An index of jazz standards ordered by how likely a tune is to be called, with every figure in it recountable from files anyone can download.
Most repertoire lists are assembled from taste and presented as fact. This one states its sources, prints what it declined as well as what it kept, and closes its arithmetic by two independent routes. Where a judgment was made, the judgment is named as one.
Two things here are free and open. Neither requires the book.
Order a set so no two tunes in a row are the same kind of tune. Six weighted axes scored at every joint, caps on blues, bossa, groove and ballad, and a control for tunes a singer can front.
Build a set → the evidenceEvery data file, the screening prompt with its checksum, and a verification script that recomputes ninety-four printed figures and tells you if one is wrong.
Open the repository →Clone the repository and run one command. The script reads nothing outside the repository and needs only Python 3.
python3 code/verify.py
It prints one line per check and exits zero when all of them pass. Twelve of the checks rebuild a join from the source files rather than read a stored column, because a column cannot confirm itself. A second command verifies that the screening prompt printed in the book is the same file, byte for byte, as the one archived under a DOI.
None of this asks you to agree with a single seating decision.
The volume states the procedure for disagreeing with it: declare your own filter, run the same screen, print your own residues, and publish the difference. A divergence traceable to a declared filter is not a failure of either run. It is the object the book says it wants and cannot produce alone.
The volume is in preparation. The data files and the verification script are published under
CC BY 4.0 and the MIT License respectively; the transcribed source lists remain the property
of their compilers and are credited in the repository.
© 2026 Geoffrey Jacob Caputo