Sightlines is an atlas of roughly 1,500 films — a way to find cinema by what a film is actually about, by how it feels, and by the lines of influence that connect it to the films around it. It is a personal project, not a database dump: every film was chosen, written up, and wired into the web by hand and with care.
There are several ways into the collection, all drawing on the same underlying layer of facets, dossiers, and lineage:
Every film also has a junction page: its facets, a scholarly reading of the film (the dossier), and its place in the craft graph — three lenses on the same title.
The heart of the system is a dossier written for every film against a fixed template — industry & production, technology, technique, narrative, genre, authorship, movement, era, themes, and reception. Because each card covers every dimension, facets can be re-extracted as the vocabulary grows without ever rewriting the cards.
The browseable facets come from two layers working together. A rule layer resolves curated facets from signals already attached to each film — TMDB keywords and genres, director, original-title script, and craft markers from the dossiers. On top of that, language-model passes (using Claude) read each film's plot and its full dossier to infer the things keywords miss — that a film is about justice, or addiction, or the atomic age — and to fold thousands of fragmentary keywords into a few dozen coherent concept dials. The lineage graph is built the same way: specific craft debts between films, not vague resemblance.
Under the hood it's plain Python and static HTML/JavaScript, with one small live endpoint for the mood finder and off-collection lineage tracing. Film metadata and posters come from TMDB.
Sightlines was designed, written, and built by Adrian Chan — a service and AI-interaction designer — at gravity7.com, where he also publishes research on AI and language. It's an ongoing experiment in tracing how influence travels through film over time. If it sends you toward a film you'd never have found, it's doing its job.