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
Two-group next-generation matrix R0: target-group size weighting · case 01
Small high-contact groups dominate R0 even when few susceptibles are in them.
Two-group next-generation matrix R0: target-group size weighting · case 02
Small high-contact groups dominate R0 even when few susceptibles are in them.
Two-group next-generation matrix R0: target-group size weighting · case 03
Small high-contact groups dominate R0 even when few susceptibles are in them.
Two-group next-generation matrix R0: target-group size weighting · case 04
Small high-contact groups dominate R0 even when few susceptibles are in them.
Two-group next-generation matrix R0: target-group size weighting · case 05
Small high-contact groups dominate R0 even when few susceptibles are in them.
Two-group next-generation matrix R0: infector infectious period · case 01
R0 is wrong whenever the groups recover at different rates.
Two-group next-generation matrix R0: infector infectious period · case 02
R0 is wrong whenever the groups recover at different rates.
Two-group next-generation matrix R0: infector infectious period · case 03
R0 is wrong whenever the groups recover at different rates.
Two-group next-generation matrix R0: infector infectious period · case 04
R0 is wrong whenever the groups recover at different rates.
Two-group next-generation matrix R0: infector infectious period · case 05
R0 is wrong whenever the groups recover at different rates.
Two-group next-generation matrix R0: spectral radius · case 01
Cross-group transmission is ignored when computing R0.
Two-group next-generation matrix R0: spectral radius · case 02
Cross-group transmission is ignored when computing R0.
Two-group next-generation matrix R0: spectral radius · case 03
Cross-group transmission is ignored when computing R0.
Two-group next-generation matrix R0: spectral radius · case 04
Cross-group transmission is ignored when computing R0.
Two-group next-generation matrix R0: spectral radius · case 05
Cross-group transmission is ignored when computing R0.
Two-group next-generation matrix R0: determinant sign · case 01
Strong cross-coupling lowers R0 instead of raising it.
Two-group next-generation matrix R0: determinant sign · case 02
Strong cross-coupling lowers R0 instead of raising it.
Two-group next-generation matrix R0: determinant sign · case 03
Strong cross-coupling lowers R0 instead of raising it.
Two-group next-generation matrix R0: determinant sign · case 04
Strong cross-coupling lowers R0 instead of raising it.
Two-group next-generation matrix R0: determinant sign · case 05
Strong cross-coupling lowers R0 instead of raising it.
Two-group next-generation matrix R0: eigenvector component · case 01
The leading group flips for asymmetric coupling.
Two-group next-generation matrix R0: eigenvector component · case 02
The leading group flips for asymmetric coupling.
Two-group next-generation matrix R0: eigenvector component · case 03
The leading group flips for asymmetric coupling.
Two-group next-generation matrix R0: eigenvector component · case 04
The leading group flips for asymmetric coupling.
Two-group next-generation matrix R0: eigenvector component · case 05
The leading group flips for asymmetric coupling.
Two-group next-generation matrix R0: decoupled tie-break · case 01
Two identical decoupled groups report group 1 as leading.
Two-group next-generation matrix R0: decoupled tie-break · case 02
Two identical decoupled groups report group 1 as leading.
Two-group next-generation matrix R0: decoupled tie-break · case 03
Two identical decoupled groups report group 1 as leading.
Two-group next-generation matrix R0: decoupled tie-break · case 04
Two identical decoupled groups report group 1 as leading.
Two-group next-generation matrix R0: decoupled tie-break · case 05
Two identical decoupled groups report group 1 as leading.
Final epidemic size by bisection: subcritical regime · case 01
Subcritical pathogens report a large positive overshoot.
Final epidemic size by bisection: subcritical regime · case 02
Subcritical pathogens report a large positive overshoot.
Final epidemic size by bisection: subcritical regime · case 03
Subcritical pathogens report a large positive overshoot.
Final epidemic size by bisection: subcritical regime · case 04
Subcritical pathogens report a large positive overshoot.
Final epidemic size by bisection: subcritical regime · case 05
Subcritical pathogens report a large positive overshoot.
Final epidemic size by bisection: root side test · case 01
Bisection converges to the trivial root and reports no epidemic.
Final epidemic size by bisection: root side test · case 02
Bisection converges to the trivial root and reports no epidemic.
Final epidemic size by bisection: root side test · case 03
Bisection converges to the trivial root and reports no epidemic.
Final epidemic size by bisection: root side test · case 04
Bisection converges to the trivial root and reports no epidemic.
Final epidemic size by bisection: root side test · case 05
Bisection converges to the trivial root and reports no epidemic.
Final epidemic size by bisection: upper bracket · case 01
The attack fraction never exceeds the herd-immunity threshold (no overshoot).
Final epidemic size by bisection: upper bracket · case 02
The attack fraction never exceeds the herd-immunity threshold (no overshoot).
Final epidemic size by bisection: upper bracket · case 03
The attack fraction never exceeds the herd-immunity threshold (no overshoot).
Final epidemic size by bisection: upper bracket · case 04
The attack fraction never exceeds the herd-immunity threshold (no overshoot).
Final epidemic size by bisection: upper bracket · case 05
The attack fraction never exceeds the herd-immunity threshold (no overshoot).
Final epidemic size by bisection: convergence tolerance · case 01
Final sizes are only accurate to three decimals regardless of the requested tolerance.
Final epidemic size by bisection: convergence tolerance · case 02
Final sizes are only accurate to three decimals regardless of the requested tolerance.
Final epidemic size by bisection: convergence tolerance · case 03
Final sizes are only accurate to three decimals regardless of the requested tolerance.
Final epidemic size by bisection: convergence tolerance · case 04
Final sizes are only accurate to three decimals regardless of the requested tolerance.
Final epidemic size by bisection: convergence tolerance · case 05
Final sizes are only accurate to three decimals regardless of the requested tolerance.
Final epidemic size by bisection: case count rounding · case 01
Expected case counts are one lower than the nearest integer.
Final epidemic size by bisection: case count rounding · case 02
Expected case counts are one lower than the nearest integer.
Final epidemic size by bisection: case count rounding · case 03
Expected case counts are one lower than the nearest integer.
Final epidemic size by bisection: case count rounding · case 04
Expected case counts are one lower than the nearest integer.
Final epidemic size by bisection: case count rounding · case 05
Expected case counts are one lower than the nearest integer.
Final epidemic size by bisection: overshoot reference · case 01
Overshoot values are large even for mild epidemics.
Final epidemic size by bisection: overshoot reference · case 02
Overshoot values are large even for mild epidemics.
Final epidemic size by bisection: overshoot reference · case 03
Overshoot values are large even for mild epidemics.
Final epidemic size by bisection: overshoot reference · case 04
Overshoot values are large even for mild epidemics.
Final epidemic size by bisection: overshoot reference · case 05
Overshoot values are large even for mild epidemics.
SIR with daily all-or-nothing vaccination campaign: campaign start · case 01
The campaign starts one day late.
SIR with daily all-or-nothing vaccination campaign: campaign start · case 02
The campaign starts one day late.
SIR with daily all-or-nothing vaccination campaign: campaign start · case 03
The campaign starts one day late.
SIR with daily all-or-nothing vaccination campaign: campaign start · case 04
The campaign starts one day late.
SIR with daily all-or-nothing vaccination campaign: campaign start · case 05
The campaign starts one day late.
SIR with daily all-or-nothing vaccination campaign: dose supply cap · case 01
Susceptibles go negative once supply exceeds the remaining pool.
SIR with daily all-or-nothing vaccination campaign: dose supply cap · case 02
Susceptibles go negative once supply exceeds the remaining pool.
SIR with daily all-or-nothing vaccination campaign: dose supply cap · case 03
Susceptibles go negative once supply exceeds the remaining pool.
SIR with daily all-or-nothing vaccination campaign: dose supply cap · case 04
Susceptibles go negative once supply exceeds the remaining pool.
SIR with daily all-or-nothing vaccination campaign: dose supply cap · case 05
Susceptibles go negative once supply exceeds the remaining pool.
SIR with daily all-or-nothing vaccination campaign: efficacy application · case 01
Vaccine failures vanish from the population.
SIR with daily all-or-nothing vaccination campaign: efficacy application · case 02
Vaccine failures vanish from the population.
SIR with daily all-or-nothing vaccination campaign: efficacy application · case 03
Vaccine failures vanish from the population.
SIR with daily all-or-nothing vaccination campaign: efficacy application · case 04
Vaccine failures vanish from the population.
SIR with daily all-or-nothing vaccination campaign: efficacy application · case 05
Vaccine failures vanish from the population.
SIR with daily all-or-nothing vaccination campaign: vaccination ordering · case 01
Same-day vaccinations do not reduce that day's transmission.
SIR with daily all-or-nothing vaccination campaign: vaccination ordering · case 02
Same-day vaccinations do not reduce that day's transmission.
SIR with daily all-or-nothing vaccination campaign: vaccination ordering · case 03
Same-day vaccinations do not reduce that day's transmission.
SIR with daily all-or-nothing vaccination campaign: vaccination ordering · case 04
Same-day vaccinations do not reduce that day's transmission.
SIR with daily all-or-nothing vaccination campaign: vaccination ordering · case 05
Same-day vaccinations do not reduce that day's transmission.
SIR with daily all-or-nothing vaccination campaign: administered dose tally · case 01
Administered doses match protected people rather than shots given.
SIR with daily all-or-nothing vaccination campaign: administered dose tally · case 02
Administered doses match protected people rather than shots given.
SIR with daily all-or-nothing vaccination campaign: administered dose tally · case 03
Administered doses match protected people rather than shots given.
SIR with daily all-or-nothing vaccination campaign: administered dose tally · case 04
Administered doses match protected people rather than shots given.
SIR with daily all-or-nothing vaccination campaign: administered dose tally · case 05
Administered doses match protected people rather than shots given.
SIR with daily all-or-nothing vaccination campaign: mixing denominator · case 01
Vaccination makes remaining susceptibles meet infectives more often.
SIR with daily all-or-nothing vaccination campaign: mixing denominator · case 02
Vaccination makes remaining susceptibles meet infectives more often.
SIR with daily all-or-nothing vaccination campaign: mixing denominator · case 03
Vaccination makes remaining susceptibles meet infectives more often.
SIR with daily all-or-nothing vaccination campaign: mixing denominator · case 04
Vaccination makes remaining susceptibles meet infectives more often.
SIR with daily all-or-nothing vaccination campaign: mixing denominator · case 05
Vaccination makes remaining susceptibles meet infectives more often.
Age-structured force of infection: contact orientation · case 01
Pressure on children uses the contacts adults report making with children.
Age-structured force of infection: contact orientation · case 02
Pressure on children uses the contacts adults report making with children.
Age-structured force of infection: contact orientation · case 03
Pressure on children uses the contacts adults report making with children.
Age-structured force of infection: contact orientation · case 04
Pressure on children uses the contacts adults report making with children.
Age-structured force of infection: contact orientation · case 05
Pressure on children uses the contacts adults report making with children.
Age-structured force of infection: prevalence denominator · case 01
Small age bands receive inflated force of infection from large ones.
Age-structured force of infection: prevalence denominator · case 02
Small age bands receive inflated force of infection from large ones.
Age-structured force of infection: prevalence denominator · case 03
Small age bands receive inflated force of infection from large ones.
Age-structured force of infection: prevalence denominator · case 04
Small age bands receive inflated force of infection from large ones.
Age-structured force of infection: prevalence denominator · case 05
Small age bands receive inflated force of infection from large ones.
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 ↗