These are the questions the lab works on. Each one says what we believe now. Its page lists the tests of that question on our public record. A test is on the public record when we set it down before it ran and it used only novels from before 1931 or our own pages.
Which structures under the surface make readers love a story? We look at rules, at the breaking of rules and at the forces that pull characters back.
We have not found them yet. Measures of style and surface did not separate the books readers love. Our reading of a novel must get better before our writing can.
0 tests · 0 lessons
How the belief got here →Can the system read a novel at every level well enough to build on?
Yes for the parts we use. Scenes now have a closed list of functions and the library is searchable. None of this predicts which books readers love yet.
8 tests · 9 lessons
How the belief got here →Can we plan a story from the shapes of human novels and keep it consistent as it grows?
Plans built from the library now feed the writer. We do not yet know if they make better stories.
0 tests · 0 lessons
How the belief got here →What makes our pages read like a person wrote them?
Our model's own sentences are the clearest sign of a machine. Only human text with small edits has read as human so far.
17 tests · 25 lessons
How the belief got here →Which checks and judges are right about our pages and how do we know?
The best judge we have reports but does not decide. Listeners on a platform are the only true test.
0 tests · 0 lessons
How the belief got here →Can we run this work as a lab whose public record shows the system getting better?
This page is the start. Each test is set before it runs and each result is kept.
0 tests · 0 lessons
How the belief got here →What do listeners tell us about our published episodes?
We have no listener numbers yet.
0 tests · 0 lessons
How the belief got here →