RPO Draft Circulation
As part of refining the RPO draft, I circulated my paper among peers for structured feedback. To make the insights clearer, I’ve grouped them into four key areas, each showing what worked, what needs work, and what to consider next.
Reflecting on the peer feedback, I can see that my RPO has a strong foundation in its clarity, structure, and the way the three pillars flow into one another. The feedback highlighted that the introduction, use of diagrams, and design thinking framework are working well to guide readers, which reassures me that the direction is clear and engaging. At the same time, the comments point to important areas that need further depth, particularly around methods, prototyping, and the rationale behind visualising algorithmic identity.
Moving forward, I need to sharpen how I define and apply prototyping in my context, show more explicitly how the critical journal guides my enquiry, and articulate the kinds of insights and impact I am aiming for. It also feels crucial to consider the broader scope of identity beyond online personas, and to design ways of testing or iterating my approach that make the processes of algorithmic identity more perceptible.
Rethink "Prototype"
One key piece of feedback I received was to think more critically about what “prototyping” means in the context of my project. I’d initially approached it as a way to make something visual or interactive — but I’m learning that in design research, prototyping is more about exploring ideas through making. It’s a tool for thinking, testing assumptions, revealing systems, and asking questions that aren’t always visible through theory alone.
In my project, this shift in thinking opens up a wider range of possibilities. Rather than jumping straight to polished outputs, I’m exploring how different types of prototypes can help reveal how platforms categorise us, and how our digital selves are shaped over time. Here are some directions I'm considering:
⑴ Conceptual Prototypes
• Create speculative "what if" cases, e.g., how would
my identity shift if only three data points defined me?
• Building "algorithmic personas"
that reflect how algorithms might see me, and comparing that with how I see myself
⑵ Material/Visual Prototypes
• Using p5.js to let interactions modify a
generated digital identity or identity code
• Mapping how small behaviours (likes, searches)
funnel into increasingly narrow content recommendations
⑶ Experiential Prototypes
• Performing controlled behaviours (e.g., only
engaging with sports content) on social media platforms to track how identities are shaped
•
Let users input text or images, then distorting or classifying them based on platform-like logic
to expose how filtering happens
⑷ Critical/Spectulative Prototypes
• Mock-ups of social media feeds that show
hidden metadata; why content is shown, what traits are inferred
• Interactive experiences
that exaggerate algorithmic nudging; for exmaple, gradually removing choices until only one
option remains
Andreas' feedback challenged me to think beyond interaction as a mechanic and focus more on experience, narrative, and intent. For example, in Prototype 1: Feed Your Garden, he asked: why different flowers? Can users adjust their inputs? This made me realised the metaphor I used wasn't fully thought through. I hadn't defined what the elements represented or what kind of agency the user should have.
How can I design a system where the algorithm's influence is experienced viscerally? It is the idea of making users feel the dilemma, not just showing it. The suggestion to lean into a gallery-like, curated format shifted my thinking. Rather than disconnected sketches, I'm now considering how these fragments could form an exhibition where each piece contributing to a broader narrative about algorithmic identity.
The critique of my current visual approach also pushed me to explore references like norm.to and The Straits Times' interactive features. It's not just about aesthetics, but crafting a tone, building curiousity, and framing the project as an experience.
Moving forward, I'll focus on curation, narrative, and designing interactions that expose, rather than explain the algorithm's personality.