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
Tidal stream diamond table: Extrapolated rate can go negative below neaps · case 01
On very small ranges the stream rate becomes negative.
Tidal stream diamond table: Extrapolated rate can go negative below neaps · case 02
On very small ranges the stream rate becomes negative.
Tidal stream diamond table: Extrapolated rate can go negative below neaps · case 03
On very small ranges the stream rate becomes negative.
Tidal stream diamond table: Extrapolated rate can go negative below neaps · case 04
On very small ranges the stream rate becomes negative.
Tidal stream diamond table: Extrapolated rate can go negative below neaps · case 05
On very small ranges the stream rate becomes negative.
Tidal stream diamond table: The half-hour margins of the table are rejected · case 01
Between HW+6 and HW+6.5 no stream is reported although a row applies.
Tidal stream diamond table: The half-hour margins of the table are rejected · case 02
Between HW+6 and HW+6.5 no stream is reported although a row applies.
Tidal stream diamond table: The half-hour margins of the table are rejected · case 03
Between HW+6 and HW+6.5 no stream is reported although a row applies.
Tidal stream diamond table: The half-hour margins of the table are rejected · case 04
Between HW+6 and HW+6.5 no stream is reported although a row applies.
Tidal stream diamond table: The half-hour margins of the table are rejected · case 05
Between HW+6 and HW+6.5 no stream is reported although a row applies.
Tidal stream diamond table: Tabulated tenths are reported as knots · case 01
Stream rates are ten times too large.
Tidal stream diamond table: Tabulated tenths are reported as knots · case 02
Stream rates are ten times too large.
Tidal stream diamond table: Tabulated tenths are reported as knots · case 03
Stream rates are ten times too large.
Tidal stream diamond table: Tabulated tenths are reported as knots · case 04
Stream rates are ten times too large.
Tidal stream diamond table: Tabulated tenths are reported as knots · case 05
Stream rates are ten times too large.
Course to steer across a stream: Heading correction steers with the stream · case 01
The vessel is set twice as far off track.
Course to steer across a stream: Heading correction steers with the stream · case 02
The vessel is set twice as far off track.
Course to steer across a stream: Heading correction steers with the stream · case 03
The vessel is set twice as far off track.
Course to steer across a stream: Heading correction steers with the stream · case 04
The vessel is set twice as far off track.
Course to steer across a stream: Heading correction steers with the stream · case 05
The vessel is set twice as far off track.
Course to steer across a stream: Headings near north are not wrapped · case 01
Courses left of north come out negative or above 360.
Course to steer across a stream: Headings near north are not wrapped · case 02
Courses left of north come out negative or above 360.
Course to steer across a stream: Headings near north are not wrapped · case 03
Courses left of north come out negative or above 360.
Course to steer across a stream: Headings near north are not wrapped · case 04
Courses left of north come out negative or above 360.
Course to steer across a stream: Headings near north are not wrapped · case 05
Courses left of north come out negative or above 360.
Course to steer across a stream: Stream set is read as the direction it comes from · case 01
The boat is steered downstream and fair streams look foul.
Course to steer across a stream: Stream set is read as the direction it comes from · case 02
The boat is steered downstream and fair streams look foul.
Course to steer across a stream: Stream set is read as the direction it comes from · case 03
The boat is steered downstream and fair streams look foul.
Course to steer across a stream: Stream set is read as the direction it comes from · case 04
The boat is steered downstream and fair streams look foul.
Course to steer across a stream: Stream set is read as the direction it comes from · case 05
The boat is steered downstream and fair streams look foul.
Course to steer across a stream: Speed made good ignores the along-track stream · case 01
Fair and foul streams give the same ETA.
Course to steer across a stream: Speed made good ignores the along-track stream · case 02
Fair and foul streams give the same ETA.
Course to steer across a stream: Speed made good ignores the along-track stream · case 03
Fair and foul streams give the same ETA.
Course to steer across a stream: Speed made good ignores the along-track stream · case 04
Fair and foul streams give the same ETA.
Course to steer across a stream: Speed made good ignores the along-track stream · case 05
Fair and foul streams give the same ETA.
Course to steer across a stream: Feasibility compares the whole drift with boat speed · case 01
A strong but fair stream is declared impossible to counter.
Course to steer across a stream: Feasibility compares the whole drift with boat speed · case 02
A strong but fair stream is declared impossible to counter.
Course to steer across a stream: Feasibility compares the whole drift with boat speed · case 03
A strong but fair stream is declared impossible to counter.
Course to steer across a stream: Feasibility compares the whole drift with boat speed · case 04
A strong but fair stream is declared impossible to counter.
Course to steer across a stream: Feasibility compares the whole drift with boat speed · case 05
A strong but fair stream is declared impossible to counter.
High and low water extraction: Stand of the tide is timed at its first sample · case 01
Flat high waters are reported at the start of the stand.
High and low water extraction: Stand of the tide is timed at its first sample · case 02
Flat high waters are reported at the start of the stand.
High and low water extraction: Stand of the tide is timed at its first sample · case 03
Flat high waters are reported at the start of the stand.
High and low water extraction: Stand of the tide is timed at its first sample · case 04
Flat high waters are reported at the start of the stand.
High and low water extraction: Stand of the tide is timed at its first sample · case 05
Flat high waters are reported at the start of the stand.
High and low water extraction: The first sample is compared with the last · case 01
A spurious turning point appears at the start of the prediction window.
High and low water extraction: The first sample is compared with the last · case 02
A spurious turning point appears at the start of the prediction window.
High and low water extraction: The first sample is compared with the last · case 03
A spurious turning point appears at the start of the prediction window.
High and low water extraction: The first sample is compared with the last · case 04
A spurious turning point appears at the start of the prediction window.
High and low water extraction: The first sample is compared with the last · case 05
A spurious turning point appears at the start of the prediction window.
High and low water extraction: Ripples exactly at the threshold are discarded · case 01
A genuine event whose height difference equals minr disappears.
High and low water extraction: Ripples exactly at the threshold are discarded · case 02
A genuine event whose height difference equals minr disappears.
High and low water extraction: Ripples exactly at the threshold are discarded · case 03
A genuine event whose height difference equals minr disappears.
High and low water extraction: Ripples exactly at the threshold are discarded · case 04
A genuine event whose height difference equals minr disappears.
High and low water extraction: Ripples exactly at the threshold are discarded · case 05
A genuine event whose height difference equals minr disappears.
High and low water extraction: Double high water keeps the later peak regardless of height · case 01
The lower of two high waters is reported.
High and low water extraction: Double high water keeps the later peak regardless of height · case 02
The lower of two high waters is reported.
High and low water extraction: Double high water keeps the later peak regardless of height · case 03
The lower of two high waters is reported.
High and low water extraction: Double high water keeps the later peak regardless of height · case 04
The lower of two high waters is reported.
High and low water extraction: Double high water keeps the later peak regardless of height · case 05
The lower of two high waters is reported.
High and low water extraction: Plateau midpoint is floored to a whole sample · case 01
Even-length stands are timed half an interval early.
High and low water extraction: Plateau midpoint is floored to a whole sample · case 02
Even-length stands are timed half an interval early.
High and low water extraction: Plateau midpoint is floored to a whole sample · case 03
Even-length stands are timed half an interval early.
High and low water extraction: Plateau midpoint is floored to a whole sample · case 04
Even-length stands are timed half an interval early.
High and low water extraction: Plateau midpoint is floored to a whole sample · case 05
Even-length stands are timed half an interval early.
Tidal height window: Crossing time is interpolated from the wrong end · case 01
Windows open and close at mirrored times within the sample interval.
Tidal height window: Crossing time is interpolated from the wrong end · case 02
Windows open and close at mirrored times within the sample interval.
Tidal height window: Crossing time is interpolated from the wrong end · case 03
Windows open and close at mirrored times within the sample interval.
Tidal height window: Crossing time is interpolated from the wrong end · case 04
Windows open and close at mirrored times within the sample interval.
Tidal height window: Crossing time is interpolated from the wrong end · case 05
Windows open and close at mirrored times within the sample interval.
Tidal height window: Window opens at the next sample instead of the crossing · case 01
Windows open late by up to one sample interval.
Tidal height window: Window opens at the next sample instead of the crossing · case 02
Windows open late by up to one sample interval.
Tidal height window: Window opens at the next sample instead of the crossing · case 03
Windows open late by up to one sample interval.
Tidal height window: Window opens at the next sample instead of the crossing · case 04
Windows open late by up to one sample interval.
Tidal height window: Window opens at the next sample instead of the crossing · case 05
Windows open late by up to one sample interval.
Tidal height window: A window still open at the end of the series is dropped · case 01
The last opportunity of the day is missing from the plan.
Tidal height window: A window still open at the end of the series is dropped · case 02
The last opportunity of the day is missing from the plan.
Tidal height window: A window still open at the end of the series is dropped · case 03
The last opportunity of the day is missing from the plan.
Tidal height window: A window still open at the end of the series is dropped · case 04
The last opportunity of the day is missing from the plan.
Tidal height window: A window still open at the end of the series is dropped · case 05
The last opportunity of the day is missing from the plan.
Tidal height window: A window already open at the first sample is lost · case 01
When the tide is already high enough at the start, the first window is omitted or truncated.
Tidal height window: A window already open at the first sample is lost · case 02
When the tide is already high enough at the start, the first window is omitted or truncated.
Tidal height window: A window already open at the first sample is lost · case 03
When the tide is already high enough at the start, the first window is omitted or truncated.
Tidal height window: A window already open at the first sample is lost · case 04
When the tide is already high enough at the start, the first window is omitted or truncated.
Tidal height window: A window already open at the first sample is lost · case 05
When the tide is already high enough at the start, the first window is omitted or truncated.
Tidal height window: Window edges are rounded to the nearest minute · case 01
Published windows can open before the water is deep enough.
Tidal height window: Window edges are rounded to the nearest minute · case 02
Published windows can open before the water is deep enough.
Tidal height window: Window edges are rounded to the nearest minute · case 03
Published windows can open before the water is deep enough.
Tidal height window: Window edges are rounded to the nearest minute · case 04
Published windows can open before the water is deep enough.
Tidal height window: Window edges are rounded to the nearest minute · case 05
Published windows can open before the water is deep enough.
Vertical datum conversion: Depths are shifted like heights · case 01
Soundings referred to a higher datum get deeper instead of shallower.
Vertical datum conversion: Depths are shifted like heights · case 02
Soundings referred to a higher datum get deeper instead of shallower.
Vertical datum conversion: Depths are shifted like heights · case 03
Soundings referred to a higher datum get deeper instead of shallower.
Vertical datum conversion: Depths are shifted like heights · case 04
Soundings referred to a higher datum get deeper instead of shallower.
Vertical datum conversion: Depths are shifted like heights · case 05
Soundings referred to a higher datum get deeper instead of shallower.
Vertical datum conversion: Datum offsets are subtracted in reverse order · case 01
Every conversion moves values the wrong way by the offset difference.
Vertical datum conversion: Datum offsets are subtracted in reverse order · case 02
Every conversion moves values the wrong way by the offset difference.
Vertical datum conversion: Datum offsets are subtracted in reverse order · case 03
Every conversion moves values the wrong way by the offset difference.
Vertical datum conversion: Datum offsets are subtracted in reverse order · case 04
Every conversion moves values the wrong way by the offset difference.
Vertical datum conversion: Datum offsets are subtracted in reverse order · case 05
Every conversion moves values the wrong way by the offset difference.
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 ↗