The adaptation register as a data file
All three registers, 167 rows in total, are in one CSV. Ids starting law- are jurisdictions, lib- are libraries, and the rest are AI tools; every row carries its source and reading day, under CC BY 4.0. As of 2026-09-25.
| Column | Holds |
|---|---|
| fact_id | Register, subject, clause and a counter |
| value | The reading in a sentence |
| applies_to | The part of the source it comes from |
| source_url | The source |
| source_name | Publisher and document |
| value_since | The day read |
| verified | The day read |
Inclusion rule. Every column in the file. Order. File order.
1Working with the rows
An id such as law-ca-term is Canada's general term; lib-gutenberg-caveat is Project Gutenberg's note for readers outside the U.S.; runway-training is Runway's clause on training. A trailing number marks a second reading.
Clauses a tool's terms do not state are rows with an empty value, so they can be counted.
Quotes are copied as they appear. Some are cut mid-sentence to keep the clause itself and drop surrounding cross-references.
Terms of service change without notice. Treat each row as a pointer to the source on the day read, and read the current version before relying on it.
Statutes change more slowly, but the U.S. public-domain line moves every January 1, and the row for it carries the year it was read in.
Rows for the jurisdictions come from statutes and government guidance, rows for the libraries from their own policy pages, and rows for the tools from their terms of service; the source_name column says which.
The applies_to column says which part of a source a row comes from, such as a named section of the terms, so a reader can find the sentence again after a page is rearranged.
Download book-adaptation-rules.csv
2Registers
3Turning the book into episodes
Once the rights in “Data” are settled, SceneMixer describes the adaptation itself like this:
That is SceneMixer's core design: characters live in a persistent asset library with locked reference art, and every segment that features a character is generated against those references — same face in episode 1 and episode 50.
SceneMixer, The AI Micro Drama Production Pipeline — “Can AI keep the same actors across 50 episodes?” (read 2026-10-13)
Measured across every job that finished successfully on scenemixer.com in the thirty days to 20 September 2026: analysis of a whole book runs about 2 minutes 49 seconds at the median, one episode's storyboard 1 minute 54 seconds, one video segment 3 minutes 20 seconds, and compositing an episode 3 minutes 52 seconds.
SceneMixer, Turn a Novel into a Video Series with AI — “How long does it take to produce one episode?” (read 2026-10-13)