Once exposure logging is implemented, we should provide developers with more thorough guidance for logging exposure in their experiments to ensure accuracy and trustworthiness of the data collected.
Notes
For everyone (all user traffic) experiments (that use edge uniques for enrollment) with server-side changes, we can borrow ideas from this approach:
We can provide a recommendation for how to log exposure on the client-side, namely by using server-side logic to conditionally load client-side instrumentation which would include an exposure logging call.
Acceptance criteria
- https://wikitech.wikimedia.org/wiki/Test_Kitchen/Experiment_exposure_logging includes examples/scenarios, do's & don't's, recommended patterns / best practices to help developers:
- maximize true positives (logging exposure when exposure actually took place)
- minimize false positives (logging exposure when there wasn't any exposure) and false negatives (failing to log exposure when exposure should have been logged)
- https://wikitech.wikimedia.org/wiki/Test_Kitchen/Conduct_an_experiment links to https://wikitech.wikimedia.org/wiki/Test_Kitchen/Experiment_exposure_logging