== Background
Due to a bug in eventlogging (see T175918), we were not able to obtain correct session data for the previous hovercards a/b test on English and German wikipedias
== Pre-deployment checklist:
[x] Ensure data is removed from MySQL
- See T174815#3650853 onwards.
[x] The fix for T175918 is deployed to the group2 wikis.
- See T175918#3645697.
[] Analytics Engineering confirms that the backend for Popups events has been switched to Hadoop from MySQL.
== Acceptance Criteria
[] Launch a/b test on en- and dewiki.
[] Create a task for pausing A/B test exactly two weeks after launch date.
[] Record event on the [[ https://www.mediawiki.org/wiki/Reading/Web/Release_timeline | release timeline ]]
[] As for the previous experiments, record the sampling/bucketing rates at https://meta.wikimedia.org/wiki/Schema_talk:Popups
== Questions
[x] What are the bucket sizes for each group per project?
The bucket sizes that we used in {T172291} resulted in an average overall rate of [[ https://grafana.wikimedia.org/dashboard/db/eventlogging-schema?refresh=5m&orgId=1&var-schema=Popups&from=now-90d&to=now | roughly 200 events/second ]], but 100 events/second were deemed sufficient earlier (cf. T176469#3688649 ). Thus we should cut the bucket sizes in half for the new experiment:
| wiki | on:off:control | `$wgPopupsAnonsExperimentalGroupSize` |
| -- | -- | -- |
| enwiki | ~~0.03:0.03:0.94~~ 0.015:0.15:0.97 | ~~0.06~~ 0.03 |
| dewiki | ~~0.08:0.08:0.84~~ 0.04:0.04:0.92 | ~~0.16~~ 0.08 |