# Evidence and Reliability Guide
Owlery is a practical sensemaking garden, not a database of universal best practices.
Its reliability comes from making claims traceable, bounded, and usable.
The key questions are:
- Where does this idea come from?
- What kind of claim is it?
- How confidently should it be used?
- What would be an overreach?
- How should it be tested in context?
## Reliability does not mean certainty
Many Owlery notes deal with complex social and organisational situations. In those situations, certainty is often false confidence.
A reliable Owlery note should help people see more clearly, act more responsibly, and learn faster. It should not pretend to remove judgement.
## Use the claim type
Each important note should be understood through [[Claim Types in Owlery]].
A source-grounded concept, a practice-grounded pattern, a tool, and a teaching case each have different kinds of legitimacy.
## Confidence levels
Use confidence levels to show how firmly to hold a claim.
- High confidence: well-supported by source lineage, repeated practice, or strong conceptual grounding.
- Medium confidence: useful and coherent, but context matters and evidence may be mixed or indirect.
- Low confidence: plausible, experimental, or early-stage. Use as a hypothesis.
## Source lineage
Important concepts should show their lineage.
A good lineage block names:
- primary sources or traditions
- Owlery synthesis built from them
- connected notes and tools
- what the concept should not be used to claim
See [[Source Lineage Standards]].
## Use with care
Central concepts should include a use-with-care section when misuse is likely.
This is especially important for psychological safety, complexity, motivation, social loafing, accountability, governance, system debt, metrics, and cognitive load.
## Teaching cases are not proof
A field note can make a pattern vivid. It should not be treated as evidence that the pattern is generally true.
Stories are useful for recognition. They are not substitutes for source lineage or observation.
## Tools are tested in use
Owlery tools should be treated as safe-to-try practices. Their value is judged by whether they improve observation, decision quality, participation, learning, or flow.
## See Also
- [[Claim Types in Owlery]]
- [[Source Lineage Standards]]
- [[Core Claims Register]]
- [[Using Owlery Responsibly]]
- [[Principles]]
## Counter-indicators
Reliability also means knowing when not to use a frame.
For central concepts and tools, add counter-indicators: conditions where the concept may be wrong, premature, unsafe, misleading, or too costly.
See [[Counter Indicators and Responsible Use]].
## Applied claim-status pass
A first targeted application pass has added claim status, source lineage, Owlery synthesis, and use-with-care sections to selected load-bearing concepts. See [[Claim Status Source Lineage Application - 2026-05-27]].