An enlightenment learning engine: the next bit of information you get is chosen to resolve your biggest current uncertainty, from a model of where you stand, and it remembers you.
Ask a question. Unveil works out the crux behind it, asks where you stand, answers to move that, and checks where you landed.
A project is one core uncertainty plus hypotheses. Each visit ranks what to test or read next by how much uncertainty it would resolve.
A few new things each day, chosen to move your open cruxes; it grows with what you read, skip and vote.
Objective: uncertainty resolution, not surprise. Memory: per person, across days. Benchmark: how well the model predicts your next question and your own stated update. See Benchmark.
Unveil, an enlightenment learning engine. MVP. Spec on the hub; code in meno-sh/Unveil.