Autopoietic Architecture

This is an experimental research project to integrate computational generative architecture and theory hybridized between autopoiesis, technology and cognition in a framework of microbial and AI remedial actions. It considers buildings as intelligent agents. The framework for this dialectic practice and design praxis samples real-world phenomenal nature, living systems, technology, and botany for data appropriate to metabolic buildings. See also: https://uic-es.academia.edu/DennisDollens

Thursday, December 6, 2018

Vasulka’s Tokyo: Media, Machines & Architecture: A Twenty-Year Appreciation

Vasulka’s Tokyo: Media, Machines & Architecture: 
A Twenty-Year Appreciation

Dennis Dollens


Figure 2. Woody Vasulka. Table 1: Translocations. Plotter with 
rails, screens, projectors. 




 First published in the catalog for: 
“Woody Vasulka — The Brotherhood: A Series of Six Interactive Media Constructions. 
NTT InterCommunications Center. Tokyo, Japan 1998. Rewritten and updated 2018..
Key Words: Woody Vasulka. Media Installations. Media Art. Sculpture. Architecture. 
Artificial Intelligence. Algorithmic Art. Object-Oriented Ontology. Architecture Theory. 
Extended Cognition. Epigenetics. Should Trees Have Standing? Environment. 

TABLE: 2. A set of facts or figures systematically 
displayed, especially in columns. [e.g.] “the population 
has grown, as shown in table 1” 3. A flat surface.
Second and third definitions, Google Dictionary

Woody Vasulka’s Bruderbund media structures — The Brotherhood Tables — are hybrid architectures corrupted by subordination. “Corrupted" in Heidegger’s sense that masters are themselves subverted when subordinating others into servitude. In the tables’ case, servitude is to their designer’s intellectual priorities and his imposed functions as spatial, perceptual, and political manipulators. Therefore, in a studio or gallery context, the tables oscillate between machines, sculpture, technology, and architecture — between complex metaphors and simple signs. Just who (the designer?) is mastering what? — Just what (the machines?) is mastering whom? Each table’s unstable position of betweenness in my mind’s-eye is metaphorically symbiotic, making them devices by which we can analyze architecture, biology, and machines as recursively intelligent with us. Through them we can mix programming and infrastructure as in Turing/Wittgenstein (algorithmic/linguistic) propositions1. For example, the tables mediate between machines and bodies, machines and language, machine and pictures, machines and buildings, machines and media, and machines and environment, depending on one’s subjective and physical occupation of space.


Figure 1. Woody Vasulka. Table 6: The Maiden in Construction.

        Military machines organizing geopolitical space do this too: they mediate between humans, electronics, environment, targets, drone-flight paths, and politics. As collections of systems, military equipment is hybrid and prosthetic, yet it is also environmentally embedded with psychological power and bluff — something between political rhetoric, video games, and actual bombing. The tables’ messages thus carry some of this same mechanical/psychological posturing that leads me to think of them as generative intelligences subordinating cultural, political, and feral nature. This is particularly true of Table 6, which Vasulka calls The Maiden (Fig 1), and to a lesser degree of Table 5, The Scribe — where The Maiden is the provocateur and The Scribe our collective memory. 

In most views, including the photographic, the tables reveal industrial machinery, electronic components, pneumatic tubes, sensors, and video projections that occupy interior space as infrastructural, mechanical, and digital innards. Individually, the six tables may be approached and worked with, or they may be observed as a suite of spatial manipulators, sometimes giving viewers a sampling of being inside an evolutionary architecture, a war-room or battle simulation. 

Table 1, Translocations (Fig 2), a matrix bombing plotter from the Vietnam War (recovered from a scrap-yard), is the command unit for dual rails configured as a corridor with video projectors and screens running unpredictably along its rails. The screens and projectors are sensor-manipulated from the plotter and by visitors’ movements along, through, and between the instillation. Yet, the tracking, moving images are only part of Translocation’s power. The structural framework for this 20-foot long, free-standing corridor evokes power, strength, and intelligence colliding with destabilizing video imagery and sound. 


This skeletal corridor, with its shifting physicality and rapid-fire video, not only overpowers interior space, it reorients it. Table 1 charges that space with a mixture of sensory content that includes machine animation, sonic disorientation, and visual confusion, while its three screens and projectors feed real-time and captured images to it and to you. A flux-state of ontological confusion results. To me this combination is taunting. In my experience Table 1 is the closest Vasulka has come to matching virtual space, robotics, and architecture to power and fright. 



Figure 3. Woody Vasulka. Table i: Translocations. rails, screens, & 
projectors during testing.

       In Translocations, the screens are the crucial point of ambiguity fragmenting machine image and viewer perception: they disqualify the structure as machine-in-the-garden (art galleries) and disrupt interior space, never letting Translocations rest fully as media or machine — but always as both. If you immerse yourself in Table 1, (Fig 3) you will realize that its linear and non-linear power are akin to a carnival’s centrifugal spinner: it’s strong and scary and disorienting and seductive.

In a sense, the tables manifest an indeterminate and confusing sight — rhyzomic or hybrid. My argument posits them as bio-machinic works strategically underpinning visual extrapolations, and propositions. Obviously, Vasulka’s theory is written in steel, code, and video. Within this frame, the tables are electronic and engineering agent provocateurs subverting their gallery site (maybe subverting philosophy and criticism too). They will signal corrupt states to you if you let them. Vasulka’s stochastic video programing — imagery projecting severed and serving, feminine/masculine (e.g. The Maiden/The Scribe), hidden and seen, watched and watching, power and control. This video programming celebrates industrial strength, then takes cover behind moving video screens/targets. Furthermore, programming and sensors entrap you as data/image, thereby appropriating you into the culture of surveillance, war, and technology. Unexpectedly, you’re on screen or tripping switches: just being there, you’re implicated and culpable for The Brotherhood diaspora whereafter you join or reject its critique and dialectic. When you enter Vasulka’s work, you don’t know where you are, what you are doing, what is expected of you, or who/what is watching — all prophetic we now recognize of Amazon and Google.

This is architectural power. Think Nazi Germany. Think Nuremberg. (Think Fox News.) Speer’s architecture was the site for the demonstration of power, the spatial event for group-think manifested in rallies in order to manipulate public opinion. For the Third Reich, choreography of image and message was event — ultimately transforming architecture into stage sets for the solidification of control by way of descriptive radio broadcasts, photographs/journalism, and Reifensthal’s cinematic transmission of power. The repeatable choreography of mass media appropriating architecture and staged event was the delivery vehicle responsible for stirring viewer’s allegiance. 

Yet, if Vasulka’s tables critique fascism, their own power firmly lies in the ethical choices they thrust at the attentive individual. They invite users to explore how vulnerable they are to media-machine propaganda controlled by place-based and networked media infrastructure that uses fear tactics from ruthless politicians in collusion with the horrors of war zone images. Today we may think of our own Fake News delivered from cynical newsrooms worldwide. And, if we are attentive, The Brotherhood alerts us in past-is-present scenarios to guard against neofascist rhetoric emerging in current media and politics. For example, Table 6 allows us to analyze The Maiden as anthropomorphic, a nascent feminist in a grotesque posture of seduction recalling the final sentence of “A Cyborg Manifesto’s” “I would rather be a cyborg than a goddess2.” This is never more true than when The Maiden is musically, robotically engaged in performance. Still and all, she is an apparatchik-member of The Brotherhood helping to render the viewer as viewed as well as documented by voyeuristic systems that today could be Facebook, Google, the state, or corporations. 

Vasulka erected sites of generative machine and code cognition/perception that I see as responders to and harbingers of epigenetic change3. In theory, the aphorism, “Machines for living in” (or with), did this too. When Le Corbusier coined that maxim he was serious. He theorized an industrial-based aesthetic that supported propositions for evolving space/form/machine into a mode of social habitation he thought compatible with aspirations for 20th-century life. Le Corbusier’s theoretical vision articulated one of the primary conceptual breakthroughs for Modernism. Vasulka’s work posits a parallel unstated maxim: The Brotherhood tables are machines for thinking. 

Relatedly, I have heard Vasulka use the term epigenetic when referencing art’s role in idea generation and evolution. I’m not off target by speculating that Vasulka’s work evolved similar epigenetic consequences beyond Le Corbusier’s words by investigating electronic control, network surveillance, digital power, and machine learning that edge toward metabolic life situated between hybrid algorithmic AI and organic cognition. 

By forging machines with Heideggerian overtones, Vasulka created cognition-altering partnerships (machine/viewers) imbued with massive industrial strength. His machine/code questions violence and veracity communicated through objects, space, infrastructure, and image in order to investigate phenomena of media, motives, and cognition. The tables are perpetually oscillating — being/Being — no more able to resolve their dilemma than Heidegger was able to his. If Heidegger’s self-acknowledged failure was the inability to resolve being/Being, I don’t think we will find building/machine/mind resolution. Nevertheless that is still our task as heirs of The Brotherhood. We can move toward recuperating and learning the ideas the tables engendered in the Tokyo exhibition. What we initially find is the manipulation of machines and electronics as physical/cognitive instruments and thinking prompts. Such use of tools is unique to human technological progression — stick-to-stone-to-language-to-code-to-computer illustrates how objects have been evolved along pathways as implements in the trajectory of civilization. Those realizations have produced machines for cultivating both thinking and making as acts of nature merging with technology — extended phenotypes.

If you allow that Vasulka intentionally set up the oscillations and that such oscillation leads to sets of binary questions (some already noted), then the machinery/architecture of the tables constitutes a unity. Each table has been designed as an intellectual conceptual device in the union of The Brotherhood. The tables are tools for questioning social cohesion and cognitive ruptures. We arrive at their perimeters to find them environmental agents in formation to practice what cognitive science calls extended cognition4. As such, I’m using them to prompt debate on architecture that encompasses reality/perception, mask/masque, sign/semiotic, self/master, and organism/AI as conceptual sites for an exploration of form and thinking where objects are cognitive environmental participants. 

Specifically, Vasulka has created physical/cognitive infrastructures that are both subject and object — composites and unities generating autopoietic domains5. As infrastructures the tables are machines hacked and recalibrated for explorations communicating intelligence — they are part of our environment, part of our extended cognition, as well as expressions of our social and personal militancy. The tables consequently change the perceptual (epigenetic) capacity of architecture/media by counter-indoctrination, warning viewers of compiled-truths and alt-messages as forms of power — the machines in the building, electronics in the machines, and their ontological being in us, in nature. 

Today, by extrapolation, we can see machines (Fox News) in society and their shadows in the cognitive machinery of our minds. Here is meta-architecture — object-oriented ontology6 where the philosophical unresolvable oscillates between perception, gaming, truth, and manipulation — all exemplified in Hope Hick’s White House phrase, “alternative facts.” Twenty-years after Vasulka’s Tokyo exhibition, his machine megastructures still oscillate for new viewers on YouTube and Google. Those oscillations morph the tables beyond sculptural notability to the position of open source, commonweal-intelligences, nurtured in machinic niches where AI, ALife, and synthetic biology collide with cellular (microbe, plant, animal) perception, cognition, and intelligence.

The tables further suggest ways to question the contradictions of architectural appearance/function in order to consider site, materials, and intelligence as environmentally and phenomenally dispersed among organisms, matter, and force. Woody Vasulka has created machines obviously open to other readings. So, I push my reading toward interpreting ontological and environmental questions embedded in his steel and electronic machines. I read them as coded thinking — evolved, extended phenotypes of Vasulka’s own cognative / perceptual / phenomenal being. The machines are him — self-portraits — Wittgenstein pictures analyzed as extended-human phenotypes, I transfer and appropriate them for residency in my thinking as exploratory devices grappling with questions of power, presence, intelligence, mapping, space, cognition, (epi)genetics, and urban occupation.

I’m valorizing an architecture of AI-robotic, servo-mechanical intelligence as proto-autonomous environmental/cognitive agents. I’m reading the tables because I think they were genuinely an avant guard example of transdisciplinary debate. I see them as the expression off an artist/engineer — a puppet master — working out life’s prickly philosophical questions in the vocabulary of ontological machines even as they echo Le Corbusier’s, “machines for living in.” For those reasons I see the tables as architectural texts open to study by designers concerned with space, control, and AI/ALife. They are new models, not of the old either/or but of the BOTH — biointelligence and machines. In my mind’s-eye, the tables are prototype architectures — Smart City antecedents — merging digital, metabolic, and physical space as it morphs with ways-of-being and way of communicating/negotiating justice in nature7.

Contemplating the tables ultimately leads me back to Heidegger’s corrosive delusion of those who master, those who repress, and those masters corrupted by the subordination of others into a role of servitude. I think the tables, epigenetically recalibrated today, force us to look to abused environments and subsequently to abused humanity not as inevitable conditions and states, but as oscillations between hostile and predatory acts of governments and corporations. We can see predatory acts as war against the environment and be prompted to protect oceans and wild lands and migrations and species justified by the rights of nature7. We can thereby begin to reject the process of the master’s corrosive delusions when we admit that the Trump White House is a seedbed of fascism, corrupted by servitude to self-interest. 

Hereafter, the tables may retrospectively conjure something akin to sets in dystopic films. Think Metropolis. Think 1984. Think Blade Runner. Accordingly, the overriding dialectic here challenges us to configured extended cognition as a system of thinking inclusively on cognition, architecture, technology, war, environment, nature, and perception — ways-of-being in the world — or in Graham Harmon’s sense, “the study of being5” in object-oriented ontology (OOO). Vasulka’s six machines — Translocations, Automata, Friendly Fire, Stealth, Scribe, and The Maiden — deposit us in a frame between space and existence where servitude is our choice, betweenness our state, and evolution subordinated to the consequences of Valsuka’s epigenetic actions translated to nature. In this domaine most of us are replicants politically/environmentally run amok — only a few, such as Woody Vasulka are blade runners.

References

1. Dollens, Dennis. 2018. “Calculating Turing Thinking Wittgenstein: AI, ‘the Case,’ & Metabolic Architectures.” https://www.academia.edu/37266358/Calculating_Turing_Thinking_Wittgenstein_AI_the_Case_and_Metabolic_Architectures

2. Haraway, Donna J. 1991. “A Cyborg Manifesto: Science, Technology, and Socialist-Feminism in the Late Twentieth Century.” Simians, Cyborgs, and Women: The Reinvention of Nature. Routledge.

3. Rosenfield, Henry Israel & Ziff, Edward. “Epigenetics: The Evolution Revolution.” The New York Review of Books. Pp. 36-38. (7 June 2018. LXV;10.)

4. Clark, Andy. 2008. Supersizing the Mind: Embodiment, Action, and Cognitive Extension. New York. Oxford University Press. 

5. Dollens, Dennis. 2018. “Varela's ‘Patterns of Life,’ AI, & Bioremedial Architecture.” https://www.academia.edu/37638563/Varelas_Patterns_of_Life_AI_and_Bioremedial_Architecture
6. Harman, Graham. 2018. Object-Oriented Ontology. A New Theory of Everything. Pelican Books.

7. Stone, Christopher, D. 1972. “Should Trees Have Standing: Law, Morality, and the Environment.” California Law Review


Tuesday, November 20, 2018

AI-to-Microbe Architecture: Simulation, Intelligence, Consciousness

AI-to-Microbe Architecture: Simulation, Intelligence, Consciousness 
Dennis Dollens • exodesic @ mac.com


Abstract:
This paper probes Ludwig Wittgenstein’s impact and Alan Turing’s biodigital plant simulations in light of computational generative architecture and design for bio-intelligent and metabolic architecture today. I’m using scant records and testimony from contemporary and later chronicles, interpreted through firsthand readings of each’s theoretical writings, architecture, and simulations. This text is then a device for considering ways-of-being within, and ways-of-thinking about theory and practice in a framework of philosophy and computation for biodigital design. I propose that fusing biological-to-biosynthetic intelligences (microbes, plants, animals, machines) will allow the theoretical development of bioremedial environmental cleanup addressing climate change. This proposition supports biomimetic design methods to harvest metabolically-driven intelligence and AI in the production of environmentally performative art and architecture. That ontological pathway stems from issues surrounding machine learning and digital simulation for repercussions at object and urban scales necessary to address Smart Cities. Wielding text association and picture identification, recent breakthroughs in heuristic AI and synthetic life stirred me to question past accomplishments in relation to current experimental practices and link future developments in living technologies (ALife) and artificial and biological intelligences as they may be hybridized into performative AI skyscrapers participating in Smart City environmental and communication networks. 
Keywords:Artificial Intelligence (AI), ALife, Art & Architecture, Genetic Architecture, Alan Turing, Ludwig Wittgenstein, Computational Design, Smart Cities, Machine Intelligence, Generative Architecture, Plant Neurobiology, Extended Cognition. 

Propositions Toward Metabolic Architectures
Research strategies to regulate exploratory sets of actions are called upon here for reasoning the inclusion of AI, synthetic life, and bio-algorithmic generation into the production of metabolic architectures. Those exploratory tactics underwrites hypothesizing metabolic architectures as parts of nature. Therefore, to link theory, reason, and observation I evolve stratagems for the investigation of matter, forces, and phenomena starting with symbolic languages to frame architectural intelligence vis-a-vis nature. Specifically, concepts-terms such as “atomic facts,” “form,” “objects,” “substance,” and especially “picture” — anomalously inhabiting Wittgenstein’s Tractatus-world — are appropriated for design research. His word anomalies most convincingly suit design analysis, reinforced when the househe designed for his sister is decrypted to factor in Tractatuslogic examined in “Calculating Turing Thinking Wittgenstein.”2So, for this paper, the Wittgenstein house stands as an intermediary agent between words, realms, pictures, objects, thinking, and design practice to drive philosophical propositions materially realized here to support concepts of metabolic architectures.
Wittgenstein’s philosophy thus serves design-research propositions, especially when updated by extended cognition,3extended phenotypes,4and Turing’s algorithmic botany.5That mix is used to prompt theoretical design emergence involving AI and microbial sensory perception. And that emergence enables conceptualizing metabolic architectures through Wittgenstein’s logic and Turing’s computational simulations as ways to visualize bioAI performance for a building.5As precepts, Wittgenstein/Turing underpinnings justify combining theoretical texts, algorithms, and drawings as operational procedures to explore metabolic architectures influenced by research outside of design realms, such as Life’s Engines.6
The product of propositional thought used to stimulate design visualization enables architects to factor in: (1) the synthesis and mutability of life and matter; (2) differing typologies of intelligence existing in microbes, plants, animals, and (some) machines with (3) simulation potential in computational design. All three areas have parallel interactions in Turing’s7writings and generative biology that impact new lines of propositions, questions, and axioms here grafted onto Wittgenstein’s8ideas as foundations for metabolic architecture. Selectively deployed, this hypothesis uses resulting axioms and questions to reveal first-hand research observations as emergent data. The Turing/Wittgenstein reformulations thereby nurture propositional pathways over which buildings may be considered as intelligent agents. Consequently, users may similarly configure their own research goals to strategize hybridizing organisms with technology.15,16
Together, Wittgenstein’s propositions and Turing’s computational tactics then enable a design research patois to frame bioarchitectural investigation. That research reveals biological attributes of living organisms suited to translation into architectural simulations and models. Viewed through propositional tactics, the Wittgenstein house, designed approximately ten years after the Tractatus,is a manifestation of his philosophical vision resolved in physical form: book-to-building, mind-to-matter. Therefore mind-to-matter scrutiny locates the house in a relationship with nature and biointelligence parallel to that of bees with beehives, wasps with combs, termites with termitaries, and swallows with adobe nests.
Design — formulated through propositions — sanction cognitive states supporting mind-to-matter enactments. It seeks biological performance by questioning nature’s ways to implement sensory biointelligence. For example, asking: How can architects obtain insight into biological intelligence and phenomena for buildings to enact environmental bioremediation made conceivable by human intelligence/creativity? And, if remedial strategies are sourced in nature: How can architects employ theoretical procedures to interpret an organisms’ intelligence and communicate with it for monitoring environmental toxicity? Such questions become recursive — referencing history and philosophy to evolve thinking and data appropriate to biomechanic/biosynthetic hybridization for building components. The architect may then set the building’s objective as receptive to living metabolic operations assisted by AI and synthetic biology.
For metabolic architectures, design development is theoretically guided to include observational and morphological insight derived from nature — cognitively regenerated through algorithms — to foster research between living/synthetic and urban/natural systems. In this step, behaviors of AI/microbe hybrids may be studied as cellular-intelligent agents for partnering buildings to achieve bioremedial behavior. First examples of design for metabolic architectures are organized around theoretical biology via autopoiesis10,11(auto=self; poiesis=making), interpreted to validate plant/microbe organisms as cognitive systems. My attempt to ground theory for such propositions tracks back to Alan Turing’s digital plant observations/simulations11and his theories of machine intelligence,12I filter through autopoiesis updated by extended cognition, AI, and biointelligence.
This cybernetic-like mode of design is linked to biology, with biology linked to code, and code linked to generative architecture looping feedback/feedforth questions born when Turing asked: “Can machines think?”7From Turing’s starting point, designers may fast-forward observations and data to contemplate intelligent systems programmed through computational biology necessary to generate building/machine behavior. In such cognitive/computationally-based processes, theoretical propositions and corollaries emerge as communication tools that push thinking (conceptualization, visualization, design) to situate metabolic buildings as human-extended phenotypes.4,13
Observation of nature’s biochemical and physiological processes illustrate that cellular agents — microbes, plants, and AI — can be integrated into living/synthetic materials provisioned in an architectural core or facade. The resulting architectures operate not only for human occupation but also as hosts for cellular living organisms communicating between urban infrastructures, global networks, and their biological ecosystems. For such hybrids, a designer needs biological data rendered in code or biochemicals to simulate living conditions that, until now, have been genetically programmed only by nature. Such observations and programming presuppose subsystems and sub-subsystems that house collective microbe behaviors designers (working with biologists), must coax into AI partnerships in order to ask: Can buildings biologically remediate pollution? In turn, these post-Turing questions take for granted a machine’s (a building’s) intelligence that, for example, deploy bacteria to feed on carbon (converting toxins to energy) in ways biochemically pioneered by oil-spill cleanup.
The thought behind bioalgorithmic programming and the exchange of metabolic data, is in keeping with the quest for fusing AI, microbes, and biology for intelligent, biochemical tasks. New interactions arise between living, AI, and synthetic life when cellular organisms become capable of genetically enabling environmental responses. Biocellular-AI thereafter resides in a building’s material infrastructure, monitoring, for example, pollution while metabolically reacting to waterborne or airborne toxins atmospherically circulated.
By studying and extrapolating from Turing-inspired biological simulations, I repurpose similar biology-to-code investigations he observed and documented between matter, life, phenomena, and machine processing.5,15 Procedurally, the lineage studies Turing’s observations of plants and radiolarian to unfold new functions for algorithmic simulations. And, in keeping with how Wittgenstein8used propositions to identify the world as the “case,” the Tractatusis culled for systems of logic, cognition, and language. Analyzed and extrapolated data — sometimes pictures in Wittgenstein’s sense — are thus interpreted for simulating responses compatible with AI, biochemical signaling, and biointelligence. Pictured — cognize, as in mind’s-eyevisualized thought — is then accredited as a mode of simulation. My design experiments for tall metabolic buildings11preview this process [Figs. 1. 2], and are propositionally restated and tested in studios I teach in the Biodigital Architecture Master program at the Universitat Internacional de Catalunya.15,16
Designer observations, enhanced with their 3D scanning and SEM imagery are case-relevant to the Tractatus’s first line: “The world is everything that is the case.” Student’s evolve their own “case” when data is retrieved from organisms and applied to buildings in the process of design — inarguably a realm of cognitive nature. In this situation, philosophical procedures supplement material and environmental data to feed design research goals15aimed at building performance. Following preliminary research, biocomputational data and imagery from microscopes and scanning instruments16nurture code transporting insights from AI/microbe symbionts to computational simuations. Wittgenstein/Turing symbiotic pairing — as outlined — substantiates biocomputational simulation with Python or Grasshopper prior to experimental prototyping and consequent fablab models.15,16 
To unpack the preceding two paragraphs requires us to: (1) theorize technology, ecology, architecture, and computation in terms of bio/synthetic propositions. Doing so positions us to hybridized AI with cellular intelligences targeted to generative architectures and, (2) tasks resulting theory to support programming pathways that address climate, soil, and water restoration in the context of a building’s ecology and its own self-maintenance. Reasoning logical propositions for programming then connect those two pathways into a network of living and synthetic potential for, (3) envisioning and embedding buildings with autonomous sensing, thinking, and reacting intelligences.
The above bio-to-building debate15,16characterizes emergent buildings not only as intelligent, but as potentially autonomous. Transforming airborne carbons for instance, into energy, is an example whereby actions of metabolic architectures requires intelligent, organismlike analysis of toxins in order to orchestrate environmental response. Indexed and routed to architectural functions, algorithmic simulations, and cellular biochemical reactions frame how conceptualization of metabolic buildings gives rise to new questions concerning design/technology in nature. For instance, queries such as: If a metabolically enhanced, intelligent building subjectively experiences nature by drawing upon sensory plant or microbial life, does it enter phenomenal realms of plant or AI consciousness / biointelligence?
From such interrogations — not frequently probed, yet lurking behind, for example, Google’s AlphaGo,17— I anticipate ontological nature-to-machine unity. Such questions are credible after Google’s AI succeeded at learning, playing, and winning Atari video games, beating world champions at the Game of Go, and triumphing over human and machine chess players.18, 19, 20Still questions remain: If programmers cannot fully understand how AI learns, are some species of AI existentially, phenomenally, and/or experientially thinking?20Associatively: Is that degree of AI processing sufficient to suspect machines are thinking? That a subset of specialized machines are approaching cognitive abilities and conscious subjectivity?22, 23,24— qualities humans have considered exclusive to themselves and a few other animal species.
This is not to suggest one-to-one microbe or plantlike parity between biological or even heuristic AI.23Rather, I point out repercussions of behavior rooted in computationally originated phenomena if paired with, or genetically edited into, biological cells reproducing synthetic-life expressed in next-generation cellular inheritance — e.g. Venterlike biosynthetic-cellular reproduction. Metabolic buildings, set in motion with hybrid intelligences, could be conceived as environmentally proactive biosynthetic agents. Thereafter, unforeseen propositions resulting from genetic-reproductive systems would jolt architectural possibility. Through machine consciousness,24ALife25,26or CRISPR-edited DNA — biological reproduction8of synthetic cells begins to blur distinctions between life, AI, and machines23long after Turing. Ironically, the above processes are happening in parallel with our recognization of plant, tree, and soil intelligence, communication, and cooperation.31,32,33
To be clear, I’m saying that Turing/Wittgenstein theoretical propositions are themselves members of intelligent nature, they are phenomenal agents-of-thought licensed in philosophy and theories of cognitive science.3Such propositions are tools/agents of thinking — facilitators and outgrowths from human cognition as it evolves new categories (species) of intelligence3— in this paper’s “case”8as metabolic architectures.11Those agents-of-thought, manifested in the Tractatus, are read here as design propositions, axioms, and demonstrations that look to me like extended (idea-text-drawing) phenotypes.4, 13From that viewport, idea-text-drawing-propositions are extendable3to buildings resulting in built forms and AI devoting life-support systems to metabolic architectural agency. 
However jarringly, constructed species of AI and ALife9, 25, 26extend and complicate Turing’s question, “Can machines think?”7, 12, 28His question (and Wittgenstein’s too27), if answered positively, gives credence to the proposition: Buildings, as big machines, can think. Thinking symbolically, communicated in philosophy by Wittgenstein and in computation by Turing, allows us to conceive symbolic, linguistic, mathematic, biochemical, and biocomputational goals for metabolic architectures and theory. Design research goals can then be perused as ontological searching for ways-of-being / modes of debate / types-of-intelligence / circumstances-of-design / and responses-to-climate change. Ontological searching, as an aspect of design research, afterwards, contextualizes and justifies the philosophical and theoretical extension of cognitive functions to buildings, and consequently, metabolic buildings as agents of extended cognition.3
Metabolic architectures are thereby first ideas, then propositions and sketches (or codes) developing cognitive exchange between design, phenomena, material, and nature. In this lineage, propositional analysis and communication are realizedas descendent not only from the Tractatus,8but from Latin res(thing) as in res extensa; and idea as in ideinfrom Greek, to see. The paradox — things seen cognitively — or cognitive things (res cogitans) — underpins design (ideas/propositions) translated to, or simulated in matter, as Wittgenstein’s “case” defining thinking domains for performative ecological architectures. 
In theory, I situate the “case” in service to biodigital simulation begun by Turing’s translation of plant attributes to algorithms.12, 28Simulation then enters the framework not only through Turing, but through Wittgenstein’s term, to “picture” (residein). Similarly, propositions themselves can be a category of simulation defined from their Latin simulationem orsimulareto underpin practices of thinking, designing, imitating, matching, or mimicking — simulating — as crafted or constructed generative thought. 
Turing’s algorithms simulated parts of nature via computation and are now examples appropriate to design-theoretical thinking and its role in simulating acts of nature as architecturally performative. In this scenario, we confront results of the verb simulation and the noun simulacra to discern the imitation/intimation of life/cognition. Thus, simulations for this usage are human-extended phenotypes.13They are objects of thought/simulation outside of Baudaraud’s29virtual phenomena existing as cognitive/computed numbers transportable into symbolic, thinking objects potentially animate. In autopoietic10terms, they are symbols —autopoietic unities or domains — that include Wittgenstein’s picturing8to underpin language and thought constructions realizable in phenomenal and material worlds.
To give bearing to this paper, I locate Turing as the agent from whom we learned the primary stages of algorithmic nature and phenomenal simulation as observational components for programming and later, for machine output — seeing (res) and imagining (ideate)for computationally simulating nature.In the process, we evolve architectures after examples Turing established for machines by extending his processes from computational biology11to realms where machines and architectures are extended phenotypes, growing autonomously intelligent. By subscribing to precepts of Dawkins’s4and Turner’s13extended phenotypes as pathways for intermingling observations of nature3(ideas, matter, and things cognitively active), leads to metabolic architectures as descendants of Turing Machines.30The design territory then encompasses observation-to-mathematical simulations,15cultivated in a Tractatus-influenced dialectic demonstrated as computable by Turing’s experiments with botanic simulation11 his theory of reaction/diffusion,12and interest in phyllotaxis.28
Natural functions, for instance phyllotaxis in fircones or daisies, as observed by Turing in his drawings, algorithms, and programming5,28have been culturally and technologically assimilated. However, they may be further recast — computationally reprogrammed and resimulated, biofabricated for morphological support systems not possible in Turing’s time. In that frame, a further paradox is apparent whereby we are simulating nature as an act and continuum of intelligent/phenomenal nature itself. And, at junctures of observation, simulation, and performance other questions arise: If AI can simulate life, can it think it is alive or conscious?24Or: Can bioAI convince single cell and multicellular life that it is part of their metabolic systems? Such questions are not so projectively strange after human systems have been surgically and pharmacologically adapted to pacemakers, cochlear implants, and transplanted organs. Considering those accomplishments, resultant metabolic architectural species may be contemplated as transmitting coded/decoded instructions for the operation of environmentally mediating buildings that enter zones of autonomous, biodigital agency.15,16

Conclusion
New iterations of AI — biological, synthetic, and hybrid — are being developed in laboratories around the world. Plant intelligence predicated on sensing, signaling, and biochemically reacting are equally subjects of international research.22, 31Theory for metabolic buildings should recognize that collective bio/AI intelligences will impact ways-of-being for performative architectures in smart cities. We should therefore interrogate bio/AI systems and methodology not as questions of sustainability, but in the context of extended cognition,3metabolic buildings, and hybrid AI/organisms aimed at biointlligently eradicating pollution.
From theories of extended cognition, extended phenotypes, and autopoiesis4,10,13protocols for analysis may be transferred to studio practices15,16involving multidisciplinary design fields. Designers may apply technologically enabled observation in order to construct research agendas that introduce bioAI intelligence into architectural visualization and fabrication. Simultaneously, designers may investigate AI and/or living intelligence that address questions of how big machines — buildings — can function like organisms. For example, one direction for application will appear when AI manages and anticipates life-support for microbial populations that metabolize — “eat” — pollution and thus contribute to systemic homeostasis. In this sense, AI and cellular life become endosymbiotic — microbial-AI — cells producing energy by biochemically digesting pollution, for example, as robotically discussed by Jonathan Rossiter,14but applied to metabolic architectures addressing toxic environments.

References
1. Leitner, Bernhard. 1976.The Architecture of Ludwig Wittgenstein: A Documentation. New York. New York University Press.

2. Dollens, Dennis. 2019. Forthcoming” “Calculating Turing Thinking Wittgenstein: AI, ‘the Case,’ & Metabolic Architectures.” Virtual, Informal, and Built Landscapes. Bogota. Pontifica University.

3. Clark, Andy. 2008. Supersizing the Mind: Embodiment, Action, and Cognitive Extension. New York. Oxford University Press.

4. Dawkins, Richard. 1982. Extended Phenotypes: The Long Reach of the Gene. New York. Oxford University Press.

5. Dollens, Dennis. 2014 “Alan Turing’s Drawings, Autopoiesis and Can Buildings Think.” Leonardo: The International Society for the Arts, Sciences and Technology. Cambridge, MA. The MIT Press. 47:3. 249-253.

6. Falkowski, Paul G. 2015.Life’s Engines: How Microbes Made Earth Habitable. Princeton. Princeton University Press.

7. Turing, Alan M. 1951. “Can Digital Computers Think?” TS with AMT annotations of a talk broadcast on BBC Third Programme. 15 May 1951. In: Cooper, S. Barry & van Leeuwen, Jan. Eds. (2013) Alan Turing: His Work and Impact. Amsterdam. Elsevier. 660.

8. Wittgenstein, Ludwig. 1922. Ogden, C.K. Tractatus Logico-Philosophicus. London. Kegan Paul.

9. Church, George and Regis, Ed. 2012. Regenesis: How Synthetic Biology Will Reinvent Nature and Ourselves. New York. Basic Books. 

10. Maturana, Humberto & Varela, Francisco. 1980. Autopoiesis and Cognition: The Realization of the Living. Dordrecht, Holland. D. Reidel Publishing Company.  

11. Dollens, Dennis. 2017. Metabolic Architectures: Turing, Sullivan, Autopoiesis, & AI. Barcelona. Universitat Internacional de Catalunya. 

12. Turing, Alan M. 1952. “The Chemical Basis of Morphogenesis.” Philosophical Transactions of the Royal Society B. 237. 37-72. Reprinted in: Saunders, P.T. (1992) Collected Works of A.M. Turing: Morphogenesis. London. North-Holland.

13. Turner, Scott J. 2000. The Extended Organism: The Physiology of Animal-Built Structures. Cambridge, MA. Harvard University Press.

14. Rossiter, Jonathan. 2016. “A Robot That Eats Pollution.” TED Talks. March 2016. http:// www.ted.com/talks/jonathan_rossiter_a_robot_that_eats_pollution?utm_source=newsletter_ weekly_2017-02-25&utm_campaign=newsletter_weekly&utm_medium=email&utm_ content=talk_of_the_week_button 

15. Dollens, Dennis. 2018. “Studio: Digital Morphology for Metabolic Architectures.” Prepublication. https://www.academia.edu/36491725/Studio_Digital_Morphology_for_Metabolic_Architectures
16. Dennis L Dollens, Alberto T. Estévez. (2017) "Metabolic Architecture: Genetics, Digital Morphology, Turing, & AI". In: VV.AA. Conference Proceedings, Kine[SiS]tem'17 - From Nature to Architectural Matter International Conference. ISCTE - IUL, Lisbon (Portugal), June 19th-20th, 2017. P. 248-256.
17. Burton-Hill, Clemency. 2016. “The Superhero of Artificial Intelligence: Can This Genius Keep it in Check?” London, UK. The Guardian. 16 February 2016. https://www.theguardian.com/technology/2016/feb/16/demis-hassabis-artificial-intelligence-deepmind-alphago
18. Sample, Ian. 2017. It's able to create knowledge itself': Google unveils AI that learns on its own. The Guardian. 18 October 2017.  https://www.theguardian.com/science/2017/oct/18/its-able-to-create-knowledge-itself-google-unveils-ai-learns-all-on-its-own

19. Sample, Ian. 2017. Google’s DeepMind Makes AI Program That Can Learn Like a Human. The Guardian. 14 March 2017. https://www.theguardian.com/global/2017/mar/14/googles-deepmind-makes-ai-program-that-can-learn-like-a-human

20Agence France-Presse, Shanghai. 2017. “World's best Go player flummoxed by Google’s ‘godlike’ AlphaGo AI.” The Guardian. 23 May 2017.https://www.theguardian.com/technology/2017/may/23/alphago-google-ai-beats-ke-jie-china-go

21. Kuang, Ciff. 2017. “Can A.I. be Taught to Explain Itself?” New York. The New York Times Sunday Magazine. 26 November 2017.

22. Dollens, Dennis. 2015. “Autopoiesis + Extended Cognition + Nature = Can Buildings Think?” Communicative and Integrative Biology. 31 August 2015. www.ncbi.nlm.nih.gov/pmc/articles/PMC4594259/

23. Sample, Ian. 2017 “Organisms Created with Synthetic DNA Pave Way For Entirely New Life Forms.” The Guardian. 24 January 2017. https://www.theguardian.com/science/2017/jan/23/organisms-created-with-synthetic-dna-pave-way-for-new-entirely-new-life-forms

24. Dehaene, Stanislas. Lau, Hakwan. Kouider, Sid. 2018. “What is consciousness, and could machines have it?” Science. 358. 486–492. 27 October 2017. http://science.sciencemag.org/

25. Bedau, Mark A. McCaskill, John S. Packard, Norman H. & Rasmussen, Steen. 2013. “Introduction to Recent Developments in Technology.” Artificial Life. Cambridge, MA. The MIT Press. 19. 291-298. Accessed: October 2013. 

26. Langton, Christopher G. (1988/1989) “Artificial Life.” In: Langton, Christopher G. 1989. Artificial Life. Santa Fe, NM. Addison Wesley & The Santa Fe Institute. 6:1-47. 

27. Wittgenstein, Ludwig. 1933/1965. The Blue and Brown Books. New York. Harper Perennial.

28. Turing, Alan, M. 1953. "Morphogen Theory of Phyllotaxis." In: Saunders, P.T. Ed. Collected Works of A.M. Turing: Morphogenesis. London. North-Holland. 
29. Baudrillard, Jean. Faria Glaser, Sheila. TR. 1981/TR: 1994. Simulacra and Simulation. The University of Michigan Press.

30. Turing, Alan M. 1936. “On Computable Numbers, With an Application to Entscheidungsproblem.” In: Cooper, S. Barry & van Leeuwen, Jan. Eds. (2013) Alan Turing: His Work and Impact. Amsterdam. Elsevier. 16

31. Pollan, Michael. 2013. “The Intelligent Plant: Scientists Debate a New Way of Understanding Flora.” The New Yorker. 23 & 30 December 2013. 92-105. 

32. Lettvin, J.Y., Maturana, H.R., McCulloch, W.S., Pitts, W.H. (1959) “What the Frog’s Eye Tells the Frog’s Brain.” Proceedings of the Institute for Radio Engineers. 47:11. November 1959: 1940-59. 


33. Stone, Christopher, D. (1972. 2000) Should Trees Have Standing: Law, Morality, and the Environment. Oxford University Press. 

Monday, November 12, 2018

Emergent Futures: Non-Deterministic Universal Turing Machines Using DNA — An Assessment for Metabolic Architectures



Figure 1. Dennis Dollens. Metabolic Ferox bioTower. 2018-ongoing. Theoretical hybrid between AI and microbial intelligent agents to 
sense and respond to transform atmospheric CO2 via computationally enhanced photosynthesis. 
L-system grown eTree tower. 3D, CT scanned Datura ferox pods hosting microbial agents.


Emergent Futures: Non-Deterministic Universal Turing Machines Using DNA — 
An Assessment for Metabolic Architectures 
Dennis Dollens. Universitat Internacional de Catalunya, Barcelona
An interdisciplinary response to: “Computing Exponentially Faster: 
Implementing a Non-Deterministic Universal Turing Machine Using DNA.” 
Interface: The Royal Society. London. 14.

The title, “Computing Exponentially Faster: Implementing a Non-Deterministic Universal Turing Machine Using DNA”1 alerts the existence of a mathematical/scientific barrier the AI-uninitiated reader may think to avoid. However, it rewards reader perseverance by opening a view into biological computing based in living organisms at molecular levels of DNA. To read the text is then a wakeup call for design professionals (and others) theorizing molecular bioremediation, artificial intelligence (AI), and artificial life (AL) in a lineage descending from generative computation and species’ intelligence. The first three sentences from “Computing Exponentially Faster’s” Abstract references the decades after Alan Turing’s 1936 “On Computable Numbers:”2

The theory of computer science is based around universal Turing machines (UTMs): abstract machines able to execute all possible algorithms. Modern digital computers are physical embodiments of classical UTMs. For the most important class of problems in computer science, non-deterministic polynomial complete problems, non-deterministic UTMs (NUTMs) are theoretically exponentially faster than both classical UTMs and quantum mechanical UTMs (QUTMs).1

Leaving aside a definition of non-deterministic polynomials for a moment, if we pair the lines above with propositions from the “Introduction,” we discover a scaffold supporting ways to conceive, visualize, and (hypothetically) interface biological processes with Turing machines. Doing so, the propositions open potentials to harness cellular organisms as computing partners, situating them as applicable to theoretical development in various disciplines. In the context of environmental bioremediation and metabolic architectures we then encounter parameters envisioned by King et al. where:

Universal Turing machines (UTMs) [are] the theoretical foundation of computer science [where] the Church–Turing thesis states that UTMs exactly define the concept of an algorithm-effective calculability. UTMs also play a fundamental role in science: the Church–Turing principle states that they are sufficient to simulate perfectly all physically realizable systems.1
The reason a designer without knowledge of Turing machines, AI, neural networks, or microbial/plant3 intelligence should continue reading is embedded in the subject of DNA’s capacity for driving a first revolution in metabolic computing. For designers contemplating the incorporation of microbe/plant abilities in buildings, “Computing Exponentially Faster” elaborates propositions that will alter the dynamics of research investigating urban, material, and environmental design processes capable of manifesting a credible design-by-research pathway. With a research pathway,4 architectural experimentation may follow toward pairing biocomputational (e.g. NUTMs) and biochemical (e.g. microbes) for the eradication of environmental toxins. 
Equally important, these computational strides will come partnered with, or embedded in, biological functions the paper (and architecture) anticipate, utilizing physical space and matter (molecules). Molecular components and genetic editing have essential roles creating spatial conditions in which DNA NUTMs coexists in living and synthetic cells5 for new material performances. Contemplated for intelligent, metabolic buildings, DNA machines would maintain life support and communication systems for microbes enabling crosstalk between organisms, substrates, soils, and NUTMs. Thereafter organisms, substrates, structures, and code host programmed molecules — DNA unities in autopoietic6 terms — capable of integrating buildings/smart city infrastructures with biological life and AI/AL infrastructures (Fig 1). Those unities are thus programmed to communicate between living cellular and computational systems. One potential route for design theory and experimentation is to accept projections in “Computing Exponentially Faster” in order to equate NUTMs as components of metabolic architectures. Computation materially embedded and systemically distributed essentially turns buildings into very big performative NUTMs in the form of — say skyscrapers or even whole neighborhoods — whose bioAI infrastructure consequently preforms with:
. . . advantages in speed, energy efficiency, and information storage over electronics: the number of operations for a desktop DNA computer could plausibly be . . . (approx.103 times faster than the fastest current supercomputer); it could execute approximately 2 X 1019 operations per joule (approx. 109 more than current supercomputers); and utilize memory with an information density 1 bit per nm3  (approx. 108 more dense than current memory). . . a DNA NUTM based computer could potentially utilize more processors that all the electronic computers in the world combined, and so outperform all standard computers.1
This is a distant vision, and while “Computing Exponentially Faster” is not directed to architecture or design, nor do the authors speculate on NUTM implementation in practical applications, the paper’s impact on design (and other disciplines) is preparatory, signaling a scenario by which effective climate monitoring, simulation, materialization, and biochemical responses may arrive for environmentally proactive architecture. DNA computation will prompt new thinking and preparation for experiments involving nature, biology, synthetic biology, artificial intelligence (AI), and artificial life (AL) — to enhance performance abilities of buildings and cities targeted to negative carbon dioxide (-CO2) emissions. 

If King et al’s., revolutionary proposal enters realms of design debate, analysis of their Non-Deterministic Universal Turing machine’s (NUTM’s) and their consequent DNA programming would impact theoretical solutions from other sciences, technologies, and design with environmental monitoring and remedial actions. One recent article in Nature spoke of Tobias Erb’s research searching for ways to build synthetic cells “that allow photosynthetic microbes to pull carbon dioxide from the environment and make sugars and other cellular building blocks.”5 p174 Micromanaging DNA NUTMs could situate synthetic cells for exponentially greater efficiency in hosting of those photosynthetic microbes, creating an autopoietic circuit or recursive homeostasis suitable for urban duty.

In retrospect we can recall that it took decades for the implications of Turing’s 1936 UTMs2 to to gain widespread credibility. In fact, that paper is still insightful as are his 1950s “morphogen” and reaction/diffusion (biological-to-computation) algorithmic simulations.7 Those plant/radiolarian simulations spearheaded plant/animal functions algorithmically coded for discrete segments of code.8,9,10 

Likewise, it took decades for AI and synthetic biology to arrive at current states ripe with potential for intelligent and autonomously intelligent metabolic buildings. So, while it may be decades before DNA NUTMs enter the desktop realm King et al. work toward, there is justification for theorizing bioarchitectural intelligence that take advantage of resulting computational magnitude — potentially shifting ecological policy, urban planning, smart city design, and social strategies to coexist within the rights of nature.11 For buildings, differences between UTMs and NUTMs are thus germane for contemplating molecular and/or microbial infrastructures as participants in hybrid bioAI and living technology scrubbing CO2 from the atmosphere.3,5 

“Computing Exponentially Faster” posits: “Digital electronic computers physically embody UTMs, but differ from them in that they have bounded memory, may only run for a bounded amount of time, [and] make errors.” In relation to NUTMs, Universal Turing Machines (UTMs) take “an input state to an output state, with computing halting if an accepting state is reached.” A Non-Deterministic Universal Turing Machine differs, in “that from one input state there may be multiple output states.” However, King et al. write that a “NUTM is a UTM that can reproduce itself” thereby establishing parallel paths “with the computation ending when one path reaches the [correct] accepting state.” They further stress, “the resource limitation in a physical NUTM is space. The speed of an NUTM’s computation increases exponentially while the amount of space available is polynomially bound [restricted by multiple performative mathematical functions and limits of matter]. . . . Computation in a physical NUTM therefore resembles an explosion.” A resulting consequence is the theory’s ability to conduct DNA computing coded as a Thue NUTM. The tradeoff between UTMs space, and NUTMs time, suggests that when:

. . . trading space for time it makes sense to use as small processors as possible: molecules. . . . It is therefore rational to expect that a molecular NUTM, through trading space for time, could outperform the world’s current fastest supercomputer, while consuming a tiny fraction of its energy.

To program NUTMs, King et al. use Thue rewriting systems they reference to Lindenmayer’s biological coding language, L-systems.12,13 Thue systems have the equivalent power of Truing machines running L-systems, but are expressible in multiple strings where, “the main differences are that in an L-system there are no symbol deletions, and multiple rules are applied simultaneously.” Here, for their NUTM: 

. . . the application of Thue rules is naturally non-deterministic: multiple Thue rules may be applied to a string, and individual Thue rules may be applied to multiple positions in a string. . . . These two features enable the practical exponential increase [over L-systems] in our NUTM design.1

Those features result in the possibility “to translate any Turing machine into a Thue system, and vice versa.” Such an advantage is necessary for working with multiple-strings of computation simultaneously, where the authors claim that, “significantly our work is an advance on all previous other work in that we present the first NUTM design.” The authors further emphasize infrastructural similarities (compatibilities) with living systems:

The mechanism of the NUTM depends on the specificity of molecular binding — as do living organisms. . . . To write programs (initial states), we use DNA synthesis technology. Accepting states are specific sequences of DNA that contain identifying certificates, and corresponding answers to the computation. We require that the accepting states be recognized from among a potential exponential number of other states generated by the NUTM. . . . [In this situation] It is helpful to divide the task of applying a single Thue rule / direction into two steps: recognition and rewriting. . . . [Identifying that] There are three types of Thue rewriting edits: transposition, insertions, and deletions.1

Consequently, “the rules required for an NUTM can be physically implemented using DNA mutagenesis. . . .” demonstrate that “the design works using both computational modeling, and in vitro molecular biology experimentation.” Still, the authors are cautionary that, among other requirements, a “NUTM programming language that compiles down to Thue systems” is still missing. Conceivably, this may be less a problem as labs worldwide search for a DNA editing language widens and synthetic biology evolves cellular operating systems.5

Explaining Thue systems by comparison with L-systems,12,13 the authors established a further link to Turing’s mathematics (reaction/diffusion) that underpin Lindenmayer’s programming language for simulating biological organisms.4 p48 Still, as far as I know, no versions of L-systems implement individual string or branch instructions. In my experience implementing single “transpositions, insertions, and [or] deletions” at specific branches required multiple L-systems models/modules running simultaneously in locked X, Y, Z axes. Running simultaneously, conditions not applied to all branches are possible, as are turning off functions of say, a tree trunk, in order to express structural self-supporting interesting branches (necessary to a physical system simulated from tree phyllotaxy generated as a structural truss without kingpins and, in the case of a tree, a repressed trunk4). 

The motivation for “Computing Exponentially Faster,” “is to engineer a general purpose way of controlling cells. The natural way cells are controlled is a very complex combination of DNA, RNA, protein, and small-molecule interactions (supplemented by epigenetics. . . .” The authors maintain that the above DNA NUTM would:

. . . in principle enable arbitrary biological processes to be programmed and executed. The NUTM could receive biological signals from the environment through interactions [“morphogens” in Turing’s words] with transcriptions factors, etc. . . . [Here then] it would seem possible to implement the core ideas in a biological substrate. 

Homeostasis in biological substrates, hybridized into a building’s facade and/or its materialization, then becomes an experimental challenge — the entire building becomes a computing, environmentally-remediating biological substrate incorporating DNA NUTMs. Conceivably, substrates (hybridized biofilms?) then located in building materials act as architectural organisms.
If multicellular life forms, objects, and environments are recognized as components of cognitive science’s extended mind hypothesis14 — our acknowledging a wider scope of environmental and object intelligence “receiv[ing] biological signals from the environment,” enables us to contemplate “biological processes to be programmed and executed.” The environmental potential of scaled-up signaling inherent in the use of DNA NUTMs then impacts single-to-multicellular organisms, communicating between neurological and non-neurological intelligences. 
Such intelligences are native across bacteria, plants, and animals and would enable DNA NUTMs to serve as biological pathways Truing’s 1950s research initiated (but did not foresee) in plant/animal observations he translated to algorithmic strings of code and drawings.7 Turing’s expression of nature, cognition (observation), botanic-to-algorithmic translation, programming, and subsequent drawing output, emphasizes an extended role for computation in design performance as a significant trajectory for future DNA NUTMs participating in environmental bioremediation.
References

1. King, Ross D. Currin, Andrew. Korovin, Konstantin. Ababi, Maria. Roper, Katherine. Kell, Douglas B. & Day, Philip J. 2018. “Computing Exponentially Faster: Implementing a Non-Deterministic Universal Turing Machine Using DNA.” Interface: The Royal Society. London. 14. 

2. Turing, Alan M. 1936. “On Computable Numbers, With an Application to Entscheidungsproblem.” In: Cooper, S. Barry & van Leeuwen, Jan. Eds. (2013) Alan Turing: His Work and Impact. Amsterdam. Elsevier.

3. Dollens, Dennis. 2015. “Autopoiesis + Extended Cognition + Nature = Can Buildings Think?” Communicative and Integrative Biology. 31 August 2015. www.ncbi.nlm.nih.gov/pmc/articles/PMC4594259/

4. Dollens, Dennis. 2017. Metabolic Architectures: Turing, Sullivan, Autopoiesis, & AI. Barcelona. Universitat Internacional de Catalunya. 

5. Powell, Kendall. 2018. “Biology from Scratch: Built from the Bottom Up, Synthetic Cells Could Reveal the Boundaries of Life.” Nature. 8 November 2018. 563.

6. Maturana, Humberto & Varela, Francisco. 1980. Autopoiesis and Cognition: The Realization of the Living. Dordrecht, Holland. D. Reidel Publishing Company.

7. Dollens, Dennis. 2014 “Alan Turing’s Drawings, Autopoiesis and Can Buildings Think.” Leonardo: The International Society for the Arts, Sciences and Technology. Cambridge, MA. The MIT Press. 47:3. 249-253.

8. Turing, Alan M. 1951. “Can Digital Computers Think?” TS with AMT annotations of a talk broadcast on BBC Third Programme. 15 May 1951. In: Cooper, S. Barry & van Leeuwen, Jan. Eds. (2013) Alan Turing: His Work and Impact. Amsterdam. Elsevier. 660.

9. Turing, Alan M. 1952. “The Chemical Basis of Morphogenesis.” Philosophical Transactions of the Royal Society B. 237. 37-72. Reprinted in: Saunders, P.T. (1992) Collected Works of A.M. Turing: Morphogenesis. London. North-Holland.

10. Turing, Alan, M. 1953. "Morphogen Theory of Phyllotaxis." In: Saunders, P.T. Ed. Collected Works of A.M. Turing: Morphogenesis. London. North-Holland. 
11. Stone, Christopher, D. (1972. 2000) Should Trees Have Standing: Law, Morality, and the Environment. Oxford University Press.

12. Lindenmayer. A. 1968. “Mathematical Models for Cellular Interaction in Development.” Journal of Theoretical Biology. 30:300-315.

13. Lindenmayer. A. & Prusinkiewicz, Przemyslaw. 1988. “Developmental Models of Multicellular Organisms: A Computer Graphics Perspective.” In: Langton, Christopher G. 1988. Artificial Life. Santa Fe, NM. Addison Wesley & The Santa Fe Institute. 6: 221-249.


14. Clark, Andy. 2008. Supersizing the Mind: Embodiment, Action, and Cognitive Extension. New York. Oxford University Press.