Understand the failure.
Verify the repair.
Small, reproducible software failures. The broken implementation, the fix that didn’t work, and the one that passed—preserved together.
Explore the cases ↓How results are verified ↗WHAT THE ARCHIVE CONTAINS
100840 executable cases. 20168 are open.
Every case records the implementation that fails, the fix that did not work, and the repair that passed its checks—with recorded outputs and source hashes. This release adds 100840 cases across 20168 failure mechanisms and 254 domains.
The open tier gives you the failure and the unsuccessful fix for one case in every mechanism. The remaining 80672 cases, 5 variants per mechanism, are member-only: the verified repair, its recorded checks, and the full fixture suite are held in the member archive. Read the methodology ↗
A RECORD OF WHAT WENT WRONG
Browse the archive / 100840
Co-orbital phasing manoeuvre: The lead angle is not split across phasing revolutions · case 01
Multi-revolution phasing orbits are far too eccentric.
Co-orbital phasing manoeuvre: The lead angle is not split across phasing revolutions · case 02
Multi-revolution phasing orbits are far too eccentric.
Co-orbital phasing manoeuvre: The lead angle is not split across phasing revolutions · case 03
Multi-revolution phasing orbits are far too eccentric.
Co-orbital phasing manoeuvre: The lead angle is not split across phasing revolutions · case 04
Multi-revolution phasing orbits are far too eccentric.
Co-orbital phasing manoeuvre: The lead angle is not split across phasing revolutions · case 05
Multi-revolution phasing orbits are far too eccentric.
Co-orbital phasing manoeuvre: Only the entry burn is budgeted · case 01
The delta-v budget omits the burn that recircularizes.
Co-orbital phasing manoeuvre: Only the entry burn is budgeted · case 02
The delta-v budget omits the burn that recircularizes.
Co-orbital phasing manoeuvre: Only the entry burn is budgeted · case 03
The delta-v budget omits the burn that recircularizes.
Co-orbital phasing manoeuvre: Only the entry burn is budgeted · case 04
The delta-v budget omits the burn that recircularizes.
Co-orbital phasing manoeuvre: Only the entry burn is budgeted · case 05
The delta-v budget omits the burn that recircularizes.
Co-orbital phasing manoeuvre: Phasing orbit speed uses swapped vis-viva radii · case 01
Burn magnitudes are wrong for every eccentric phasing orbit.
Co-orbital phasing manoeuvre: Phasing orbit speed uses swapped vis-viva radii · case 02
Burn magnitudes are wrong for every eccentric phasing orbit.
Co-orbital phasing manoeuvre: Phasing orbit speed uses swapped vis-viva radii · case 03
Burn magnitudes are wrong for every eccentric phasing orbit.
Co-orbital phasing manoeuvre: Phasing orbit speed uses swapped vis-viva radii · case 04
Burn magnitudes are wrong for every eccentric phasing orbit.
Co-orbital phasing manoeuvre: Phasing orbit speed uses swapped vis-viva radii · case 05
Burn magnitudes are wrong for every eccentric phasing orbit.
Co-orbital phasing manoeuvre: Safety check tests the semi-major axis instead of perigee · case 01
Phasing orbits that dip into the atmosphere are accepted.
Co-orbital phasing manoeuvre: Safety check tests the semi-major axis instead of perigee · case 02
Phasing orbits that dip into the atmosphere are accepted.
Co-orbital phasing manoeuvre: Safety check tests the semi-major axis instead of perigee · case 03
Phasing orbits that dip into the atmosphere are accepted.
Co-orbital phasing manoeuvre: Safety check tests the semi-major axis instead of perigee · case 04
Phasing orbits that dip into the atmosphere are accepted.
Co-orbital phasing manoeuvre: Safety check tests the semi-major axis instead of perigee · case 05
Phasing orbits that dip into the atmosphere are accepted.
Hyperbolic departure from a parking orbit: Periapsis speed uses the circular potential term · case 01
Departure burns are underestimated.
Hyperbolic departure from a parking orbit: Periapsis speed uses the circular potential term · case 02
Departure burns are underestimated.
Hyperbolic departure from a parking orbit: Periapsis speed uses the circular potential term · case 03
Departure burns are underestimated.
Hyperbolic departure from a parking orbit: Periapsis speed uses the circular potential term · case 04
Departure burns are underestimated.
Hyperbolic departure from a parking orbit: Periapsis speed uses the circular potential term · case 05
Departure burns are underestimated.
Hyperbolic departure from a parking orbit: Burn subtracts escape speed instead of circular speed · case 01
Departure burn equals only the excess over escape.
Hyperbolic departure from a parking orbit: Burn subtracts escape speed instead of circular speed · case 02
Departure burn equals only the excess over escape.
Hyperbolic departure from a parking orbit: Burn subtracts escape speed instead of circular speed · case 03
Departure burn equals only the excess over escape.
Hyperbolic departure from a parking orbit: Burn subtracts escape speed instead of circular speed · case 04
Departure burn equals only the excess over escape.
Hyperbolic departure from a parking orbit: Burn subtracts escape speed instead of circular speed · case 05
Departure burn equals only the excess over escape.
Hyperbolic departure from a parking orbit: Departure eccentricity uses unsquared excess speed · case 01
Hyperbola shape is wrong for any vinf other than one.
Hyperbolic departure from a parking orbit: Departure eccentricity uses unsquared excess speed · case 02
Hyperbola shape is wrong for any vinf other than one.
Hyperbolic departure from a parking orbit: Departure eccentricity uses unsquared excess speed · case 03
Hyperbola shape is wrong for any vinf other than one.
Hyperbolic departure from a parking orbit: Departure eccentricity uses unsquared excess speed · case 04
Hyperbola shape is wrong for any vinf other than one.
Hyperbolic departure from a parking orbit: Departure eccentricity uses unsquared excess speed · case 05
Hyperbola shape is wrong for any vinf other than one.
Hyperbolic departure from a parking orbit: Turn angle is half its value · case 01
Flyby bending is underestimated by half.
Hyperbolic departure from a parking orbit: Turn angle is half its value · case 02
Flyby bending is underestimated by half.
Hyperbolic departure from a parking orbit: Turn angle is half its value · case 03
Flyby bending is underestimated by half.
Hyperbolic departure from a parking orbit: Turn angle is half its value · case 04
Flyby bending is underestimated by half.
Hyperbolic departure from a parking orbit: Turn angle is half its value · case 05
Flyby bending is underestimated by half.
Spherical web mercator forward projection: longitude wrap · case 01
Features east of the antimeridian or given in [180, 540) ranges land off the projected world extent.
Spherical web mercator forward projection: longitude wrap · case 02
Features east of the antimeridian or given in [180, 540) ranges land off the projected world extent.
Spherical web mercator forward projection: longitude wrap · case 03
Features east of the antimeridian or given in [180, 540) ranges land off the projected world extent.
Spherical web mercator forward projection: longitude wrap · case 04
Features east of the antimeridian or given in [180, 540) ranges land off the projected world extent.
Spherical web mercator forward projection: longitude wrap · case 05
Features east of the antimeridian or given in [180, 540) ranges land off the projected world extent.
Spherical web mercator forward projection: latitude clamp · case 01
Polar latitudes project far outside the square web mercator extent.
Spherical web mercator forward projection: latitude clamp · case 02
Polar latitudes project far outside the square web mercator extent.
Spherical web mercator forward projection: latitude clamp · case 03
Polar latitudes project far outside the square web mercator extent.
Spherical web mercator forward projection: latitude clamp · case 04
Polar latitudes project far outside the square web mercator extent.
Spherical web mercator forward projection: latitude clamp · case 05
Polar latitudes project far outside the square web mercator extent.
Spherical web mercator forward projection: half-angle term · case 01
Northings are wildly exaggerated even at mid latitudes.
Spherical web mercator forward projection: half-angle term · case 02
Northings are wildly exaggerated even at mid latitudes.
Spherical web mercator forward projection: half-angle term · case 03
Northings are wildly exaggerated even at mid latitudes.
Spherical web mercator forward projection: half-angle term · case 04
Northings are wildly exaggerated even at mid latitudes.
Spherical web mercator forward projection: half-angle term · case 05
Northings are wildly exaggerated even at mid latitudes.
Spherical web mercator forward projection: easting angular unit · case 01
Eastings are 57 times too large.
Spherical web mercator forward projection: easting angular unit · case 02
Eastings are 57 times too large.
Spherical web mercator forward projection: easting angular unit · case 03
Eastings are 57 times too large.
Spherical web mercator forward projection: easting angular unit · case 04
Eastings are 57 times too large.
Spherical web mercator forward projection: easting angular unit · case 05
Eastings are 57 times too large.
Spherical web mercator forward projection: northing sign convention · case 01
The map is mirrored north-south.
Spherical web mercator forward projection: northing sign convention · case 02
The map is mirrored north-south.
Spherical web mercator forward projection: northing sign convention · case 03
The map is mirrored north-south.
Spherical web mercator forward projection: northing sign convention · case 04
The map is mirrored north-south.
Spherical web mercator forward projection: northing sign convention · case 05
The map is mirrored north-south.
Spherical web mercator inverse projection: gudermannian scaling · case 01
Every latitude away from the equator comes back halved.
Spherical web mercator inverse projection: gudermannian scaling · case 02
Every latitude away from the equator comes back halved.
Spherical web mercator inverse projection: gudermannian scaling · case 03
Every latitude away from the equator comes back halved.
Spherical web mercator inverse projection: gudermannian scaling · case 04
Every latitude away from the equator comes back halved.
Spherical web mercator inverse projection: gudermannian scaling · case 05
Every latitude away from the equator comes back halved.
Spherical web mercator inverse projection: exponent sign · case 01
Latitudes come back with the wrong hemisphere.
Spherical web mercator inverse projection: exponent sign · case 02
Latitudes come back with the wrong hemisphere.
Spherical web mercator inverse projection: exponent sign · case 03
Latitudes come back with the wrong hemisphere.
Spherical web mercator inverse projection: exponent sign · case 04
Latitudes come back with the wrong hemisphere.
Spherical web mercator inverse projection: exponent sign · case 05
Latitudes come back with the wrong hemisphere.
Spherical web mercator inverse projection: easting wrap modulus · case 01
Positions from wrapped world copies return longitudes in the western hemisphere only.
Spherical web mercator inverse projection: easting wrap modulus · case 02
Positions from wrapped world copies return longitudes in the western hemisphere only.
Spherical web mercator inverse projection: easting wrap modulus · case 03
Positions from wrapped world copies return longitudes in the western hemisphere only.
Spherical web mercator inverse projection: easting wrap modulus · case 04
Positions from wrapped world copies return longitudes in the western hemisphere only.
Spherical web mercator inverse projection: easting wrap modulus · case 05
Positions from wrapped world copies return longitudes in the western hemisphere only.
Spherical web mercator inverse projection: longitude unit conversion · case 01
Longitudes are returned in radians.
Spherical web mercator inverse projection: longitude unit conversion · case 02
Longitudes are returned in radians.
Spherical web mercator inverse projection: longitude unit conversion · case 03
Longitudes are returned in radians.
Spherical web mercator inverse projection: longitude unit conversion · case 04
Longitudes are returned in radians.
Spherical web mercator inverse projection: longitude unit conversion · case 05
Longitudes are returned in radians.
Spherical web mercator inverse projection: latitude radian conversion · case 01
Latitudes are returned in radians.
Spherical web mercator inverse projection: latitude radian conversion · case 02
Latitudes are returned in radians.
Spherical web mercator inverse projection: latitude radian conversion · case 03
Latitudes are returned in radians.
Spherical web mercator inverse projection: latitude radian conversion · case 04
Latitudes are returned in radians.
Spherical web mercator inverse projection: latitude radian conversion · case 05
Latitudes are returned in radians.
Slippy map tile index from coordinates: tiles per axis · case 01
Tile indices are half of the expected values and zoom 0 produces fractional grid sizes.
Slippy map tile index from coordinates: tiles per axis · case 02
Tile indices are half of the expected values and zoom 0 produces fractional grid sizes.
Slippy map tile index from coordinates: tiles per axis · case 03
Tile indices are half of the expected values and zoom 0 produces fractional grid sizes.
Slippy map tile index from coordinates: tiles per axis · case 04
Tile indices are half of the expected values and zoom 0 produces fractional grid sizes.
Slippy map tile index from coordinates: tiles per axis · case 05
Tile indices are half of the expected values and zoom 0 produces fractional grid sizes.
Slippy map tile index from coordinates: row axis orientation · case 01
Northern cities are fetched from southern tile rows.
Slippy map tile index from coordinates: row axis orientation · case 02
Northern cities are fetched from southern tile rows.
Slippy map tile index from coordinates: row axis orientation · case 03
Northern cities are fetched from southern tile rows.
Slippy map tile index from coordinates: row axis orientation · case 04
Northern cities are fetched from southern tile rows.
Slippy map tile index from coordinates: row axis orientation · case 05
Northern cities are fetched from southern tile rows.
INSPECTABLE BY DESIGN
Every result has a runnable source.
Runnable implementations with recorded outputs, source hashes, and explicit contracts. Related variants share a failure mechanism and belong together in evaluation splits.
Read the methodology ↗