Author: Vanya Vimalathithan

Annotated bibliography:

---

2 references drawn from the course reading list

---

1. Ways of Seeing

Citation:
Berger, J. (1972) Ways of Seeing. London: Penguin.

Images are man-made and embody ways of seeing.

Annotation:
Berger’s Ways of Seeing forms the conceptual foundation of my project, particularly his argument that images embody specific ways of seeing and that meaning is shaped through context, perception, and cultural frameworks rather than being inherent. His claim that “seeing comes before words” and that perception is never neutral informs my understanding of images as constructed rather than fixed. Rather than analysing these ideas theoretically, I treat them as a system to be tested through practice. In my iterative reworking of Venus of Urbino, I apply different “ways of seeing” through AI prompts, treating each iteration as a controlled variation in perception. This allows me to observe how meaning shifts across outputs, demonstrating that the image does not hold a singular meaning but is continuously reinterpreted through different conditions of viewing. Berger’s framework therefore moves from theory to method, shaping my process as a structured investigation into how systems of seeing construct meaning.

---

2. Design and Crime (And Other Diatribes)

Citation:
Foster, H. (2002) Design and Crime (And Other Diatribes). London: Verso.

Digital mediation transforms objects into data, allowing them to be continuously reproduced, reformatted, and consumed.

Annotation:
Foster’s discussion of digital mediation, where objects are transformed into data that can be endlessly reproduced and redesigned, directly informs my understanding of AI image-making as a system rather than a tool. His argument that design operates within continuous cycles of production and consumption parallels my iterative reworking of Venus of Urbino, where the image is repeatedly generated and altered through AI prompts. In this process, the image is no longer a fixed object, but a flow of variations shaped by computational systems. This shifts my approach away from producing a final image and towards using iteration as a method of enquiry. Foster’s framework allows me to critically position my work within these digital systems, recognising how meaning is not inherent, but constructed through processes of mediation, repetition, and transformation.

---

1 reference that is specifically related to my project in it’s theme

---

3. Bell hooks

Citation:
hooks, b. (1996) Reel to Real: Race, Sex and Class at the Movies. (Referenced via: media-studies.com)

Representation is shaped by intersecting systems of race, gender, and class within what hooks calls the “white supremacist capitalist patriarchy.

Annotation:
Bell hooks’ concept of intersectionality expands my understanding of images as sites where multiple systems of power intersect, particularly race, gender, and class. Her argument that media representations are not neutral but constructed within a “white supremacist capitalist patriarchy” challenges the idea of images as singular or objective. Instead, representation becomes a controlled system that shapes how identities are seen and understood. This directly informs my project, where I treat Venus of Urbino not as a fixed image, but as one embedded within historical and ideological systems of viewing. By iterating the image through different prompts, I begin to test how these systems shift meaning and representation. hooks’ framework pushes my enquiry beyond perception alone, introducing power as a critical layer in how images are constructed and interpreted. This allows me to recognise that systems of seeing are not neutral, but are shaped by broader cultural and political structures that influence how meaning is produced.

---

1 reference that demonstrates a critical position in context of your specific topic, medium, or method

---

4. Cindy Sherman

Citation:
Sherman, C. (1977–1980) Untitled Film Stills.

Sherman stages herself as multiple stereotypical female characters, revealing identity as constructed through repetition and visual codes.

Annotation:
Sherman’s Untitled Film Stills reveal how identity within images is constructed through repetition, performance, and cultural coding rather than inherent truth. By repeatedly staging herself as different stereotypical female roles, she exposes how femininity is produced through systems of representation shaped by the gaze and broader ideological structures. This directly informs my iterative reworking of Venus of Urbino, where the image is subjected to multiple transformations through AI prompts. Like Sherman’s use of repetition, my process generates variations of a single image, each shaped by a different system of seeing. However, while Sherman performs within existing visual codes, I extend this through computational systems, where meaning is produced not only through cultural conventions but also through algorithmic interpretation. This enables me to test how identity, power, and meaning shift when the act of seeing is mediated and automated.

---

1 reference that demonstrates a critical position in context of my specific medium/ method

---

5. AI art: the end of creativity or a new movement?

Citation:
Baxter, C. (2024) AI art: the end of creativity or a new movement? BBC Future.

AI challenges traditional ideas of creativity and authorship, positioning image-making as a collaboration between human input and machine systems.

Annotation:
This article situates AI image-making within a broader debate around creativity, authorship, and artistic intention, which directly informs my use of AI as a method of iteration. My engagement with AI began from a position of ethical uncertainty, as I questioned whether it could be considered a legitimate tool within my practice. Rather than avoiding it, I chose to test it as a system—using prompts and iteration to explore how images are produced and interpreted through it. This shifted my focus away from outcomes and towards understanding AI as a system of seeing. The article reinforces this by presenting AI not as a neutral tool, but as an active participant in image production. However, I do not treat these outputs as resolved works, but as material for further intervention. My authorship therefore emerges through how I frame, manipulate, and rework these images, allowing me to test systems of seeing while retaining control over how meaning is constructed.

---

1 wild card reference (identifying the type of relationship of the above prompts)

---

6. Marcel Duchamp

Citation:
Philadelphia Museum of Art (2017) Marcel Duchamp and the Fountain Scandal.

A mass-produced urinal was submitted as art, shifting authorship from making to selection and context.

Annotation:
Duchamp’s Fountain challenges traditional notions of artistic authorship by proposing that art can be defined through selection, positioning, and context rather than manual creation. By presenting a mass-produced object as art, he shifts the role of the artist from maker to decision-maker, foregrounding the importance of framing in constructing meaning. This informed my response to contemporary debates around AI, particularly through the BBC article, where similar tensions around creativity, authorship, and legitimacy emerge. It led me to question whether AI could be understood as a contemporary form of avant-garde practice, in the way it disrupts established ideas of image-making and artistic control. This felt like a deliberately controversial question within my own enquiry.

However, while Duchamp removes the need for making entirely, my process diverges from this position. I use AI to generate variations as a way of testing systems of seeing, but do not treat these outputs as resolved works. Instead, authorship is reintroduced through how I frame, manipulate, and rework these images. Duchamp therefore acts as a point of departure, allowing me to question authorship while ultimately repositioning it within my own process of intervention and construction.

---

My Critical Line of Enquiry

This project investigates how meaning within images is constructed and transformed through different systems of seeing. Drawing from Berger’s Ways of Seeing, I approach perception not as neutral, but as shaped by cultural, ideological, and technological frameworks. Rather than analysing these ideas theoretically, I test them through an iterative process, using AI as a system to generate multiple variations of a single image (Venus of Urbino). Each iteration applies a different condition of viewing, allowing me to observe how meaning shifts across outputs.

My use of AI emerges from a position of ethical and conceptual uncertainty, leading me to treat it not as a tool for producing final images, but as a system for experimentation. The generated outputs are not considered resolved works, but material for further intervention. Through processes of framing, cutting, and manipulation, I reassert authorship and construct meaning. The project therefore explores how images are not fixed, but continuously reinterpreted through systems of seeing, while questioning how authorship operates within computational image-making.

---

This project unfolded in three stages: imitation, hacking and synthesis. Each stage involved a process of trial, confusion and reflection, gradually shifting my understanding of augmented reality as both a tool and a medium.

1.Imitation: learning the software through repetition

The first stage of the project began with imitation. I chose an existing AR project and attempted to recreate it as accurately as possible in order to understand how the software actually works. This stage involved an extensive amount of trial and error. I repeatedly rebuilt the project in order to match what I had seen in the original work, often restarting the process several times just to understand how different elements interacted.

Through this process I learned the fundamental mechanics of the software: how assets are stacked in depth, how animation timing works, how sound can be synchronised with visual triggers, and how 2D and 3D elements coexist within the same spatial environment. What initially appeared simple was in fact very elaborate and time-consuming. Many adjustments required careful micro-tuning, and small changes often affected the entire sequence.

Although frustrating at times, this stage was extremely valuable because it created a framework for understanding how AR operates technically. By reconstructing the work piece by piece, I began to see how layering, timing and spatial relationships shape the experience. This stage ultimately set a precedent for how I would approach AR throughout the rest of the project.

2. Emotional Leakage: attempting to hack the system

The second stage asked us to move beyond imitation and begin hacking the tool, using the software in a way that subverts its intended purpose. I initially approached this by exploring AR as a navigational system that gradually breaks down. Because AR is often used for guidance and wayfinding, I thought it would be interesting to create a system that begins as a functional navigation tool but slowly deteriorates.

This experiment evolved into what I described as emotional leakage. The interface moved through stages such as minimal fracture, irritation, judgement, hesitation and eventual withdrawal of clarity, leading to a collapse of the system. The idea was that the tool itself would begin to reveal emotional instability through its behaviour.

However, when presenting this stage during Progression 2, I realised I was not fully satisfied with the result. Although the idea sounded interesting conceptually, it did not feel like a successful “hack.” Instead, it felt as though I was forcing an emotional narrative onto the tool rather than truly interrogating the medium. This created a mental block in my process because the work no longer felt aligned with the intention of the assignment.

I discussed these difficulties with my tutor, who suggested exploring other directions such as censorship or narrative control, ways in which AR could potentially intervene in how images or information are presented. While these ideas were helpful, the project still felt unresolved. Stage two left me feeling confused rather than confident about where the work should go next.

3. Final synthesis: returning to the medium

Moving into the final stage, I found myself reflecting on the earlier parts of the project, particularly the imitation phase. After stepping back and reviewing my experiments, I began to realise that some of the most interesting discoveries had already occurred during those early technical explorations. Techniques such as depth stacking and spatial layering had repeatedly appeared throughout my tests, and they seemed to offer a clearer connection between the medium and the concept.

Through conversations with my professor, I started to reconsider how these technical discoveries could inform the final outcome. Instead of trying to construct a complex narrative system or multiple AR interactions, I decided to focus on a single trigger image and explore how layering itself could become the central mechanism of the work.

This shift eventually led me to the idea of examining the male gaze within a historical painting. Rather than altering the painting directly, I used AR to fragment the image into spatial layers that only align when viewed from a particular perspective. As the viewer moves around the work, hidden elements appear behind the surface—cropped fragments, gaze markers and symbolic references that point toward the historical construction of the female body in art.

By requiring the viewer to physically move in order to reconstruct the image, the work transforms the act of looking into an active process. In this way, the final project emerges directly from the lessons learned throughout the earlier stages: experimentation with the tool, failed attempts at hacking it, and a return to the medium’s own technical possibilities as a way of generating meaning.

Reflection

Looking back, the most valuable part of this project was the process of learning through uncertainty. The assignment required not only experimentation with a new tool, but also a willingness to question and revise my own assumptions about how ideas should be developed. Several stages of the project did not work in the way I expected, and moments of technical difficulty or conceptual confusion often slowed the process down. However, these moments ultimately became productive because they forced me to return to the medium itself and reconsider what it was capable of communicating. By moving between imitation, hacking and reflection, I began to understand AR not simply as a platform for displaying content but as a system that shapes how images can be experienced. This process of learning and unlearning made the final outcome feel less like a fixed solution and more like the result of an ongoing inquiry into how tools influence the way we design, interpret and see.

Link to Project: https://contexualising-e2city.netlify.app

Our enquiry asked: How can electricity consumption at CSM be visualised across temporal scales? In dialogue with UAL’s Net Zero and Sustainability Implementation Plan (NZSIP), we focused on lighting as a measurable yet infrastructural form of consumption, where institutional decisions shape collective environmental impact.

Following preliminary research and user testing, we conducted primary data collection across all four floors of the CSM building, documenting light fixtures through systematic photography and spatial mapping. We catalogued fixtures, researched bulb types, and estimated wattage through conversations with facilities staff, producing close approximations of overall consumption. While access restrictions limited exact measurements, the process revealed something unexpected: the overwhelming density of light sources. What appeared mundane became a vast system of diffuse energy expenditure.

Our challenge was not simply to visualise data, but to do so responsibly. We were conscious of two risks: drowning the viewer in excessive information, or flattening complexity through oversimplification. To navigate this tension, we used mapping, cataloguing, and grid structures combined with compression and expansion of scale. The interactive website allows users to zoom between macro and micro perspectives. Zooming out reveals cumulative institutional load; zooming in isolates a single fixture, displaying its location, wattage, and estimated consumption. A dynamic total usage counter and seasonal time scale layer temporal context without overwhelming the interface.

By structuring access rather than presenting everything at once, we treated visualisation as interpretation, positioning graphic design as a mediator between infrastructural systems, institutional accountability, and public understanding.

---

Draft 1

Accountability in Practice

Working on this project shifted my understanding of graphic design as a relational practice. Rather than treating energy consumption as neutral data to be visualised, I became aware of how design actively reconfigures the contexts it operates within, environmental, institutional, and social.

Engaging with UAL’s Net Zero and Sustainability Implementation Plan (NZSIP) initially felt abstract. Targets, percentages, and long-term ambitions existed at a scale that felt distant from everyday practice. However, conducting on-site documentation across the CSM building—photographing, mapping, and cataloguing hundreds of light fixtures—brought those ambitions into material focus. Climate responsibility became spatial and immediate. Each bulb represented a small, seemingly insignificant decision, yet collectively formed an infrastructural system of considerable environmental weight.

This process reshaped my position as a practitioner. I became more conscious that design does not simply reflect institutional strategies; it mediates how they are understood and felt. Constructing the interface required negotiating between accessibility and complexity, resisting both spectacle and oversimplification. I realised that visualisation is never neutral, it structures perception, distributes attention, and subtly frames accountability.

Working across scales, from a single fixture to cumulative institutional load, from daily rhythms to seasonal patterns, deepened my awareness of how climate justice operates through accumulation. For me, this project clarified that responsible graphic communication is not about producing definitive answers, but about making systems legible while acknowledging their limits. It reinforced my commitment to practice as situated, iterative, and critically attentive to the contexts it inevitably reshapes.

---

Draft 2

Annotated Bibliography: Six References Informing “From Bulb to Building”

---

2 texts from the reading list:

---

Steyerl, H. (2012) ‘In Defense of the Poor Image’, in The Wretched of the Screen. Berlin: Sternberg Press.

Steyerl’s notion of the “poor image” reshaped how we thought about visibility, circulation, and value within our project. Although our focus was electricity consumption rather than image compression, Steyerl’s argument that images gain political force through distribution rather than resolution helped us reconsider fidelity. Our lighting data was incomplete due to institutional restrictions and estimation limits. Instead of perceiving this as a weakness, Steyerl’s framing allowed us to see approximation as structurally embedded within systems of access and power. The project thus became less about producing a perfect dataset and more about exposing infrastructural conditions that shape what can be seen and measured. The website interface, which layers information progressively, mirrors the idea that visibility is constructed and mediated. Steyerl’s writing deepened our awareness that transparency is never total, and that representing institutional systems involves negotiating gaps, compressions, and constraints.

---

Reinfurt, D. (2019) I-N-T-E-R-F-A-C-E: A New Program for Graphic Design. Los Angeles: Inventory Press.

Reinfurt’s conception of interface as an active structure rather than a neutral surface strongly influenced our approach to the website. He frames graphic design as programmatic—organising how information is accessed and sequenced. This resonated with our need to mediate between micro and macro scales of energy consumption. Instead of presenting all lighting data simultaneously, we structured access through zoom functionality and temporal sliders, allowing scale to become navigable rather than fixed. Reinfurt’s pedagogical emphasis on systems thinking reinforced our grid-based mapping strategy and iterative testing process. His work helped position the interface itself as a site of contextualisation: the way information is arranged shapes how responsibility and accountability are perceived. This shifted our understanding of design from representational to relational—where interaction produces meaning.

---

2 texts outside the reading list:

---

Garland, K. (1964; updated 2000) ‘First Things First’, discussed in AIGA Eye on Design (2019) ‘Why Ken Garland’s First Things First Manifesto Keeps Getting Updated’. Available at:https://eyeondesign.aiga.org/why-ken-garlands-first-things-first-manifesto-keeps-getting-updated/

Garland’s manifesto, and its continued revision, foregrounds design’s ethical responsibility within social and environmental contexts. Engaging with this text positioned our project within a lineage of designers questioning where attention and labour should be directed. By focusing on institutional energy infrastructure rather than aesthetic spectacle, our project aligns with the manifesto’s call to prioritise socially consequential communication. The updated discussions around sustainability and climate justice reframed our enquiry as more than a formal exploration of scale; it became a reflection on how graphic communication can redirect institutional focus toward environmental accountability. Garland’s insistence that design is never neutral strengthened our awareness that visualising energy data is itself a political act—shaping what is prioritised, questioned, or ignored.

---

Latour, B. (2004) ‘Why Has Critique Run out of Steam? From Matters of Fact to Matters of Concern’, Critical Inquiry, 30(2), pp. 225–248.

Latour’s distinction between “matters of fact” and “matters of concern” informed how we approached energy data. Rather than presenting electricity consumption as detached numerical fact, we aimed to frame it as a shared concern embedded within institutional systems. Latour’s argument that critique should assemble rather than merely debunk resonated with our method of mapping and layering rather than exposing failure. The project does not accuse; instead, it situates viewers within a network of infrastructural relations. By allowing users to navigate between individual fixtures and cumulative load, the interface assembles connections between small-scale actions and systemic impact. Latour’s framework strengthened our understanding of contextualisation as relational: energy consumption becomes meaningful not in isolation, but within networks of policy, architecture, and daily use.

---

2 design practices/projects:

---

Worth, J. (2015) If the Moon Were Only 1 Pixel. Available at:https://joshworth.com/dev/pixelspace/pixelspace_solarsystem.html

Worth’s project provided a crucial precedent for working with extreme shifts in scale. By translating astronomical distances into scrollable digital space, he makes incomprehensible magnitude experientially navigable. This approach directly informed our zoom-based interface, where users transition from individual light fixtures to institutional accumulation. Worth’s work demonstrates how digital interaction can convert abstract data into embodied spatial experience. Rather than compressing scale into a static infographic, he expands it temporally and spatially, allowing users to feel distance through duration. This strategy influenced our decision to incorporate temporal sliders and cumulative counters, reinforcing the experiential dimension of energy consumption. Worth’s project showed that scale can be designed as an interface condition, not merely represented visually.

---

Eliasson, O. (2023) Uncertain. Available at:https://olafureliasson.net/uncertain

Eliasson’s practice foregrounds perception, uncertainty, and environmental awareness through spatial installation. Uncertain informed our sensitivity to how viewers encounter systems rather than simply observe them. Eliasson often makes invisible forces—light, atmosphere, climate—materially perceptible without reducing their complexity. This resonated with our attempt to visualise infrastructural lighting systems without flattening nuance. His work emphasises relational experience: meaning emerges through interaction between body, space, and environment. While our medium was digital rather than architectural, the principle remains similar. By structuring zoom and temporal navigation, we aimed to produce awareness through movement rather than didactic explanation. Eliasson’s approach reinforced our understanding that climate-oriented design should cultivate attentiveness rather than overwhelm, balancing clarity with openness.

---

Draft 1
Methods of Iteration: Copying an AR Project

For this project I chose to work with Augmented Reality, a medium I had no prior experience with and one I had found intimidating because of its perceived technical and temporal demands. Although I had been curious about AR for some time, it had always seemed monumental and inaccessible within the constraints of other deadlines. This brief offered an opportunity to confront that hesitation by copying an existing AR project using KiwiCube, a platform marketed as an accessible entry point into the field.

The process immediately proved more layered and time-intensive than expected. Each illustrated element had to be produced separately, uploaded, positioned within an augmented spatial environment, distributed across multiple depths, and animated individually. Animation was particularly destabilising: small changes in axis points, angles, or timing produced disproportionately large effects. Progress took continuous cycles of trial and error watching tutorials, dissecting pre-existing templates, borrowing animation behaviours, and repeatedly testing how these translated to my own imagery.

Because much of my past work sits within food, beverage, and children’s toy packaging, I approached this project with a desire to create playful, future-facing experiences that bring a sense of magic into everyday digital encounters. This raised questions about how AR can extend surface-level graphics into spatial moments of delight, how much complexity sits behind these effects, and whether such systems genuinely open access to wonder or quietly restrict who gets to make and experience them.

Draft 2
While developing my Stage 2 AR project, I kept returning to the same question: was I producing multiple outcomes, or was I genuinely iterating? I was duplicating the same KiwiCube file, fixing the navigational grammar, and escalating tone and visual noise across different locations, but I had not yet articulated what kind of knowledge this process was generating.

Reading Conditional Design Workbook reframed this uncertainty. Andrew Blauvelt, Luna Maurer, Edo Paulus, Jonathan Puckey and Roel Wouters describe design not as the execution of a preconceived message, but as the construction of systems whose rules generate form (Blauvelt et al., 2013). Instead of expression first, they foreground constraints, procedures and repetition as the site where meaning emerges. Seen through this lens, my AR project becomes less about narrative and more about behaviour.

---

---

The work takes the language of campus wayfinding, arrows, neutral typography, QR codes, institutional phrasing, and distributes it across studios, corridors and exits. The physical triggers remain calm and administrative, but once scanned the AR layer begins to fracture: directions become partial, side-notes interrupt instructions, and visual clutter increases. Viewers can enter the system at any point, encountering either a polite interface or one already mid-collapse.

Conditional Design helps me understand this as a rule-based structure rather than six separate artworks. I fixed the navigational framework and allowed only three variables to change: tone, density and clarity. Each iteration is a minor mutation inside the same system. Authority is not declared outright; it is produced through consistency and slowly destabilised through deviation.

What became clearer through the reading is how strongly systems shape both designers and users. Repeating the same instructions forces me into a procedural role, part author, part operator, maintaining a service that is gradually failing. Meanwhile, participants are not spectators but test-cases inside a process, generating outcomes simply by following, doubting or abandoning the interface.

This has sharpened my enquiry. Rather than asking whether AR can become confrontational, I am asking how neutrality becomes persuasive in the first place. At what point does repetition turn into pressure? When does guidance slip into judgement? How much instability can an interface show before its authority collapses?

Conditional Design also made me attentive to what remains unchanged. The recognisable arrows and frames are crucial: without stability, breakdown would read as chaos rather than failure. Withholding information becomes a formal strategy rather than a narrative flourish. The interface keeps speaking, keeps occupying space, keeps performing service, while quietly refusing to complete the task it promised.

Using this text as a lens allows me to frame the project as a procedural experiment embedded in architecture. The department becomes a testing ground; each trigger image another condition; each escalation a small adjustment to the rules. In that sense, misusing AR is not a gimmick but a method, a way of letting structure, rather than intention, expose how authority and dependence are constructed through design.

---

Reference
Blauvelt, A., Maurer, L., Paulus, E., Puckey, J. and Wouters, R. (2013) Conditional Design Workbook. Amsterdam: Valiz.

Draft 3

---