What could be a way to identify the levels at which the digital is affecting our surrounding? What could be appropriate approaches to designing it? In order to find out, we will examine four core concepts of algorithmic, computational and massively distributed phenomena. In the text that follows, we will explore the claims from contemporary theorists about material consequences of computation. The four core concepts are the following: “active form” (Keller Easterling, “Extrastatecraft”, 2014), “soft thought” (Luciana Parisi, “Contagious Architecture”, 2013) “layer” (Benjamin Bratton, “The Stack”, 2016) and “metaphore”.

Active Form

In much of her work, Keller Easterling advocates the design of active form. Most notably, in Extrastatecraft (Easterling, 2014), she describes how active form is already performing important changes on the urban form. Active form generates repeatable objects anywhere in the world.

Easterling weighs active against the object form, which is what architects are attuned to making. She turns our attention to the importance of infrastructure space performance – laws, regulations, optimisations in the production of space – and the power that resides in working with or on active form. This is urban software that is performing on urban space, not on computer hardware.

The algorithms Easterling discusses are inscribed in both zoning rules and regulations, state legislation of free trade and master plans for a ‘city in a box’, 3D modelling software rendering these master plans, the aesthetic of architectural designs. The algorithmic performance of space is not about making an individual house, but “a protocol or a non-digital spatial software that is both shaping and generating the activity of making houses”  (Easterling, 2012a loc. 150). Spatial products are the outcomes of algorithmic performances on space.

Active form is an entity that determines and produces the physical bits of our cities. It performs in space and It is driven by higher order algorithms, such as incentivised urbanism.  The importance of designing active form lays in the premise that the majority of buildings today are not designed in a careful architects design process, but replicated mediocre projects that proved sufficiently successful elsewhere.

For Easterling, looking at active form is a way to “untune” our eyes from buildings and discuss performativity of infrastructures from a political and spatial perspective, offering a unique understanding of complexities involved in the relationship between the urban fabric and telecommunications.


How might we personify active form? Easterling gives us some hints by talking about what active form can do. It can create object form: “Active form may partner with and propel object form, determining how it will align with the power to travel through culture, redoubling design capacities with additional modes of authorship, time-released powers and cascading effects.” (Easterling, 2012a loc. 293).

By which means does active form act on space? Active form, thus, determines object form, but it is more generic than that: “active forms […] may move beyond the conventional architectural site.” (Easterling, 2011). It is object software itself: “active forms, like softwares, are a form for handling forms.” (Easterling, 2012b, p. 62). Architects could, Easterling believes, master this way of looking at architecture, and dedicate themselves to designing active form actively – rather than leaving it to invisible processes that come out of interactions between legislation and economic incentives.

How does it propagate? By legislation and by means of construction rationalization. Active form determines object form in such a way that it makes constructing architecture more easily more profitable. The combination of incentives given to companies in form of lower taxation, and the availability of the “city in a box”, generic city projects that are easily applicable to any location in the world, enable the acting of active form on a global scale.

How totalzing is it? Easterling describes notable buildings as “stone in the water”. The rest, is the water. And it is increasingly shaped by active form. It is thus on its way to becoming the principal source of aesthetic and structural decisions in shaping of future built architecture. Its effect on existing cities is limited to the organisation of newly implemented infrastructure and the economic viability of preserving old urban structures.

Soft Thought

Luciana Parisi introduces “Contagious Architecture” (2013) to describe digital space composed of uncertainty and incomputability. The later, Parisi insists, is intrinsic to computation. She considers chance, randomness and complexity to be the more important and interesting aspects of computational processes. Parisi’s refreshing account of algorithms casts them as not neutral, but also not entirely deterministic: they are actual, undivided, discrete entities containing infinity and imbued with randomness.

Parisi’s most important concept in the context of this text is the notion of soft thought, which stems from “the immanent ingression of incomputable data into digital programming” (Parisi, 2013, p. xvii), that are yet to be accounted for. Coming from a background in culture studies and critical media theory, Parisi uses architecture to discuss computational processes.

Architecture is not necessarily referring to practice of making buildings, but to a systemic view of processes of creation and organisation, which she considers to be contagious. She understands digital architecture as “being already involved in the building of instances of space thought” (Parisi, 2013, p. 175). Soft thought is a vehicle to convey this open-ended view of intertwined processes of digital and physical architectures.


Speculating about human thought acquiring nonhuman perspective leads to concepts that are able to account for the presence of indetermination in quantities (Parisi, 2013, p. 55). This speculative approach goes in a reverse direction from Easterling’s articulation of urban through computational processes (infrastructural spatial technologies that structure urban form like a software).

Computation is not just another function of reason and algorithms are not reducible to a human mind and sensorimotor system. Soft thought is made of transitions between spatioetmporal entities. By which means does soft thought act on space? Soft thought is at the same time a creation process and an analytic process. All digital design is, for Parisi, an instance of soft thought, a mode of thought which is actually “a form of immanent experience” (Parisi, 2013, p. 174).

She is not looking for formal nor visual (or any other sensorial) translation of digital code to explore and explain the effects of soft thought: “Number do not have to produce something […] in order to be considered as aesthetic objects” (Parisi & Portanova, 2011). In a line similar to Easterling’s, Parisi contends that architectural form and urban infrastructures are programmed by  “the digital spatiotemporalities”, constructed by algorithms.

Soft thought, however, is “not reducible to new spatiotemporal phenomenological experiences.” (Parisi, 2013, p. 175). These experiences are  new modes of perception of space or navigation through it, brought forward by contemporary technologies. Instead, soft thought has to be considered in light of Alfred North Whitehead’s mereotopological approach[1], a model of whole-to-parts connection that describes spatiotemporal entities as coming out of prehension[2].

How does it propagate? Parisi speculates as to which algorithms construct spatiotemporal actualities. Soft thought follows the purpose of intertwining, connecting and executing. How totalizing is it? Soft thought is inherent to computation. Incomputable quantities are at the core of this new algorithmic logic that the book strives to engineer. Soft thought can be observed everywhere where computation is taking place. It is not totalizing in environmental sense, like Bratton’s stack or Easterling’s active form (which are bound to describe specific interactions between a number of actors and materialities) – it is contained within the hardware (be in computer hardware or any other “thinking” entity).

How does it then affect space? First of all, Parisi argues that it affects the built environment on the level of design. Form that is rendered in digitally based design processes is affected by incomputability that she recognizes, beyond the glitch and randomness that are sometimes used by its human authors (architects, artists). Still, we do not even yet know what the digital can do, Parisi believes.

The Layer of The Stack

The Stack is a well crafted, technically informed articulation of the nomos of our networked and increasingly technocratic world. Benjamin Bratton introduces The Stack as a way to talk about “planetary-scale computation” (Bratton, 2015). The term “stack” comes from telecommunication engineering where it refers to the software implementation of communication protocols.

This fits Bratton’s strategy: to re-articulate technical terms in the larger context of their effects on the world. Such a re-articulation should reveal the politics which is inherent in these technical systems. It is not clear, however, whether Bratton considers them to be the consequence of the Internet’s flexibility or of intractable properties inscribed in it, which land themselves to totalitarian authority – to go back to Langdon Winner’s useful distinction of the ways artefacts can have politics (Winner, 1980).

Next to the Stack, Bratton instills new meaning to the Earth, the Cloud, The City, the Address, the Interface, the User – all of which form layers constitutive of his Stack. The notions of layer and platform also play an important part in this organisation. Bratton’s articulation of The Stack bears a lot of similarity to the standard reference model of the Open Systems Interconnection (OSI) protocol suite[3].

This model prescribes how software and hardware components interact with each other in network communication. The model defines seven layers of functional elements, from the physical layer up to the application layer. Communication is regulated in such way that the decisions concerning layers on the top of hierarchy (e.g. encryption of data) do not affect the ones at the bottom (e.g. establishing a physical link), and the other way around.

This grants interoperability on a global scale regardless of the underlying internal structures and technologies. In his particular style of writing, Bratton gives condensed accounts of the way our living environment has been reconfigured by digital technologies (telecommunications, sensing, all that is digitally captured and logged, stored), followed by sequences that relax the reader with telling examples of technology (mis)use.

The Stack is nobody’s plan, it is an accidental megastructure. Bratton is interested in ways this total presence: “distorts and deforms traditional Westphalian modes of political geography, jurisdiction, and sovereignty” (Bratton, 2014), distilling itself into all levels of organisation. He identifies the clash between the two logics of governance – one based on a horizontally subdivided territory (current states) and the other based on vertical stacking of (communication) layers. Unlike its’ horizontal analog, The Stack is designed to be remade. Any layer can be replaced by something else. Anything can be connected to something else.


What can a layer of The Stack do, particularly when it is replaceable? Where is the agency of the layer to be found, specifically when we keep in mind it is nobody’s plan in particular? What would be the reason to replace the User layer with something else? If we today replace copper wire with fibre-optic cable for reasons of speed and reliability of transmission, what would be the criterion for dysfunctionality of the User?

How does a Stack layer act on space? Bratton already talks about the probable procedure to replace the User layer – currently made of human and non-human network users, each identified by a unique Internet Protocol (IP) address[4]. Is anything with an IP address a user? If so, he envisions a sort of gradual ‘overtaking’ of this layer by programmed and intelligent machines.

If this is the way we are going to communicate to forthcoming systems of governance (not government any more), then anything that can speak the Internet protocol language – the “Esperanto for machines” as Tim Wu had qualified it in his book “The Master Switch” –  could take equal part in information and decision making. The Stack to come, according to Bratton, is a cold blooded killer. In the future, artificial intelligence trained on our today’s data will not be able to distinguish us (life) from other matter. It will not know us, it will not care about us. It will be arrogant, self-righteous, totalitarian.

How does a stack layer propagate? The flow of information that Bratton describes in The Stack has come to be through the broad implementation telecommunication technologies (such as CCTV or the Internet). The Stack is a layered organisation, this property inherent to any future organisation that uses its existing layers. When we replace one layer with something else, we stay within the same organisational paradigm. Some of this, arguably, can be said for the current state of pan-surveillance and data gathering on unprecedented scale.

How totalizing is a layer? A layer of the Stack transcends geographical limits and horizontal subdivisions of space, but it also seems to transcend the scope of data gathering and preservation. Bratton talks about the probability of “burring” the cloud – all our data gathered from all sorts of tracking technologies. There seems to be no limit to this pervasive collecting activity, and it seems to already have serious political implications.

The geopolitical characteristic of every layer is that it is an overall, global pattern. It is affected by and affecting a specific level of geopolitical organisation everywhere on Earth. While the effects of Easterling’s active form are reserved to places that are receiving an upgrade (newly built or reconstructed), which more often happens in areas that are under economic (re)development, Bratton’s layers are as distributed as telecommunication infrastructure. 

Metaphors at work: the synthesis of identities

I have examined computational metaphors in architecture, media theory and philosophy. The intention was not to offer a comprehensive overview of contemporary theories, but to shed some light on spatially relevant digital processes. These processes involve algorithms that perform on space, immanent experience of computational thought and massively distributed computation and intelligence. Keller Easterling introduced active form, part of infrastructure space, which is generating repeatable spatial formulas all over the world.

Luciana Parisi introduces soft thought which describes the immanent experience of architecture organised by algorithms. Benjamin Bratton introduces layers of the Stack that are at the same time super-spatial and super-local. Spatial relevance is quite obvious in the presented concepts. They either describe change or interaction with it. They describe processes that are hard to qualify, that are not related to one particular activity (such as building or designing architecture, interaction with situated technologies, execution of an algorithm on a computer).

The work on personification of the three concepts loosely followed the technique described in early user-centred research literature previously discussed. I proceeded in this process of personification by asking “what can [a concept] do?”; “how does it act on space?”; “how does it propagate” and “how totalizing is it?”.

These four questions help to put all that has been said on a comparable level – it permits comparison and conversation between the different computational metaphor personas. I observe certain regularities in the way authors have described their concepts –  the “direction” the metaphor takes.

By “direction”, I refer to the distinction I want to make between metaphors that use a computer-based process to explain the flow of information of a non-computer-based processes (e.g. urban redevelopment), and metaphors of natural processes (e.g. thinking) that are used to explain computer-based or computational processes.  Broad use of computational metaphors to explain life and the complexity of human nature is a recognizable trend in genetic research of the last decades.

In this fashion, DNA is understood as a complete description of an organism, completely codified in chromosomes. This is a reductive approach makes this (an organism) like that (a computer) by means of the smallest common denominator – taking the mathematical notion of independence or invariance of a system as an absolute and applying it to life. The problem with this approach, as other researches have pointed out, is that it disregards the fact living organisms are made of this DNA, it is therefore impossible to distinguish between the material and the logical, between hardware and software (Montévil, Mossio, Pocheville, & Longo, 2016).

What interests me here is how metaphor can be used as a productive, enriching method in explaining natural or artificial processes. Thus I do not approach these metaphors with a critique of their likeliness to the phenomenon but rather their productivity in terms of discussing it. Metaphor carries the risks of using vaguely, poorly defined notions to draw hard consequences, as we have seen in the DNA example, as well as in the tracing of brain metaphors throughout the history of science. These seem to always correlate the way the brain works with the current, most sophisticated technology (Zarkadakēs, 2016). However, metaphor also transfers and enriches the meaning of a theoretical proposition.

The reflections presented here question of the relationship between space – a physical, immutable entity, and computing – universal calculation processes producing intangible outputs. The three concepts (active form, soft thought, stack layer) do not necessarily have any direct connection with the execution of the digital. Yet those are the concepts with which we architecture and media theorists begin to interrogate spatial effects of digital processes.


[1]Mereotopology is a formal theory of relations between parts and wholes, parts, parts of parts, and boundaries between them. In philosophy, it was first articulated by Alfred North Whitehead and it describes parts that can be bigger than wholes and wholes that become parts. Whitehead’s mereotopological approach insists on the spatialization and temporalization of actual occasions.

[2]Parisi defines prehension as “The process by which an actual entity confronts infinite data through the physical and/or conceptual selection, evaluation, inclusion, exclusion, and transformation of data, and by which it thereby invests and reprograms the actual field of potentiality.” it is an act of algorithms (or another prehending entity) grasping the (in)compatibility by ordering infinite data.

[3]The architecture of the Open System Interconnection (OSI) model was published in 1984, is part of the International Organization for Standardization (ISO) project to define a unifying standard for the architecture of all networking systems

[4]An Internet Protocol (IP) address is a numerical label (a 32 or 128 bit number, for IPv4 and IPv6 respectively) assigned to each device connected to a computer network that communicates over the Internet Protocol (IP). IP is the principal communications protocol for relaying datagrams across network boundaries, establishing the Internet in this way.