think in that sense the practice, not only the projects have this kind of radical proposition. So I wanted, you know, in this spirit of kind of engagement and building, through building and making, what about scale or, you know, what are the parts of your research or practice that you think can scale up faster than other? Or is that a concern? Or, you know, you're starting to, it's clear that now the project exists more and more at the level of the envelope. Certainly a lot of the creativity in the Princeton lab happened at, you know, at the kind of, at that surface, right? Between inside and outside. What are other areas or how do you imagine scale or? Yeah, I mean, I think I've increasingly been thinking about this, you know, in our practice, but also with students here, that there's some different versions of scale. One is like the size, but one is like the impact. And so our, you know, I think by making some buildings recently or, you know, engaging that kind of level, not just an installation and prototype, but a building, you know, that allows a spectrum of size scales, you know, from the material, you know, to the assembly, to the building. And I think we've always been trying to engage a kind of ecosystem or kind of global flow scale. In terms of impact, I mean, that's a good question because although we're a research practice, I think we're also a kind of practicing small firm. And, you know, that may sound generic, but I actually think it's kind of rare to be, you know, if you go to some of the conferences that we go to to present papers, and if you engage like some of the research aspects of, you know, what academic peer-reviewed culture we have in architecture, which is not a huge amount yet. Most of the people there are not practicing. And most of the people practicing, good friends and peers of ours are not putting as much effort into this kind of reusable, you know, design intelligence idea and the papers. So I think that's something that we're going to have to, you know, I would like to keep those both going, but I think it's going to be challenging. You know, so there's a sense of scale in terms of impact, in terms of doing more projects, making more buildings, maybe having the size of the firm grow in the way of, you know, a small firm, you know, those kind of questions a small firm asks. And I think we're starting to do a little bit of that but at the same time, I think like by our essence and even by our name, I think we're going to need to stay involved in the research and the prototyping. Which is maybe the most difficult thing you'll ever have to do is being in control of the scale of your practice and I think that's, so a different question I had was in terms of the office space you designed. So that had an aspect that you didn't, you mentioned, but it was this kind of participatory almost, right where it seemed like people were quite engaged in. Was that, do you think that the result is different? How different, that's the one project that I was wondering, like how different is the result from a design where you would say, okay, we need small meeting rooms, medium meeting rooms, large meeting rooms. Which part you felt, and it's hard, which part, okay, this is really a unique discovery that I wouldn't have made without the process of participation and computation and kind of parametric, you know, sort of ranking the different options. I mean, I think I'm gonna, one way to reframe your question, which I think is valid and pointed, and I think, I don't know if you were exactly thinking this, but I've asked myself this question, it's like, doesn't that look too conventional? Or not just look like, if there was one discovery that you think wouldn't have, you know, like Europe would have, you know, you wouldn't have, you know, like your process versus making 100 foam models, and then deciding, oh, that one is the better one. So I just... But I'll tell you, because I think it may well look pretty conventional, but what's different about it is one of these things that I was describing, you know, in terms of embodied energy, is the thing you would never see from the photographs, which is that every single person is sitting in a space of, you know, sitting in a space that was, that acknowledged the data that they input into the system. I'm almost certain that that doesn't happen in 90% of the office spaces being designed today. So you look at the photographs, and it's another way in which our architectural representation is not really in line with some of the things that we're experimenting with. Now, it was the first time we had done that, in my knowledge, the first time that something of this kind of scale that was built that had this amount of complexity was tried. So I'm not certain that every single person out of 300 has a better work environment than they would have had in the one-size-fits-all approach, but I think that was more of the hypothesis. Could we design at this higher resolution with this degree of complexity with factoring a lot of people in, a lot of their desires in, not just through a focus group and a kind of summary at the end of what people want, do they want an open office or not, or half and half, but a very fine resolution consideration. So I don't think there really yet has, I don't think there was a discovery in terms of the formal arrangement. I mean, we decided that we would allow for this variety and we created a variety, but that's not particularly new. But the one other thing that I thought that we did discover with the process that I think is relevant to using this kind of approach and this technology is that you ultimately need to have a kind of discussion and a human value and a selection of the design that you wanna build when you have any problem with more than two dimensions. And it's especially poignant when you have a lot of dimensions or medium amount like six dimensions, six different goals. And that allowed us to test out this idea of taking some design options to a diverse group of stakeholders, managers, just lowly employees, even people who represented the building because this was just a tenant that we were working for, and have a discussion informed by and based on data, but also to have the discussion at all on a kind of common ground. So I think that is one of the potentials of using this technology, and it's different than the blue foam model version of iteration is that we can start as a designer, as an architect going in and saying, here's this range of options. We know that everyone has slightly different values and priorities. Here's a way to have a starting point for what will always be the human messy discussion and debate about what is the best design and what is most important. So I think that is one of the true potentials of this thing called generative design. And that's a slightly different version of it than is kind of being discussed and analyzed by most people these days. Well, I wanna leave room for some questions, but just one last thought on that, which is I think that project itself begs the question of visualization, which we've been talked about earlier with the open house, which is when something isn't immediately visible design-wise, how do we render it visible as architects so that it registers as architecture, as an architecture intervention? So I think that's quite interesting. So you should visualize that. Yeah, well, drawing the invisible, that's a great idea, of course. We have time for a few questions. Thank you, David. Thank you. It's fantastic to see it all kind of in one place and so clearly explained. I mean, I really admire the thoughtfulness and the clarity with which you explain your work, which is also rare. So kind of in that spirit, since you started us off with the Pareto curve, one of the great sort of, as you hinted a little bit, one of the more deadly instruments of the 20th century, even though it is so important for the later histories of feedback-based thinking, systems thinking, and so on. I wonder if we could go back to that for just a second as a principle, not as a sort of, not so literally, except that, but reframe it a little bit quickly, that one of the great things Pareto does is he poses this in the form of a question of taste. There's, you know, bread along one axis and wine along the other axis. So not just taste, but maybe even Christian taste, perhaps even Catholic taste. So it's highly situated in the way that I think you're alluding to in a manner that maybe, you know, vis-a-vis the other moments in which you're asked or a machine is asked to judge and the criteria of judgment are ambiguous. So when, you know, you run all the algorithms and so on, you run the variations, who has the last word? And, you know, in a design situation with clients in a public, more public environment with the bacteria and how, in that sense, this is a very, very traditional, kind of almost old-fashioned question of aesthetics, question of taste. How and what way does that kind of thinking, the thinking that we associate with judgment rather than reason, or let's say their interaction, work, literally, like when one needs to make a design decision in the way that you're doing this work. Yeah, that's a great way to put it. And I think there are a lot of different moments in which that comes in, both the design input and the judgment. And it's right to ask who is judging and who is designing. So, I mean, I sometimes have to explain to clients that there's nothing inevitable about the geometric system for one that is designed. There's nothing inevitable about the scoring that is also designed. And there's a whole range of things that can't and shouldn't be scored. And that is also kind of decided and designed. Then in choosing between possible designs, you know, more bread or more wine, as you're saying, there's judgment. And I think in better scenarios as opposed to worse scenarios, that can be an inclusive debate and discussion between stakeholders, you know, between the public and the owner and the designer and other stakeholders in the project. But it should be noted, of course, and I think this is partly what you're saying, that even the choice of bread and wine is a judgment and a decision. So, I think it's a framework where as architects, we can do some things and we can help structure the discussions to be more inclusive. But we may also need to educate a little bit people about how the model itself works and about what it means to have a version of automated design and architecture, automated exploration of iterations, that there's very little that is neutral or inevitable about that in itself. And I think there's a range of people who want that to be more neutral and inevitable and they have their own interests for that, for efficiency. There are a lot of people who would like to get a hold of some kind of generative design system so that they can eliminate the architects for one. And so I think it's it, but, sorry. I said, for those of you applying, you will be. But this is a conversation that I believe we should have at this school and in the profession because if we're not having it, then we're gonna be replaced. We're gonna be replaced without our knowing it almost. I mean, so it's, in a way, it's a complex question and discussion, but I think there is room to maneuver a little bit. I'd like to maybe continue this conversation or this topic adding that when you work with biology, often you can't design what you want. And so for example, in the project at storefront, if we had wanted to engineer top down an environmental microbiome that would be optimal for human health, we're not able to make that in a lab. And so the complex microbiomes that we've measured in those different environments, whether it be storefront or the Navy Yard, those are complex populations that we couldn't engineer in a lab. I think you can engineer bacteria to do one thing, but it's hard to engineer an entire population to do many things. And so I think that the particular approach that David has been talking about also comes with the acknowledgement that we can't necessarily create and generate all the things that we want. And so as an interaction between biology and design, what you're maybe doing is coaxing systems in a direction that you want but not necessarily designing them top down. Give a chance to some students. Wonderful lecture, I loved it, first of all. Sorry to monopolize the microphone because this is a question that I could ask you personally on Thursday. But I wanna take the advantage that all of these interesting people are present to pose a question. I wanna continue his conversation from another perspective. His concern is who gets to design. My concern is a little bit more philosophical. In a book called Instrumental Form, Wes Jones made a clear distinction. He said, performance falls short of expression, implying if you strive for an expression, you sacrifice the performance of a building. You're basically going the opposite way. You're saying that performance is the most important part of designing. When you have on one side of the balance, one third of the energy of the world required to make buildings, is expression ethically relevant anymore? This is my question. Second, this is a two-part, and I'm gonna finish real quick. In the AI argument, and all of this philosophy around the AI world, there is a metaphor about a paperclip maximizer, which is a computer that is designed to maximize the production of paperclips, and the conclusion being that the whole, all of the atoms of the world in which the computer was designed were directed towards maximizing the production of paperclips. What I'm suggesting is, if we continue through the route of performance and efficiency, do we sacrifice expression in our profession? Maybe. So performance is a strange word for me. I prefer to avoid it if possible, although I probably used it, but because you could have performance for on any number of qualities. So you could have performance in terms of low weight. So that's an example that we have been talking about. But you can also have performance in terms of social justice, potentially, you know. So I don't think it necessarily has to be like that performance or no performance is the question, but it's kind of what kind of performance. Now, for the expression versus performance, I mean, I think that aesthetics and design are always gonna be part of it and are mixed in there in a very complex way, and you can't really separate one from the other. I think one possibility that you're describing is that if you think of any two things, and you could think of performance and expression, and you could make a trade-off curve out of them to continue the Pareto example, but there are potentially a lot of things and a lot of nuance that's left out of any system like that. So, I mean, the short answer is that, I think to just go back to a previous example you used, I think there is a big role for design in the embodied energy question, and in part because it's not really a question of just using all wood instead of steel. I mean, that might help at first and in some cases, but really I think it's partly a kind of design problem, not necessarily, not only in the sense of form and aesthetics, but in the sense of design as a kind of liberal arts approach, a multidisciplinary, multi-factor endeavor that requires creativity, a balance of science and arts, holistic thinking, different scales, and I think it seems to me that it's exactly that kind of approach that's necessary to deal with something like carbon emissions and climate change and the role of something like embodied energy in that. Acknowledge the amount of risk in your research approach. Have you ever had a research question fail and do you have a specific definition of failure that is specific to your studio and sort of commonly help? I mean, yes, we've had a lot of kind of failed experiments or things that we wanted to work and they didn't quite work, and I guess this is probably a standard approach, but the way we deal with that is just quickly redirect, and maybe that ties back to something I was trying to articulate at the beginning, which is about a kind of biodiversity of a design ecosystem. So I think you get a more robust biological population or a more robust wetland or a more robust and maybe resilient city if you have many different things at play, many different organisms, many different species of bacteria, but maybe also many different ways of mitigating flooding water, and I think it's that kind of biodiverse approach to a design ecosystem that we're trying to cultivate. So in other words, if a computational technique doesn't work, then there's another technique. If a certain algorithm or species or kind of sensor doesn't work, kind of new material doesn't work, then we feel like we can turn to something else. We can use the failed results to try something different, and I think it's that kind of biodiversity of design approach itself, not in a single project, but in the studio as a whole that I think might allow for a more adaptive response to the inevitable failures. Maybe a couple of last questions, maybe just we'll answer them together. Hello, hi, I think the lecture is great. The work is so cool and so new. My question is how did you start? Like, how did you start this? Did you start this interdisciplinary work when you were in school, or do you personally know all of these fields? It's true, it's not like a simple thing. How do you, I guess, start a research topic? Do you want to pass the mic and then we'll just... So my question is completely unrelated. I was actually going to ask about the office space as well. Because of the way that it's designed, it is very highly tailored to the needs and the preferences of the people working there, but I guess this is about residency as well. Because people change or their preferences change, and it is very much a snapshot of what they would like at that point in time, how do you build in the resiliency to an adaptability to different needs over time? Go for it. Those ones. We don't want to add on like three more questions. We'll add the last one. Go ahead. I was kidding, but... Thank you. We'll add four. Thanks a lot for your lecture. I just wanted to know, because you were talking about a firm in which you do research and you do buildings. How do... And you are talking about a small firm. Like how much people do you have in your office and are they the same people that do the research on one project and then on the other project, they work on the building and the construction sites? Or are they two separate teams? I mean, how does this work? Go ahead. One last question. Sorry. Sure. Okay. Well, I just wanted to comment on matching design with living organisms and artificial intelligence. It reminds me of the movie Transcendence. Have you seen it with Johnny Depp? You gotta watch that movie because Johnny Depp, he was like a mad scientist and you kind of gave me those vibes. No, like with all due respect, with all due respect. Because I think it's actually genius because like the root of creativity is taking two different fields that are just like unpractical or just don't even mesh together and taking something from that field to gain. I think in marketing, what people have found out has been successful in marketing, they will look at a different field. So say like Nike, they sell shoes but they'll look at how Michael Jordan jumps or LeBron jumps and they will take that factor and add it to the element to make their marketing successful. And I think that you explained that pretty well when it came to design and using living organisms to create and expand. Because as humans, that's what we are doing. That's what we are creating is we're expanding. We just call it architecture. We're building, you know? So watch Transcendence and for real like getting inspired just don't take over the world. Well, in a way, that connects a little bit, I think to the first question. You know, the idea about looking at different fields, but not only that, but having the almost naive thought that maybe an outsider could come up with a solution to a traditionally insider problem. So I think that's not always gonna work, but I think it's often something that's worth trying. So in other words, you know, there are hundreds of engineers at Airbus who are looking at solving this problem of light weighting. But most of them had never kind of thought about the problem or applied certain techniques in the same way that we came up with. We're no experts, right? And we needed to work with them as the experts in order to verify our kind of radical proposal. But I do have a kind of belief that an outsider approach and a kind of cross idea approach, cross discipline approach can often be valuable, especially if we acknowledge that our world is changing very quickly and that the old approaches and even disciplines themselves may be getting a little outdated for the types of things we have to do inside each discipline alone. Related to the first question about like, I'm not an expert in any of those technologies that we're using, but I think that that kind of crossover, that kind of idea or faith that one can assess a new situation and a problem in a thoughtful, analytical, critical way and contribute to the discussion is something that I have and I believe a lot of people have. My individual personal background is basically in liberal arts education. So I didn't train as an architect as an undergrad, I mean, in other words. And that I think undergrad education and the way that I was able to very easily continue that line of education, that expansiveness at GSAP I think has helped me. And then a question about the people in the office and so we have about 10 people and they're all I think equally interested, analytical, creative, critical and like-minded to me. And we work, everyone basically works on everything. We don't have a lot of specialization. We do have one person in our office who's trained as a computer scientist, but he comes to all the design meetings and participates in all aspects of the process as well. You know, if there's one thing that I'm trying to cultivate a little more of, it's more of a diversity of approaches because while there's something productive in some sense and certainly comfortable in having a kind of common ground of like here are things that we've all thought about, here are things we've discussed before, here are approaches we've used, here are personal interests and there's a kind of nice focus to that. I'm very aware that the ideal situation would be to have more diversity there. Again, returning to my mantra of a biodiversity in the design ecosystem itself. I think there's one question I didn't answer yet. Resilience of space as people move through. Oh yeah, yeah, that's very important. So yeah, in the office space that's highly tailored to one situation, how can that adapt to new situations and if you design just for certain people, like what if they leave, what if they change their minds, what if they change who they're working with? And that's a great question. The short answer is the approach is actually not so bad for that because as the data changes, as people change, as preferences change, as new people come in, the system which we designed and it's not inevitable, but it can process that and it can actually do that pretty quickly. I think the trickier question is where that intersects physical stuff matter and I think this is something that we proposed for the project and ultimately it was not implemented, which was to have flexible space itself, to have sizes of meeting rooms. You talked about you have to have this many meeting rooms, but this company and others are questioning that as well and do we need to have that many meeting rooms and this size of meeting rooms? And so I think the most interesting possibility would be is if the physical stuff itself was more reconfigurable because the system as a whole, if the teams change, we could immediately see if people are still getting the kind of spaces that they prefer or not, including other context, if a new building gets constructed across the street and blocks the light, immediately you can filter that. And in that way I think this process is well suited for changing data, but it needs a corresponding well suitedness for changing physical stuff. Well, David, you've been incredibly generous in all ways tonight, so what's the name of the movie again? Transcendence. Transcendence, watch Transcendence and I hope to see some of you. And most of you six design. I hope to see some of you sometime in the near future. Thank you. Thank you.