The Failed Experiment Nobody Can Learn From.

When only success becomes visible, everyone else can unknowingly pay to repeat the same failure.

The problem

Negative results, abandoned prototypes and failed replications often contain high-value knowledge, but publication and incentive systems favour successful outcomes. The absence of visible failure distorts what future teams think is known.

What breaks

• Publication Bias
• Memory Failure
• Provenance Loss
• Reuse Failure

Who holds part of the picture

• researchers
• funders
• reviewers
• future teams

With Koali

Koali preserves validated negative knowledge as first-class artifacts: method, provenance, uncertainty, why the team stopped, and what a future attempt would need to change.

Human authority

Authorized humans and institutions retain their existing legal, professional and democratic authority. Koali structures knowledge, coordination and traceability; it does not take sovereign or professional decisions.

What success looks like

• The relevant state remains visible across handoffs and time.
• Responsibilities, unresolved questions and escalation paths are explicit.
• The outcome produces reusable memory for the next comparable case.

Real-world parallels

Cochrane review — Publication bias in clinical trials

The review found positive trials were published more often and more quickly than negative or null trials.

Relation to scenario: DIRECT
Source: https://pmc.ncbi.nlm.nih.gov/articles/PMC8276556/

Research Waste in Randomized Clinical Trials

The study describes non-completion and insufficient reporting of randomized trials as research waste.

Relation to scenario: DIRECT
Source: https://pmc.ncbi.nlm.nih.gov/articles/PMC7572905/

Boundary

The documented cases illustrate analogous failure mechanisms. They do not establish that Koali would have prevented the event or guaranteed a different outcome.

Koali loop

signals → shared meaning → evidence → choice → responsibility → action → verification → memory → better knowledge