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
Preferred magnitude selection: local distance limit · case 01
Local events almost never report ML.
Preferred magnitude selection: local distance limit · case 02
Local events almost never report ML.
Preferred magnitude selection: local distance limit · case 03
Local events almost never report ML.
Preferred magnitude selection: local distance limit · case 04
Local events almost never report ML.
Preferred magnitude selection: local distance limit · case 05
Local events almost never report ML.
Preferred magnitude selection: zero ML presence · case 01
A microearthquake with ML 0.0 falls through to Md.
Preferred magnitude selection: zero ML presence · case 02
A microearthquake with ML 0.0 falls through to Md.
Preferred magnitude selection: zero ML presence · case 03
A microearthquake with ML 0.0 falls through to Md.
Preferred magnitude selection: zero ML presence · case 04
A microearthquake with ML 0.0 falls through to Md.
Preferred magnitude selection: zero ML presence · case 05
A microearthquake with ML 0.0 falls through to Md.
Preferred magnitude selection: Mw authority first · case 01
Deep events with a moment magnitude publish mb instead.
Preferred magnitude selection: Mw authority first · case 02
Deep events with a moment magnitude publish mb instead.
Preferred magnitude selection: Mw authority first · case 03
Deep events with a moment magnitude publish mb instead.
Preferred magnitude selection: Mw authority first · case 04
Deep events with a moment magnitude publish mb instead.
Preferred magnitude selection: Mw authority first · case 05
Deep events with a moment magnitude publish mb instead.
Omori-Utsu expected aftershock count: p equals one logarithm · case 01
Forecasts for p = 1 sequences are 2.3 times too low.
Omori-Utsu expected aftershock count: p equals one logarithm · case 02
Forecasts for p = 1 sequences are 2.3 times too low.
Omori-Utsu expected aftershock count: p equals one logarithm · case 03
Forecasts for p = 1 sequences are 2.3 times too low.
Omori-Utsu expected aftershock count: p equals one logarithm · case 04
Forecasts for p = 1 sequences are 2.3 times too low.
Omori-Utsu expected aftershock count: p equals one logarithm · case 05
Forecasts for p = 1 sequences are 2.3 times too low.
Omori-Utsu expected aftershock count: integral bound order · case 01
Expected aftershock counts are negative.
Omori-Utsu expected aftershock count: integral bound order · case 02
Expected aftershock counts are negative.
Omori-Utsu expected aftershock count: integral bound order · case 03
Expected aftershock counts are negative.
Omori-Utsu expected aftershock count: integral bound order · case 04
Expected aftershock counts are negative.
Omori-Utsu expected aftershock count: integral bound order · case 05
Expected aftershock counts are negative.
Omori-Utsu expected aftershock count: empty window rejection · case 01
A zero-length window returns a zero forecast instead of None.
Omori-Utsu expected aftershock count: empty window rejection · case 02
A zero-length window returns a zero forecast instead of None.
Omori-Utsu expected aftershock count: empty window rejection · case 03
A zero-length window returns a zero forecast instead of None.
Omori-Utsu expected aftershock count: empty window rejection · case 04
A zero-length window returns a zero forecast instead of None.
Omori-Utsu expected aftershock count: empty window rejection · case 05
A zero-length window returns a zero forecast instead of None.
Omori-Utsu expected aftershock count: c offset in power law · case 01
Early-window forecasts are badly inflated.
Omori-Utsu expected aftershock count: c offset in power law · case 02
Early-window forecasts are badly inflated.
Omori-Utsu expected aftershock count: c offset in power law · case 03
Early-window forecasts are badly inflated.
Omori-Utsu expected aftershock count: c offset in power law · case 04
Early-window forecasts are badly inflated.
Omori-Utsu expected aftershock count: c offset in power law · case 05
Early-window forecasts are badly inflated.
GNSS peak ground displacement magnitude: magnitude inversion · case 01
PGD magnitudes are several units too high and distance-independent.
GNSS peak ground displacement magnitude: magnitude inversion · case 02
PGD magnitudes are several units too high and distance-independent.
GNSS peak ground displacement magnitude: magnitude inversion · case 03
PGD magnitudes are several units too high and distance-independent.
GNSS peak ground displacement magnitude: magnitude inversion · case 04
PGD magnitudes are several units too high and distance-independent.
GNSS peak ground displacement magnitude: magnitude inversion · case 05
PGD magnitudes are several units too high and distance-independent.
GNSS peak ground displacement magnitude: metre to centimetre · case 01
Every station magnitude is about one unit high and small signals pass the noise gate.
GNSS peak ground displacement magnitude: metre to centimetre · case 02
Every station magnitude is about one unit high and small signals pass the noise gate.
GNSS peak ground displacement magnitude: metre to centimetre · case 03
Every station magnitude is about one unit high and small signals pass the noise gate.
GNSS peak ground displacement magnitude: metre to centimetre · case 04
Every station magnitude is about one unit high and small signals pass the noise gate.
GNSS peak ground displacement magnitude: metre to centimetre · case 05
Every station magnitude is about one unit high and small signals pass the noise gate.
GNSS peak ground displacement magnitude: noise floor inclusivity · case 01
Stations at exactly the 2 cm floor are dropped.
GNSS peak ground displacement magnitude: noise floor inclusivity · case 02
Stations at exactly the 2 cm floor are dropped.
GNSS peak ground displacement magnitude: noise floor inclusivity · case 03
Stations at exactly the 2 cm floor are dropped.
GNSS peak ground displacement magnitude: noise floor inclusivity · case 04
Stations at exactly the 2 cm floor are dropped.
GNSS peak ground displacement magnitude: noise floor inclusivity · case 05
Stations at exactly the 2 cm floor are dropped.
GNSS peak ground displacement magnitude: hypocentral distance · case 01
Nearby stations of deep events give inflated magnitudes.
GNSS peak ground displacement magnitude: hypocentral distance · case 02
Nearby stations of deep events give inflated magnitudes.
GNSS peak ground displacement magnitude: hypocentral distance · case 03
Nearby stations of deep events give inflated magnitudes.
GNSS peak ground displacement magnitude: hypocentral distance · case 04
Nearby stations of deep events give inflated magnitudes.
GNSS peak ground displacement magnitude: hypocentral distance · case 05
Nearby stations of deep events give inflated magnitudes.
Estimated Mercalli intensity: Roman numeral index · case 01
Intensity V is labelled VI.
Estimated Mercalli intensity: Roman numeral index · case 02
Intensity V is labelled VI.
Estimated Mercalli intensity: Roman numeral index · case 03
Intensity V is labelled VI.
Estimated Mercalli intensity: Roman numeral index · case 04
Intensity V is labelled VI.
Estimated Mercalli intensity: Roman numeral index · case 05
Intensity V is labelled VI.
Estimated Mercalli intensity: lower intensity clamp · case 01
Distant small events show negative intensities and wrap around the numeral table.
Estimated Mercalli intensity: lower intensity clamp · case 02
Distant small events show negative intensities and wrap around the numeral table.
Estimated Mercalli intensity: lower intensity clamp · case 03
Distant small events show negative intensities and wrap around the numeral table.
Estimated Mercalli intensity: lower intensity clamp · case 04
Distant small events show negative intensities and wrap around the numeral table.
Estimated Mercalli intensity: lower intensity clamp · case 05
Distant small events show negative intensities and wrap around the numeral table.
Estimated Mercalli intensity: intensity rounding · case 01
Intensity 6.7 is reported as VI.
Estimated Mercalli intensity: intensity rounding · case 02
Intensity 6.7 is reported as VI.
Estimated Mercalli intensity: intensity rounding · case 03
Intensity 6.7 is reported as VI.
Estimated Mercalli intensity: intensity rounding · case 04
Intensity 6.7 is reported as VI.
Estimated Mercalli intensity: intensity rounding · case 05
Intensity 6.7 is reported as VI.
Estimated Mercalli intensity: near-field distance floor · case 01
Very shallow events directly under a site produce extreme intensities.
Estimated Mercalli intensity: near-field distance floor · case 02
Very shallow events directly under a site produce extreme intensities.
Estimated Mercalli intensity: near-field distance floor · case 03
Very shallow events directly under a site produce extreme intensities.
Estimated Mercalli intensity: near-field distance floor · case 04
Very shallow events directly under a site produce extreme intensities.
Estimated Mercalli intensity: near-field distance floor · case 05
Very shallow events directly under a site produce extreme intensities.
Brune source radius and stress drop: shear velocity units · case 01
Source radii are a thousand times too small and stress drops astronomically high.
Brune source radius and stress drop: shear velocity units · case 02
Source radii are a thousand times too small and stress drops astronomically high.
Brune source radius and stress drop: shear velocity units · case 03
Source radii are a thousand times too small and stress drops astronomically high.
Brune source radius and stress drop: shear velocity units · case 04
Source radii are a thousand times too small and stress drops astronomically high.
Brune source radius and stress drop: shear velocity units · case 05
Source radii are a thousand times too small and stress drops astronomically high.
Brune source radius and stress drop: angular corner frequency · case 01
Radii are 6.3 times too large and stress drops 250 times too small.
Brune source radius and stress drop: angular corner frequency · case 02
Radii are 6.3 times too large and stress drops 250 times too small.
Brune source radius and stress drop: angular corner frequency · case 03
Radii are 6.3 times too large and stress drops 250 times too small.
Brune source radius and stress drop: angular corner frequency · case 04
Radii are 6.3 times too large and stress drops 250 times too small.
Brune source radius and stress drop: angular corner frequency · case 05
Radii are 6.3 times too large and stress drops 250 times too small.
Brune source radius and stress drop: circular crack constant · case 01
Every stress drop is exactly zero.
Brune source radius and stress drop: circular crack constant · case 02
Every stress drop is exactly zero.
Brune source radius and stress drop: circular crack constant · case 03
Every stress drop is exactly zero.
Brune source radius and stress drop: circular crack constant · case 04
Every stress drop is exactly zero.
Brune source radius and stress drop: circular crack constant · case 05
Every stress drop is exactly zero.
Brune source radius and stress drop: radius cube · case 01
Stress drops carry units of force per length and are orders of magnitude off.
Brune source radius and stress drop: radius cube · case 02
Stress drops carry units of force per length and are orders of magnitude off.
Brune source radius and stress drop: radius cube · case 03
Stress drops carry units of force per length and are orders of magnitude off.
Brune source radius and stress drop: radius cube · case 04
Stress drops carry units of force per length and are orders of magnitude off.
Brune source radius and stress drop: radius cube · case 05
Stress drops carry units of force per length and are orders of magnitude off.
Brune source radius and stress drop: pascal to megapascal · case 01
Stress drops are reported ten times too large.
Brune source radius and stress drop: pascal to megapascal · case 02
Stress drops are reported ten times too large.
Brune source radius and stress drop: pascal to megapascal · case 03
Stress drops are reported ten times too large.
Brune source radius and stress drop: pascal to megapascal · case 04
Stress drops are reported ten times too large.
Brune source radius and stress drop: pascal to megapascal · case 05
Stress drops are reported ten times too large.
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 ↗