Observation point
Identify whether the measurement describes the application, gateway or edge. A token can be visible at one point and absent at another.
Understand which experimental records are available and what they can support.
The project reports a 150-request proof of concept and a separate three-request functional check. The records below have different scopes and are not combined into one validated experiment.
| Record | What it addresses | Evidence limit |
|---|---|---|
| Reported 150-request experiment | Selected-action visibility and reconstruction in a controlled HTTP path. | Earlier summaries require reconciliation with the matching logs and analyzer. A stripping problem was reported for this run; it is not evidence of successful stripping across 150 requests. |
| Separate preliminary three-request check | One implicit, one explicit and one strip-mode request. The reported strip observation shows the token at the gateway and absent at the edge. | A small functional check. It does not establish a latency distribution or a general overhead benchmark. |
Identify whether the measurement describes the application, gateway or edge. A token can be visible at one point and absent at another.
State the reconstruction rule and candidate-set assumptions. Four equally likely candidates yield 2 bits under that model; this is not automatically an empirical entropy estimate.
A complete result needs the run-specific logs, analyzer version, request joins, exclusions and mode counts. A consolidated reproduction package is not published here.