How we label evidence

Our session library records what practitioners presented and adds editorial interpretation. A published claim is not automatically an independently verified result.

Speaker statement

A speaker's account, judgement or reported result. Attribute it to the speaker and the session date. A backtest described in a talk remains a reported backtest; it does not establish future investment performance.

Demonstrated behaviour

A workflow actually shown in the session. A presenter's description of a capability remains a speaker statement unless the supporting record shows it operating. Its inputs, conditions and limitations matter. A demo does not establish independent reproducibility, general accuracy or current product availability.

Editorial synthesis

Our interpretation of the session: summaries, takeaways and cross-session decision briefs. Use the dated source links to inspect the underlying statements. Interpretive conclusions and unresolved questions should remain distinguishable from measured results.

Independent verification

A result checked against identified evidence by someone independent of the original claim. We do not apply this label to a session merely because its transcript or summary is published. The current session library does not establish independent validation of vendor performance.

Event-report verification

Registration and programme records establish registration and programme facts within their stated coverage. Registrations do not prove attendance; attendance does not prove useful follow-up, commercial outcomes or technical claims. Unknown outcomes stay unknown.

Product descriptions reflect the session date. Follow the source record and seek current evidence before relying on a capability.