The app shelf as a data file
All 255 readings are in one CSV file, one row per app, field and item, with the listing address and the day read. It is released under CC BY-SA 4.0. As of 2026-09-25.
| Column | Holds |
|---|---|
| fact-id | App, field and a number |
| value | The reading in a sentence |
| applies-to | The section of the listing it comes from |
| source-url | The listing address |
| source-name | App and store |
| value-since | The day the value was first recorded |
| verified | The day the listing was last checked |
Inclusion rule. Every column in the file. Order. File order.
1Using the file
Filter by field to see every app's age rating or tracking list together; filter by app to see one listing. A missing Android version is a row too, with the file's empty wording and the address of the search that found nothing.
Store listings change without notice: a new version can bring a new rating, new privacy answers or new languages. Each row describes the listing on the day read.
Google Play readings sit in the same file. Their rows carry the Play listing address, and the value names the developer as Google Play shows it, which can differ in spelling from the App Store seller.
Content notes under an age rating are kept in one value, grouped the way the listing groups them: infrequent, frequent and contains. Splitting them into separate rows would lose which group each note belongs to.
Rows are in the order of the app pages, one app after another.
The seller reading's value is the legal name the store shows. Where a maker sells through a company with another name, that name is what appears, and the app page explains which brand it belongs to.
Download app-store-listings.csv
2Aisles
3Making the drama, wherever it runs
Away from the store listing in “Data”, SceneMixer runs in a browser and describes its pipeline like this:
SceneMixer writes dialogue into the script analysis, assigns each line to a shot with a speaker and a short delivery note, and the video model performs it — picture and audio generated together, so the mouth forms that sentence in the language you chose.
SceneMixer, AI Dialogue Generator (read 2026-10-13)
Typically 60–120 seconds, always 9:16 vertical, ending on a hook. SceneMixer's episode outlines are generated with that pacing, and setting a project to 9:16 makes every segment vertical (the default is 16:9).
SceneMixer, How to Make ReelShort-Style Dramas with AI — “How long is a ReelShort-style episode?” (read 2026-10-13)
The quote column. Each reading in the file also carries the words on its source page that it was read from, so any row can be checked against that page. Rows recorded as not stated on the pages read leave it empty.