This task involves the work of establishing a baseline understanding of how and who is using the Tools and Appearance menus while using the visual editor.
Learning about how people use these menus will enable us – the Editing and Web Teams – to confirm the following assumption we're holding: People who have published ≤100 cumulative edits will not be disrupted by losing access to the Tools and Appearance menus while editing with the visual editor on desktop.
Research questions
| Research question | Metric(s) |
|---|---|
| Who is/is not using the "Tools" and "Appearance" menus? | 1. What percentage of people elect to hide the Tools and/or Appearance menus? 2. Of the people who published ≥1 unreverted edit in the main namespace, what percentage of them used the Tools and/or Appearance menu while editing using the visual editor (desktop)? 3. Of all the unreverted edits people make in the main namespace using the visual editor, what proportion of edits involved someone using the Tools and/or Appearance menus? |
| How heavily do people depend on the "Tools" and "Appearance" menus while editing? | What percentage of editing sessions in the main namespace that result in an unreverted published edit (broken out by experience level) do people use the Tools and/or Appearance menu within? |
NOTE: each of the metrics will be limited to people who have made ≤100 cumulative edits as they are the people impacted by this change.
Decision(s) to be made
- Will we deviate from current plan to hide the Tools and Appearance menus for people who have the potential to see an Edit Check (https://phabricator.wikimedia.org/T379443)?
Open questions
- 1. What – if any – new instrumentation would need to be added for us to be able to answer the questions above?
- Per T378036#10304363, no additional instrumentation is needed to conduct the analysis scoped in this task.
Background
This investigation is prompted by a desire for two things to be true for volunteers:
- They feel like they have sufficient space (width) to make changes
- They are able to see Edit Checks in close proximity to the content they're related to without any part of the editable content being obfuscated
References
- We recently conducted a similar "feature use" analysis of Citoid in T368988.