Page MenuHomePhabricator

Process research findings from Q2
Closed, ResolvedPublic


Daisy Chen conducted design research for Suggested edits V3 and beyond. The following items have been identified as relevant for V4.

01) Add more tasks (T240196)

Other microcontribution task types (...) Implement as many as possible that interest users.

👍We’re introducing a new task: image tagging.

02) Think about in-article editing (T240195)

7 of 8 participants prefer to make edits while browsing Wikipedia, as opposed to editing from a discrete feed.

👍 Image tagging in V4 can be accomplished via in-article CTA. We’re currently thinking about bundling open tasks for a specific article, but are waiting on analytics data to change the current way of presenting a task in T239706.

03) RTL (T240194)

For translations, make the RTL UI easier to determine/navigate.

👍 RTL improvements will be part of Suggested edits V4.

04) Clearer language (T238840, T231450)

Use clearer language that does not assume previous knowledge (e.g. ‘reverts’).

👍 Revert copy will be revised in V4. In addition, we’re working on an improved, clearer version of the FAQ page.

05) Improve UI in previous versions (T240197)

[Users] don't understand the left and right arrows. other options or other things to thank?

👍 Left and right arrows will be optimized with a standard material design component for going to the next/previous suggestion. Labeling next and previous will also be considered.

06) In-app FAQ page

Additional information screen should be integrated within the app.

👎 We’re aware that leading users out of the app is suboptimal at the moment. We’re keeping it as is due to maintenance reasons and because of the fact the we’re currently experimenting with the contents of the FAQ page.

07) Customizable home page

Allow for the dashboard and motivational app elements (opt-in for the latter) to be as customizable as possible. This way, users can choose to engage more fully with all the available options or get just a small snapshot of exactly what they want to know.

Allow for users to hide task types that aren’t preferred, or auto-sort task types in descending order according to user’s engagement/# of edits.

👎 “Settings vs no settings” is an ongoing discussion within the Android team. As there are currently only 4 different type of profile stats shown, I don’t think they’re too obtrusive. This will definitely be considered once we’re adding more profile stats and/or achievements. We can bundle these motivational elements in a setting and let users either choose to enable or disable at some point.

👎 Noted but not a priority as it’s still overseeable with the upcoming 3rd task.

08) Increase affordance of tooltips

Make stat boxes look clickable, and decide whether to stick with tooltips or allow users to go to a next screen with data breakdowns and/or more information (i.e. for contributions, pageviews, reverts). If the former, make tooltips more visually distinct.

👎 After consolidating the design team, we came to the conclusion that these tooltips do have a rather low priority in the interface as they are additional information. We’re going to leave them as is for now, especially since we’re planning to add a contribution history view and with it, specifically increase affordance of the “Contributions” stat. No immediate actions to take here.

09) Consistency in profile stats

Make contributions and pageviews both either cumulative or for the past 30 days for consistency.

👎 We settled on pageviews for the last 30 days due to performance reasons (technical restrictions). No immediate actions can be taken here.

10) Vandalism protection

If needed, consider ways of allaying the concern that the tasks are ‘too easy’ and that ‘anyone’ could do it (how have similar efforts to increase editing in the past address potential for additional vandalism?).

👎 As per November 21, vandalism is not an issue. Also we’ve added an additional layer of security to Suggested edits based on reverted edits. There are no immediate actions to take here.