This task involves the work of defining the requirements that will come to shape/define the controlled experiment of the mobile web section editing dead-end controlled experiment (T409112)
Open questions
- What (if any) requirements beyond those listed in the === Experiment requirements need to be defined?
- All requirements have been defined as of 3 Dec 2025.
Experiment requirements
- The following new events will need to be added/instrumented:
- All events to be added/instrumented are documented in the Instrumentation Spec
- Event name proposal for the new “Edit full page” button is documented in T410319#11402693
- The experiment will be run on the following wikis:
- The 100 largest Wikipedias, as measured by monthly editors per 2025 wiki comparison data
- Groups of people that will be included in the experiment:
- This experiment analysis will focus on registered users with 100 or fewer cumulative edits and logged-out users.
- The current recommended approach is to make everyone eligible for the experiment and then focus the analysis on logged-out users or registered users who have an edit count of 100 or fewer edits at first exposure.
- Bucketing logic/criteria:
- Bucketing should include both registered and unregistered users at all included Wikipedias [i]. This will require the use of edge-unique as the identifier type. Note: This means enrollment will be per-device.
- All users, who are editing a mobile web main namespace page (NS:0) using the visual editor, at any of the Wikipedias should have a 50% chance of being included/bucketed into the A/B test's control or treatment group.
- Note: people who initiate a full-page edit using mobile web in the main namespace page (NS:0) using the visual editor will also be included in the experiment. Although, their behavior will be excluded from the metrics we will compute to evaluate the impact of the intervention.
- We're including this additional data so that we have the option of comparing section editing to full-page editing.
- The treatment group should have the new Mobile section editing button enabled when a user opens the section editor. The control group will not and have the default section editing experience.
- Each device should remain in the same test group for the duration of the test (and across sessions and pages). Note: If the cookie is cleared manually or automatically – by, for example, a privacy enhanced browser – the client is issued a new wmf-uniq cookie, and they may or may not end up enrolled or even assigned to the same group.
i. The 100 largest Wikipedias, as measured by monthly editors per 2025 wiki comparison data.
Test instructions
Console command to force experiment:
mw.xLab.overrideExperimentGroup('fy25-26-we-1-1-19-mobile-section-dead-end', 'treatment');Done
- Answers to all "Open questions" are documented
- All Experiment requirements are defined and documented
- Update https://mpic.wikimedia.org/create-experiment with name of experiment and configuration values (≥24 hours before start of test)