#longform ### **Fictions: A Speculative Account of Design Mediums** ## 2016 The Albertian paradigm of architecture as an allographic practice implies that architectural [[Design]] comprises forms of notation and representation. It would seem that mediums2 are all that architects engage with. Architecture is primarily a cultural, visual practice that operates through design, understood as composition and the arrangement of relations. While architects work with drawings and models, they primarily produce images. With the advent of the digital, according to media scholar Lev Manovich, other media (print, photography, radio, film, etc) have been collapsed and integrated into _software_ as a meta-medium; in almost all areas of contemporary life, [[Software]] takes command.3 According to Marshall McLuhan, when a new medium appears and before it inevitably supersedes the preceding one, it does its best to simulate it. Hence, when cinema emerged in the late 19th century, its formal vocabulary was that of the theatre until it discovered its own medium specificity – montage and movable camera.4 Importantly, a change comes to the old medium as well. After cinema took over some of the classical representational and, in that sense, political responsibilities of theatre, the focus of theatrical production was shifting more and more towards the participatory and the situational elements, retaining and honing the literary component while moving away from the visual. Eventually, these developments gave rise to modernist theatre and other complex forms. The specificity of theatre was rediscovered in focusing on the living presence of an actor’s body and voice. In another example[\[1\]](A%20Speculative%20Acount%20of%20Design%20Mediums.md#_msocom_1) , with the introduction of photography, painting was introduced to a new specificity in the form of abstraction. [\[2\]](A%20Speculative%20Acount%20of%20Design%20Mediums.md#_msocom_2) What happens when a medium contains all other mediums? Everything changes, yet the issue of software's specificity is rarely addressed. In response, Clement Greenberg’s notion of medium specificity can be reintroduced with regard to the problem of architectural design understood as a software practice. The question becomes: What is it that software can do, that no other medium can? Firstly, it is important to make a distinction between the conditions _for_ and the effects _of_ mediums. In the case of cinema, the technical conditions of the medium involve employing a number of discrete units (images) to create an illusion of movement as an effect. The fact that film operates with discrete units does not prevent its effect from being perceived as continuous. The same goes for software. Its dependence on hardware that currently operates in binary states (since we still do not have quantum computers), tells us nothing of the vast field of sensorial effects that it engenders. Similarly, abstract data, infinities and random values may be at the core of computation, but computation only becomes available as a problem in design methodology once its conditions and effects are coded in such a way so as to be readable. This is where questions of code and its relation to language in general come to the fore. Yet, coding as a practice and code in general, its apparent similarity to writing notwithstanding, does not immediately lend itself to any kind of aesthetic analysis. Coding depends on the axiomatic, mathematical model and belongs to a different semiology. Only when the outcomes of code become visible as something else than code, does software (and computation) become interesting for design. The crucial question is _how_ these outcomes become visible, and under what circumstances this visibility operates? Algorithms and Interfaces Algorithms form the core of software's medium specificity, and they produce crucial effects, like interactivity. The discussion on the nature of algorithms in relation to architecture becomes possible only when algorithms become visible, that is, only when an _interface_ is involved. This is precisely why the computational question in architecture should never be equated with its numerical basis, i.e. the _quantities_, but with how these quantities firstly become manifest optically and then as visual and semiotic _qualities_. Code is the basic interface, and yet, architectural design is a visual, cultural practice defined by its focus on compositional issues. Design procedures in the digital age are computational inasmuch as they depend on functions of language as code, and code as a representation of space in the forms of design software. In other words, the conditions of a medium become important only when a question of composition comes to the fore, and only if the conditions themselves can be shown as being composed and composing. For instance, there is no point in looking into the chemical processes of film in order to understand the film's meaning, yet the film stock properties leave a definitive imprint on the composition of an image, already working towards an image composition long prior to the film’s treatment in post-production. To avoid the pitfalls and obfuscation stemming from confusing a visual, cultural practice such as architecture with a scientific practice such as chemistry, a simple rule can be applied: If a procedure is not available for forms of reading in relation to composition (here understood in the broadest sense as the act of combining parts to form a whole), it is not relevant. As with any medium, in the case of design software there exist conditions that operate as composers of space well before any input from the user. Hence, what does software, understood in these terms, mean for architecture? Architecture relies on its traditional modes of representation for design, which mostly comprise various projection-based imagery, either orthographic or perspectival. As a concept, projection lies at the core of architectural design and continues to do so with software as well. Yet software introduces other modes of projection, of which active projection is by far the most important, since it enables interactivity. Every projection is coupled with a _gaze,_ and every projection operates on _grids,_ which are the primary design objects_._ More importantly, the algorithmic nature of software enables these projections to be populated with new and unforeseen _grids_. [\[3\]](A%20Speculative%20Acount%20of%20Design%20Mediums.md#_msocom_3) Principally, a grid is any digital object that has become visible as an interface. In this sense, any projection is always already compositional, and in turn, always already political. Hence, the true value of software in architecture is that it constructs new modes of projection and new modes of vision, as well as that it enables new models of grids. Of all the sub-mediums architects usually work with, only renderings convey a degree of complexity in terms of ambiance, mood and atmosphere. Architects are content in making only the necessary documents that their discipline demands, thus leaving the whole world of new, virtual and interactive spatiality to others. The enormous size, complexity, richness and attention to detail of some contemporary computer game worlds exemplify what this new spatiality can be. The overwhelming feeling of immersion and saturation within these worlds is the result of spatial design. Hence, as far as design methodology is concerned, architects may have as much to learn from computer game and software designers as from the histories of architecture or the vast majority of contemporary practices. Yet, it is not only that the architectural discipline will find itself in an era where humanity will inhabit and experience artificial worlds in a way not at all different to how it experiences ‘real life’, but the very conditions in which the new architectural additions to ‘real life’ are being produced, organised and disseminated are already completely set in the virtuality of digital and algorithmic worlds. The dominance of design software packages that come out of the legacy of Computer Aided Design maintains traditional architectural design methodology and ensures the endless reproduction of traditional design notations and their elements, some of which have already become almost obsolete (scale, for example). The latest iteration of the CAD paradigm is BIM, which may yet prove to be the greatest threat to the discipline as a design practice – since the BIM paradigm is principally about project management rather than design. It is no wonder then, that the most interesting design work today comes from the use of exotic and custom-made software or software whose original area of application is not architecture: Maya, ZBrush, Softimage, Houdini, Unity – or directly from programming languages like Processing. Hence, the role of software in architecture has been largely misunderstood: firstly, by disregarding software specificity and focusing on simulations of the traditional design medium in software in an attempt to preserve the discipline as it was historically, and out of necessity, defined; secondly, by amplifying the incidental and non-disciplinary effects of software, through using it as a tool for simulation of natural processes. What is needed is a radical embrace of software specificity understood as a new _visuality_ – that is, a radically new _vision system_ for architecture, and as a new ground for _architectural fictions_.5 A speculative history of design media Historically, the medium specificity of paper was given by its flatness and expendable nature which provided the perfect conditions for the rise of very specific architectural sub-mediums: orthographic projection-based plans, sections, elevations, perspectives, iso- and axonometric drawings. Since Alberti, architects have dealt with forms of representation without having to worry whether or not representations will take command and trounce ‘reality’. Architects use software principally as a simulation tool, which is particularly apparent in the practice of rendering. Rendering is simply perspectival drawing made on a computer. Insofar as Alberti was right in saying that architects do not build but make representations of buildings, renders can be seen as the key product of an architectural practice. Renders are images that trace their lineage to the rules of perspective and come out of the long tradition of mimetic representation that characterised pre-modern painting. They are the perfect example of a purely software-based phenomenon used and regarded as if it has nothing to do with software. Contemporary layer-based digital image making (which is the basis of software like Photoshop) forms the basis of architectural representation today, and yet even when it draws lessons from twentieth century cinematography, by design it remains locked firmly in the tradition of passive representation, that of a photo collage. Renders are expected to be nothing more than idealised images of a new architectural reality; they come with their own vocabulary and tropes (balloons, children, vegetation on roofs, cherry blossom trees…). When, in rare cases, an office uses a video presentation of its architecture, this still retains the passivity of an image. The use of a VR platform6 allows the clients the joy of inhabitation where the ‘body’ of a possible architecture participates through telepresence. Although virtual reality might eventually replace renders as the primary means of representation of an architectural space, it will change architecture as a discipline only if a similar environment becomes available as a _design_ medium as well. The chronic delay of fabrication and building technologies in comparison to design technologies presents an incredible bottleneck for the discipline. This paradox is shown by the fact that architects do not build yet are obsessed with the imperative of buildability. The Albertian ideal of an architect as a pure maker of spatial ideas will be fully actuated when this paradox is resolved through the flattening of design space with the real space, either by 3D printing or the _utility fog7_ or a similar idea. The Gaze and the Grid Spatial visual media can be analysed based on two concepts which both have to do with projections: The _Gaze_ and the _Grid_. The gaze is our visual access to the model space, which in turn depends on projection. The gaze is never objective, far from disinterested and always intentional. Traditionally, the need for notating the space for the purpose of preserving the design intention as well as information for building construction has ensured that orthographic projections rule the architectural design process. This implies that there is such a thing as an _orthographic gaze_ as well as a _perspectival gaze._ As with any other tool, orthographic projections are not devoid of aesthetic and political implications. They produce very specific spatial outcomes as they depend on a very specific set of presumptions. Historically, the architectural discipline has been identified with a special skill set that relies heavily on planar thinking. The word _plan_ testifies to this; it has both the meaning of a plan and planning. To plan in architecture is to partake in a political practice enabled by the medium of a plan. More specifically, this political practice is engendered by the gaze that this medium affords. This gaze can generically be identified as a top view, in case of plans, and as a side view, in cases of sections and elevations. A top view implies the idea of total control of the model space and is particularly good at enabling any idea that has to do with a central hierarchy and centralised political authority, as exemplified by Roman city planning (Cardo and Decumanus); centrally planned temples; the ideal villa of Palladio; the nine square and the four square grids. Here symmetry is particularly important since it is producible exclusively in the planar mode. The _ground_ is another important notion: In the planar top view, the ground becomes abstracted into a background (which is the original meaning of ‘ground’ in a figure-ground problem in perception), and this becomes clear with the introduction of the Nolli map. In sections and elevations another kind of relationship becomes apparent: the hierarchical dependence on the ground datum. In other words, the disciplinary problem of the figure-ground relationship is twofold since it originates both in the top and side views but has different implications in each. The role of orthographic representations is to ensure the preservation of dimensions, but an unseen consequence is that they impose the flat organisational and compositional principles on the model space, thus saturating the outcomes with abstraction. Perspectives are a different concept as they are usually made after the fact of design, to add atmosphere and an illusion of life. Perspectival projection hints at the idea of subjective space where the vanishing point is inverted into the eye of the observer. One of its effects is that it enables a specific reading of a picture plane. It not only organises the space but reorganises the observer as well and engenders a specific form of relationship between the two that can be conceptualised as _entanglement_. The notion of gaze arises in the form of a mutual gaze facilitated through an abstract _grid_ diagram. There is evidence that perspectival projection has been used as a design tool as well, for example in the Renaissance,8 but perspectives have historically been understood as ‘too subjective’ and, more importantly, imprecise to be used as design tools. Modernism introduced parallel-projection based representations as an assumed _objective_ mode of looking at the model space. The gaze embedded in parallel projection promoted another variant of the totalizing ‘god mode’ look that preserved the dimensions of the plan while simulating three-dimensionality. It follows that the compositional problems in architecture are inescapably governed by the mediums: planar and volumetric representations. These are not merely representations in the usual sense of the words, they are themselves projective systems, systems that generate a spatial outcome instead of just recording one. This relationship is an interesting inversion of the Renaissance invention of drawing as ray casting; it could be said that any act of design is ray casting in reverse. If so, Alberti's and Dürer's ‘prince of rays_’_ has been conflated with a reverse-direction ‘god ray_’_ of the designer's gaze. Interactivity in software starts with the gaze, in other words, with a specific visual access to the model. Unlike traditional perspective or axonometry, the software 'perspective view' imposes a fully actuated, fully accessible model space. Though still restricted to the two-dimensional plane of a screen, it is an interactive projection that liberates spatial outcomes from the constraints of fixed projection systems. The user can move and orbit around, zoom in and out of the model space and thus gain access to every aspect of its spatiality. Plans and sections are restrictive because they do not afford this access. The move to software is a move from the flat organisational diagram into a volumetric diagram. This implies a more fluid relationship with the underlying organisation and a less rigid set of rules. In contemporary practice and due to the use of software, plans and sections are increasingly made after the design, and practically no design is ever done only from a plan. The model space enabled by software collapses the different projection spaces into a volumetric diagram. This engenders very specific spatial outcomes, for example, the traditional relationship between the plan and the façade becomes obscured and the façade becomes either an intentional cut through the volume or is literally a three-dimensional envelope. This unification of model space enables a different outcome than what was possible in a time when space was modelled in separate orthographic views. For one thing, it does away with the idea that uni-directionality is a compositional and organisational default. The traditional, orthographic space can thus be described as a disassociated, fragmented space that had to be _stitched_ together, and it is precisely in this stitching that the traditional practice found its modus operandi. The modernist grid is a perfect example: An endless, equal potential space that only actually functions in two dimensions and makes stacking a solution to every height problem. The _Fondation Louis Vuitton_ in Paris by Frank Gehry is a building that has been designed solely on the basis of a volumetric diagram. The plan is no longer a generator; it is merely generated. Hence, orthographic mediums are exposed as ultimate abstractions of living space that have become misidentified as guarantors for disciplinary specificity. Still, when the traditional and contemporary mediums do not share the same type of gaze anymore, they have a common base for their spatial models: The grid. The base of every modelling software is the grid, just as it has always been for any graphic procedure based on projection. If software is still dependent on projection, its grid is now separated from the gaze. In other words, the projection is interactive and does not restrict the model space to a two-dimensional space but engenders a fully actuated, interactive projection of total immersion. This means that for the first time in the history of architecture, there is the possibility of visually inhabiting a design space. Historically, the grid has undergone a series of transformations, and this history can be found in software in a compressed and accelerated version. The orthographic, perspectival and isometric grids have been replaced by the active and volumetric grids of software. From the traditional grids of early 2D CAD packages, to the contemporary high-resolution (aka high-poly) grids of various sculpting and procedural-based software, design is not the design of objects, but of grids, in grids and on grids. Any model designed in software is a grid: A mesh or NURBS based, low- or high-resolution, uniform or deformed grid. And every model is instantiated on another grid, the ubiquitous _ground grid_. The Ground Grid Perhaps the history of ground grids is the most interesting, as they have been present at least since the Renaissance. [\[4\]](A%20Speculative%20Acount%20of%20Design%20Mediums.md#_msocom_4) This ground as flat grid is first found in the so-called _Prevedari engraving_ of 1481, made by Bernandino Prevedari after Donato Bramante and named _Interior of a Temple with Figures_. It, or its variations, are always found in the depictions of ideal places, such as in Raphael's _The Marriage of The Virgin_. These places are never simple landscapes; they are the originators of urbanity, yet they always have an atmosphere of an ideal nature. A flat grid indicates a perfect balance between tamed, lived-in nature and enlightened city dwelling – a negotiation between opposites and a sign of harmony. It is a utopian sign _par excellence_, and it is no wonder that it comes back in modernity with such force. It is this inescapability of the grid that has haunted architecture in the late twentieth century. In this sense, deconstruction was primarily a move against the isotropic grid, featuring instead fractured, broken grids supposed to engender new, non-privileged subjectivities. The flat grid has since become ubiquitous, its latest iterations being equated with popular depictions of the digital realm, such as the one found in the film _Tron_ of 1982 and 2010. However, it is precisely in 3D design software that the grid finally takes over. Ultimately, the image of the digital is the image of an endless grid accommodating other grids, manipulated by an omniscient and omnipresent designer. In recent years, another metaphor, that of a _cloud_, has come to represent the _digital_ _regime_, almost as an attempted escape from this perceived artificiality. Yet, maybe it is the case that these two ideal images are collapsed: it is as if somehow, in a perfect union between the natural and artificial, the world has become an endless, reflective grid mirroring the clouded sky above. In this sense, and with seeing the rise of planetary-scale computation in hindsight, Superstudio's _Supersurface_ of 1972, a project that has been understood as a conceptual, ironic utopia which projects an endless, isotropic grid taking over the world, can now be actually read as _the_ realist project for the twenty-first century. Games Software flattens the field of visual effects, and enables messy encounters between drawing, painting, video and games. As a direct descendant of traditional design mediums, design software prescribes a very specific role to the user: that of a disinterested, disembodied subject that has a full access to any projection space, that operates on a spectrum of full visibility and full zoom-in. This approach continues and vastly expands a specific subjectivity of an architect operating in the ‘god-mode’ of the traditional discipline. An architect is now an omnipresent and omniscient entity with full control over the design space, which supposedly ensures that his authorship is visibly imprinted. This notion of total visual empowerment is a heritage of the military roots of the digital regime9, and leads ultimately to a very problematic and unexamined political outcomes of design processes, best witnessed in totalising fictions like Parametricism. Unlike other software, computer games tend to problematise the notion of subjective agency through either exposing and putting into question the ability of a player, or by disturbing the mere notion of a goal. Because of their full spectrum deployment of interactivity, games could be thought of as the most medium-specific type of software. Play does not have to be always goal-oriented, and although most games do have a goal (the ‘win’ state), more and more the inherent specificity of experience leads to the player being content with merely ‘existing’ within a game. Immersion does not depend on, and is more likely even disturbed by, direct calls for action towards reaching a goal. The notions of agency and authorship are thus perceived in a different manner, which enables loosening up the idea of control. It is precisely the notion of loose control that can be postulated as a new authorial model. Rather than depending on guaranteed outcomes that either come out of total control of the medium, or out of a system-based logic of computation, this notion puts the possibility of a new subject first. A subject that is aware of his own entanglement with other, non-human forms of agency, and is willing to explore new configurations coming out of this flat, non-hierarchical relationship. This notion of obstructing prevalent, established rational and positivistic methods through employing game-like scenarios and practices is not new. It was used extensively by the Surrealists [\[5\]](A%20Speculative%20Acount%20of%20Design%20Mediums.md#_msocom_5) as a means of disestablishing the aesthetic and political implications of well-known models. In the surrealist games such as the Exquisite Corpse10, any notion of systematic, rational form-making is erased. The games were derived as means of freeing the creative process of conscious control. In August 2015, Autodesk presented a new software package that will allow architects to finally inhabit the spaces before they are actually built. Named _Stingray_, the software is actually a game engine, in the tradition of Unity 3D and Unreal. Exactly like those 3D applications as well as others, _Stingray_ employs a grid as a ground and affords a gaze, yet this time a very specific one. Its default view is that of an endless, walkable grid, observed from first person, enclosed by an endless, clouded sky. Endnotes: 1 Mario Carpo, in _The Alphabet and the Algorithm_, (MIT Press, 2011). 2 ‘Media’ and ‘mediums’ are optional terms. Here ‘mediums’ is used to ensure difference from broadcasting. 3 Lev Manovich, _Software Takes Command_, (Bloomsbury, 2013). 4 ‘A new medium is never an addition to an old one, nor does it leave the old one in peace. It never ceases to oppress the older media until it finds new shapes and positions for them,’ in _Essential McLuhan,_ Edited by Eric McLuhan & Frank Zingrone (Routledge 1997), 278. 5 The notion of _formats_ – a project or a building becomes just one format in a flat ontology of formats, where software is the medium. See David Joselit, _After Art_, (Princeton University Press, 2013), 55. 6 For example, Oculus Rift and HTC Vive offer unparalleled possibilities of immersion. 7 The Utility fog (coined by Dr. John Storrs Hall in 1993) is a hypothetical collection of tiny robots that can replicate a physical structure. As such, it is a form of self-reconfiguring modular robotics. Source: [https://en.wikipedia.org/wiki/Utility\_fog](https://en.wikipedia.org/wiki/Utility_fog) 8 Erwin Panofsky, _Perspective as Symbolic Form_, (MIT Press, Zone Books, 1996). 9 Friedrich Kittler, ‘Computer Graphics: A Semi-Technical Introduction’, _Grey Room_ (2001), 31. 10 Alastair Brotchie and Mel Gooding, _The Book of Surrealist Games_ (Shambhala, 1995), 25.