Week 8: Writing the Thesis

Overview

This week shifted from building to writing. The 1-on-1 with Camila confirmed the direction: the narrative layer is the right move, the 8 curated spaces are strong, and the paper needs to land by Sunday. Most of the week was spent revising the thesis draft, not just polishing sentences but clarifying what the project actually argues.

1-on-1 Feedback (3/22)

Camila’s feedback reinforced several things and surfaced a few new ones:

On the experience direction:

She liked the idea of narrated spaces over a gallery format. The touching moments from the first iteration felt more like immersive storytelling than browsing.
The sunset scene resonated. She pointed out that as a lighting designer, I probably notice the way light changes on skin in ways most people don’t, and that this is worth leaning into.
She suggested the narration could go poetic rather than purely descriptive.

On presentation:

A reminder to always start with the one-sentence framing before showing the work. At Demo Day, I jumped straight into the demo without grounding the audience in what the project is about.

On the paper:

Two diagrams would help: one for the experience flow (viewer’s journey through the 8 spaces), one for the technical system (capture pipeline, UE, VR).
Keep APA format. Don’t forget list of figures, vita, page numbers, title page.
Draft due between Friday and Sunday 6pm. She’ll read them as they come in.

Next week:

Bring the headset. Camila wants to experience it firsthand.
Class sharing: present where the project is now.

Thesis Draft Revision (3/26-3/28)

The bulk of the week was rewriting the thesis draft. Not just copy-editing but rethinking what the paper actually says.

Abstract

Rewrote from scratch. The old abstract summarized what I did. The new one makes an argument in three parts:

What I did: Built a curated VR archive of personally meaningful places using 3DGS, with light, sound, and narration drawn from my background as a lighting designer.
What I found: Presence works, but it’s not enough. The design language that makes a captured space feel like a shared memory comes not from technology but from how we actually remember: light, sound, a reason to be there.
What it means: The tools exist. The question is how to design within them.

3DGS as Memory (new section in 3.1)

Added a paragraph about the material quality of 3DGS itself. Gaussian Splatting is not opaque. Up close, it is translucent, soft, almost smoke-like. Edges dissolve. Details blur into abstraction. This is not a limitation. It carries a quality closer to how we actually remember. Memory is not fixed; it shifts over time, changes with experience. The medium mirrors the message.

This became one of the key arguments threaded through the paper: the design language is not imposed on the space from outside. It comes from memory itself.

Light and Sound (merged in 4.1)

Previously, Light, Sound, and Narration were three separate subsections. Light and Sound now form one section: they both rebuild the sensory atmosphere of a remembered place. Narration is separate because it does something different: it brings meaning and context.

The Okinawa example threads through both: the light field recreates the sunset, and the narration says “When I see a sunset, I always raise my hands to catch the light, watching the orange and pink colors flow on my skin.” That gesture is the memory. The light field is an attempt to let someone else stand in the same warmth.

Conclusion (5.1 revised)

Updated the reflection to land the new argument: the design language comes from memory itself. The elements that make a captured space feel like a place (light, sound, a voice sharing why it matters) are the same elements we carry in every memory but rarely name. 3DGS even mirrors this in its material: translucent, not fixed, closer to how we actually remember than to a perfect copy.

Other changes

Figure renumbering: swapped pipeline (F2) and temple (F3), removed old Sharp figure, added UE lighting setup screenshot (F11). Now F1-F12.
Removed Limitations section (5.2). Will add back if needed.
Fixed figure references throughout (1a-d mapped to correct images in 2.3).

Current State

The draft is complete: Abstract through Chapter 5, References, and Vita. 12 figures. APA format with front matter (title page, approval, committee, vita, abstract, TOC, list of figures).

Remaining before submission:

Update onboarding screenshots (Phase 2: 3 steps instead of 4)
Resolve WIP language in Section 4.2
Final page count check in Word
Update TOC in Word

What’s Next

Submit thesis draft (deadline: Sunday 3/29 6pm)
Bring headset to class for Camila
Continue refining narration scripts for the 8 spaces
Prepare class presentation on current state