A Surviving-Observer Model of Computation
April 17, 2019
This note uses the phrase "surviving observer" for a practical constraint on declarative computation: the model has to connect to an observer, example, user, or physical situation that makes the result testable.
A formal model can be elegant and still fail to carry useful meaning if there is no concrete boundary where its outputs matter.
Computation-model noteOriginally posted 2019-04-17; expanded here around concrete observers and testable examples.
A declarative model describes what should be true. That description still needs a boundary where results are observed, checked, and used.
Without concrete examples, the model can drift into symbols that are internally arranged but externally untested. The "surviving observer" language points to the missing boundary: what observes the result, what state changes, and what constraint decides whether the computation mattered.
Why Examples Matter
Concrete examples are not decoration. They force a model to choose inputs, outputs, state, update rules, and failure cases. They also expose when a clean theory is not yet a usable computation model.
The useful version of this note is simple: a programming model should be tested against the situations it claims to describe.