ABOUT THIS SITE
A small darkroom you can walk into.
Amoment is a website about photographs, film, and the moments in between. It is not a gallery: you choose a camera, load a film stock, develop a photograph, and print it on paper. The whole thing is slow. That is on purpose.
Why I built it
Most photo sites want you to scroll past a hundred images and leave. I wanted the opposite: a place that asks you to do something, and to wait. You pull the cord, you choose the paper, the photograph is only finished once it is printed. None of these steps are efficient. All of them make you stay.
The name carries the same idea. Film holds a moment one frame at a time; light holds it in a single exposure. Amoment, held.
Four rooms
-
ROOM 01
Cover · the darkroom light
The site opens in the dark with a single safelight and a pull cord. Pull it, the room lights up, and you step inside.
-
ROOM 02
Camera room · pick your gear
Six cameras and seven film stocks, each one you can drag to rotate. Every stock has its own colour character: Gold runs warm, HP5 is black and white.
-
ROOM 03
Workspace · develop
Drop in one of your own photographs, choose a film look, then drag the divider to compare before and after. The photo never leaves your browser.
-
ROOM 04
Darkroom · print
Choose a paper border, print the frame, and export a JPG. Only at this point is the photograph finished.
How it is made
- StructureSemantic HTML throughout: header, main, section, aside, footer, real buttons for real actions, and a readable label on every control.
- StylingOne shared stylesheet, colours and spacing driven by CSS variables. Layout with grid and flexbox. No framework.
- BehaviourVanilla JavaScript split into ES modules by feature: cover, camera room, workspace and printing each live in their own file.
- 3D camerasThe cameras and film rolls are glTF models rendered with Three.js. Both the library and the models live in the repository, so nothing depends on a CDN.
- Photo processingFilm looks are drawn on a canvas: the photograph first, then a graded layer computed from CSS filters and colour matrices. Compare stacks the two canvases and cuts them with a divider.
- DeploymentThe code lives on GitHub. Every push to main triggers a GitHub Actions workflow that publishes the dist folder to GitHub Pages. No build step, no server, no database.
Where the photographs come from
The sample photographs come from Unsplash under the Unsplash License, and every photographer is credited in the ⓘ panel on the cover. A photograph you import yourself is never uploaded and never leaves the browser.
What I would change next
First, the workspace only grades colour right now; there is no real grain or sharpness model yet, and I would calibrate against actual scanned film. Second, a film look only lives for the current session, so it should be saved and shareable. Third, printing only exports a JPG; connecting it to a real print service would close the loop.
About me
I am Qiang Xu, a computer science student at Hainan Bielefeld University of Applied Sciences (BiUH). This site is my personal project for the Web Technologies course. The code is on GitHub, history included.