Skip to content

Read an evidence record

Last updated View as MarkdownAgent index

Create a short evidence summary from three fictional records. This exercise teaches the product's evidence model without requiring an account, weather downloads, or a running app. The records are teaching examples, not observations from a real storm or payloads for a deployed API.

Classify the records

Use this fictional set:

Record Content
A A radar-derived estimate belongs to event revision 1
B An external ground report arrives and is included in revision 2
C A sample property intersects the estimated event footprint in revision 2
  1. Write a classification beside each record. Use the evidence labels. Keep the source observation separate from a conclusion derived from it.

  2. Compare your answer with the expected interpretation.

    Record Interpretation
    A radar-derived; the estimate is not a field inspection
    B report-observed; a report does not confirm conditions at every nearby property
    C A modeled exposure relationship; it does not confirm physical damage
  3. Add an uncertainty statement to your summary. Use "confidence not supplied" for this exercise. Do not invent a numerical probability or promote a radar-derived record to field-confirmed because it appears convincing.

Preserve the revision

  1. Record that an earlier decision used revision 1. Record the new report separately under revision 2; do not rewrite what was available earlier.

  2. List the provenance you would need to verify the summary: source identity, observation and acquisition times, source version, event revision, and the derivation method. Mark values absent from the exercise as unknown.

  3. Write the final summary. It should communicate the following result:

    Revision 1 contained radar-derived evidence. Revision 2 added an external report. The sample property intersects estimated exposure in revision 2; physical damage remains unconfirmed. Source details and confidence were not supplied in this exercise.

You now have a summary that preserves evidence type, uncertainty, and changing knowledge. Compare it with Architecture concepts to understand how the same distinctions carry into action plans and outcomes.