T239676 introduces a prompt that invites people who are attempting to open mobile VE, knowingly or not, the source editor when the former takes >2 seconds to load.

This task involves the work of leveraging the instrumentation T243641 implemented to understand when this fallback is being shown and how/if people are engaging with it.

Decision(s) to be made

• 1. What – if any – adjustments will be made to how peoples' editing preferences are/are not impacted by them opting to Switch to source editor?

Research questions

1. How often are individual people encountering the "fallback" prompt T239676 introduced? E.g. Do they see it 1 out of every 2 edit attempts? 1 out of 10? Etc.
2. Of the times the "fallback" prompt is shown, how often are people opting to Switch to the source editor?
3. What proportion of mobile web edits are abandoned in the time between someone taps edit (read: an init event is emitted) and the editing interface is ready for them to use (read: a ready event is emitted)?
   1. Of the edit sessions that result in people abandoning the edit they initiated before the editing interface is ready, what proportion of those "session abandonments" happen at various timing intervals?
   2. How do "3" and "3A" vary between "normal" visual edits and suggested edits?

Done

Reference

• T239676#9201515