Observation#
This page is generated from the code and must not be edited by hand.
Something read from storage. Immutable, and never revised.
An observation’s only claim is that a named observer, at a named version, read this from this location at this time. A later scan produces a *new* observation and both are kept, because the difference between two readings is itself evidence.
Carries extraction_tier, which measures distance from the bytes. It never carries an inference tier: an observation is not a conclusion, and the two axes are kept apart by keeping the models apart.
Fields#
Field |
Type |
Required |
If left out |
|---|---|---|---|
|
text |
yes |
|
|
|
no |
|
|
text or null |
no |
|
|
text or null |
no |
|
|
|
no |
|
|
text |
yes |
|
|
list of |
no |
|
|
text |
yes |
|
|
text |
yes |
|
|
text |
yes |
|
|
text |
yes |
|
|
text |
yes |
|
|
text |
no |
|
|
|
no |
|
|
yes or no or null |
no |
|
|
whole number or null |
no |
|
|
|
no |
|
|
number or null |
no |
|
|
text or null |
no |
|
|
text or null |
no |
|
The types named above are described on the shared types page: ExtractionTier, IdentityScope, LocatorStep, SourceScope, ValueKind.