Image Inhabitant

Generating an image from a composition of physical objects is a common technique for abstraction in both pre and post digital worlds. Pre-digital forms of this abstraction, when deployed by painters, required the former to find or construct a composition. The painter would build up layers of paint starting from the background working his way through to the foreground of the composition until the entire scene was rendered. The painter’s interpretation of the composition would then be complete, and the painting became a new object, representing the world as seen by the artist.

This technique for abstraction when working digitally has a similar process. In an instant, a photographer’s camera captures a scene of physical artefacts, the information is stored on the camera until it is transferred to the computer. The recorded data can then be processed to generate new variations of the original scene. While it is true that contemporary camera and image processing software pick up on centuries old abstraction methods, the techniques for processing image data have significantly evolved.  Now designers can freely manipulate the qualities of rendered images. These new techniques and software have reshaped the ways in which images operate in our world.

Stacking Layers, Blurring Edges

Digital images are edited and worked on using image processing software, opening these up to a new technical apparatus, the layer logic organisation. The Layer logic organisation abstracts imagery as content or data, allowing software users to edit composition content to generate new images through “The Stacks” organisation, native to Photoshop.  Layers of filters and content float analogically in a digital milieu, organized by their relationship to the user viewing editor.

The contribution Image_Inhabitant seeks to propose a new understanding of layers in physical space through exploring parts of a digital image’s ontology and technical capability. This is achieved by two means:

  1. 1. Firstly by understanding the ways that the layer logics of a Photoshop file generate the characteristics of images, how those logics serve as allegories for the ways in which post-digital software users work on objects to create images
  2. 2. Sceondly how to begin to think about translating images into physical space.
  3. The project presupposes that images and Architecture share some basic organisational logics, as both are assemblies of many components into whole objects.  The joints where these components begin to entangle are either emphasised to play up the depth or are flattened to render in two dimensions the connection between the individual components.
  4. Within images, the components that make up assemblies are called layers. Layers are bi-dimensional abstract objects that are then worked on by the user through image processing software. The purpose of understanding layers as assemblies of component layer elements is to better understand the relationship between objects that make up an image, and the way that depth is structured through parts of an image.
  5. Digital images are worked on through Photoshop, or other image processing software that manipulate the qualities of the two dimensional objects or layers. These are then organised into structural systems (The Stack) that manipulate the composition of the image through Layer management and transparency, which play on edges, and depth.
  6. The following essay contains case studies that examine modes of working on objects through image processing software, and the design for a sculpture park that attempts to simulate some of the technical qualities of software environments, as a potential physical space.

Gifs provide an opportunity to examine an artefact from both a three-dimensional and a two-dimensional persoective. In the gif above a three-dimensional object rotates around a fixed point in space. In every other frame, the object flattens its relationship to the camera, pointing to both the three-dimensional rendered object as something that exists somewhere, while also pointing to the gif as an object assembled of two-dimensional layers. Flatness in an image somehow represents the formers ability to point to itself, and to distract from the viewing of the original composition. Exaggerated depth is not explored in the gif, but the effect of “hyper depth” is similar to flatness, revealing the two-dimensional nature of the image. The “two and half” dimensional quality of the gif shows the potential for an object to blur the line between object and image.

Layer History – Cubism to Photoshop

Edges and depth are topics that have historically been explored by numerous Architects. The manipulation of depth and edges was a focus for Peter Eisenmann, Colin Rowe, and Robert Slutzky who took up studying abstract art and Cubist Painting. Layers significantly entered architectural discourse through the writings of the above mentioned architects in the ‘Transparency Articles’ written in the middle of the 20thcentury.[1]  Through an understanding of the body in space, the authors here employed formal analysis to investigate the relationship between the viewing subject and the object on view. Rowe, Slutzky, and Eisenman’s interest manifested itself as a result of the analysis of the work of cubist painters, with the aim of understanding the depth that they employed in their paintings, particularly the way they subverted the traditional reading of this.

The similarities between the modes by which a viewer reads a cubist painting and the ways in which the layers of a Photoshop file are structured are startling. Axonometric diagrams offer a critical reading of the layer logics of a cubist painting. By establishing a relation to an axonometric diagram delaminating the layers of a Photoshop file, reveal the same use of planar objects, implied space, and subject-object relationships. Both diagrams engage in a perceptual reading of planar elements in a stack that move back in space, while remaining parallel to one another. Photoshop files leverage similar perceptual systems to those inherent in the formal analysis done by Rowe and Slutzky of cubist paintings. Content and modifications are organized into layers and then offset in the ambiguous milieu native to two-dimensional graphics software.

[1]Colin Rowe and Robert Slutzky. “Transparency: Literal and Phenomenal.” Perspecta 8 (1963): 46-48.

In another experiment, a series of ‘wall objects’ exist perpendicular to one another in two directions, creating a pseudo-garden. The garden encloses a series of artefacts that exist within the milieu of the walls. Neon lights exist on the perimeter of the translucent coloured walls, and a reflective ground reflects the neon light up into space, or off the wall surfaces.

The view exploited is the axonometric. Those viewing the render get a sense for the whole and are able to visualise a series of interlocking walls that create spaces of various sizes on the inside of the garden.  Here the axonometric provides a reading that encloses the ground of the project. What’s missing are the images of the garden spaces. Translucency and light reveal to the viewer glimpses into the qualities of some spaces, but ultimately do not foil the Axonometric view. The viewer cannot see between the layers.

This study reveals some shortcomings of the single axonometric view and attempts to take elements of planar formal studies and deploy digital effects onto those formal studies.  

Image_Layer – Form and Augmentation

While it’s interesting that the articles Rowe, Slutzky and Eisenman wrote as a part of their research on transparency share similarities with Photoshop software, the characteristics that distinguish Photoshop are necessary for moving this type of perceptual system into contemporary discourse[1].  In his book, Software Takes Command, Lev Manovich provides a breakdown of the technical systems native to Photoshop that differentiate it from the pre-digital media types. Layers are instrumental in generating several of the characteristics native to images. In his 2013 book Lev Manovich provides a definition of an image: “An image is thus redefined as a provisional composite of both content elements and various modification operations that are conceptually separate from these elements.”[2]

There are two important characteristic of imagery that are products of layer organisation.

  1. 1. The first quality is concerned with how layers separate a composition into many different elements, allowing a designer to work on those elements individually and non-linearly to produce a composition.
  2. 2. The second characteristic is the leveraging of “augmentation” or modifications by giving augmentation layers the same digital and structural weight as content layers.[3]


[2]Lev Manovich. “Understanding Metamedia.” In Software Takes Command, 142.


The design project, Image_Layer seeks to examine the relationship between augmentation and mass elements by separating mass or form from augmentation or effect.  Mass objects and effects are scattered around the grounds of a sculpture park, arrayed in a series of different conditions.

The project calls for a complex of sculptures, objects, and programs, set inside another larger sculpture park, which acts as a backdrop, a background layer. The first category of objects, Mass objects, are white matte artefacts meant to simulate immaterial form. These take the shape of rock formations and building forms. The second category, Effect Objects, appropriate techniques like color fields, gradients, reflective surfaces, texture mapping, and vector graphics. These become the translucent walls, and neon lights.

Visitors to the sculpture park orbit around the sculptures and programs of the complex through an infrastructural system that structures views onto mass objects and allows visitors to interact and edit the effects acting upon the mass objects. The infrastructural system’s objecthood is negotiated by its role as an interface for the environment of the sculpture park, rather than a formal or material category. The system presents visitors with the opportunity to experience the sculpture park through both standing just inside the spaces of the park, and just outside looking in. As visitors move around the space they are offered a series of moments where effect and mass align to form seemingly completed images, and other moments where objects are misaligned. The spaces of the complex are loosely analogous to a compositional or logic system imagined for rhino space, or the space of a Photoshop file. The park is both segmented into parts by the infrastructural system that runs through the center, and whose arms reach out into the separate program spaces, and then generate views.

Image_Inhabitant – Living (in) images

In Photoshop files, layers exist as separate elements analogically floating above and below one another in digital space, never touching. Individual elements and layers are worked on separately from the original Photoshop hierarchy and then re-inserted and tested into the composition when the desired conditions have been met. This new methodology is in line with the idea of working in “real time,”[1]rather than a traditional linear approach.

The Pre-digital approach required decisions to be made with each new element being added to a composition. Decision-making was linear, one element had to be placed before placing another. At best, the penalties for interrupting this process required designers to rethink their composition, and at worst to restart their work, or at the very least contend with erasing their mistakes from the composition. The Photoshop layer logics creates a new paradigm for designers allowing to them to make tentative decisions about compositional elements without committing to something or contending with erasure. The environment of a Photoshop file serves as much as a digital workspace for design experiments as it does for the final composition. Layers of promising textures, compositional elements, and more can be worked on, evaluated, then turned off or on with no repercussion to the overall composition, which allows a designer to play around without committing to a configuration. Pre-digital work would have seen a slow buildup of content or layers and required that the designer carefully manipulate the applied materials to not ruin the composition. The significance of layer organization is that it allows for a new flexibility and a looser connection to the components being structured.


In the last case study two gifs are rendered, one shows the exterior of a house, whilst the other features an object in front of three material planes. The house form is remapped with a series of different materials, and then the series of images of the house form a cycle, presenting the viewer with several different readings of the same subject. In the second gif a few different material combinations cycle through and give different readings of the space between them, as the qualities of the spaces that they exist next to change, while also recoding the object behind the three material panes. These two gifs attempt to simulate interior and exterior relationships of a residential project. These studies signal the fluidity with which augmentations can shift and slide around a form object.

A residential program became an important instrument within the experiments executed by the authors and architects of the transparency articles, which eventually were used to construct diagrams of spatial relationships born from a critical analysis of cubist paintings, and research into pre-modern villas. In this same way, Image_inhabitant seeks to take on aspects of new modes of abstraction through attempting to take on residential program. The next step in translating the layer structures of an image into physical artifacts should come in developing the connections between interior and exterior space. The house study above plays with ideas of envelope as boundary object, and programs that float around in the space inside the envelope.   Layers of substrate hold graphics, and at time hides interior program, and other times reveal the contents of the residence.

Photoshop layer logics separate content elements from augmentation elements, creating two categories. Augmentation layers take the form of filters and effects and serve as mechanisms for working on the composition of a Photoshop file through the relationship of the augmentation layers with other subsequent layers. Pre-digital media dealt with augmentation of content through direct application of the augmentation to content, making the two inseparable from one another. Augmentation of content was synonymous with the content itself.  When content was augmented in a painting or other pre-digital media, those augmentations were subject to the penalties that a designer would be if they edited content elements. Augmentation layers can be applied to a Photoshop file by generating a layer that hosts that effect, which can then be manipulated to create the desired result or outcome. Photoshop has a variety of preloaded effects that users can mobilize to create the affect that augments the image. The position of these layers can then be moved within the layer stack with the same freedom as content layers, shifting their locations to affect layers underneath them. Layers can also be directly hosted to a single content layer, ignoring layers above and below the desired content layer. Working with layers in this way changes the user’s interpretation of the content they’re acting upon. Similarly, layer organization also allows for the separation of content and effect is that users of Photoshop software can now work on both categories of layers autonomously, further abstracting the content from its origin.