Tensions in Practice: A precise corrected reading in tension with preserving the target state¶
A slider with two invented measurement procedures
The slider begins at position 10. A contact gauge shifts it by a known two units before reading 8 exactly; adding two recovers the initial position, but leaves the slider at 8. A noncontact proxy reads 10 with an error bound of one unit and leaves the slider unchanged in this toy model. It preserves the next state but only locates the initial position within [9, 11].
Recover a precise initial value
Use a known interaction correction to infer the initial position.
Preserve the state for later use
Avoid the modeled displacement at the cost of a wider inference interval.
Why these aims pull against each other
Correcting a reading for a known disturbance can recover the earlier value without undoing the disturbance itself. Accurate inference and preserving the future state are different obligations.
Choose an arrangement to see what changes and what remains difficult.
All positions use the same arbitrary unit. Initial estimate is what the procedure infers about the earlier state; Next state is the position left for later use. The bracketed pair is an inclusive uncertainty interval.
What this choice protects
What it costs
When it fits
Compare the arrangements
Correct a contact reading
The stipulated gauge first subtracts 2 from position, then reads the resulting position without error. The inference adds 2 back.
| Reading | Initial estimate | Next state | |
|---|---|---|---|
| Slider | 8 | 10 | 8 |
- What it protects
- The initial value is recovered exactly under the declared displacement rule.
- What it costs
- The physical state remains displaced by 2; later use receives position 8 unless a separate restoration occurs.
- When it fits
- Fits a procedure that can tolerate the displaced state and has a trusted interaction correction.
Illustration note: The correction is in the inferred value, not a restoration action. Unknown displacement would remove the claimed exactness.
Use an indirect proxy
The stipulated proxy causes no position change in this model. Its observed reading is 10 and its error is bounded by ±1.
| Reading | Initial estimate | Next state | |
|---|---|---|---|
| Slider | 10 | [9, 11] | 10 |
- What it protects
- Later use receives the original state, position 10.
- What it costs
- The inferred initial value is an interval rather than a point; the proxy error bound must be justified.
- When it fits
- Fits later use that needs the state preserved and can tolerate this inference uncertainty.
Illustration note: Zero displacement is a declared idealization here, not a universal physical claim about noncontact sensing or a guarantee that indirect methods are always less accurate.
What this illustration does—and does not—establish
The source supplies the structural tension. This bounded example makes a particular relation inspectable; the aims, conditions and residual costs are part of the comparison.
- Positions and apparatus rules are invented. They illustrate modeled interaction, not empirical instrument specifications.
- The known actual initial position is disclosed to the reader; the procedure reports only what its own reading and assumptions warrant.
- This is not an irreducible precision–disturbance law. A different instrument could change either cost, and a separate restoration would have its own requirements.
Source entries
Measurement
The canonical tension motivates this comparison. The setting and arrangements are declared editorial illustrations, not observed findings.
Measurement as Intervention (Bidirectional Coupling)
Every instrument interacts with its target, so measurement is in part an intervention; the question is whether the disturbance is negligible at the precision of interest, not whether it is zero.
The source operation
A second structural fact is that every measurement is in part an *intervention*: the instrument interacts with the target, and the interaction is part of the phenomenon.