Page MenuHomePhabricator

Introduce a way for people to explain why they are declining to act on a Check
Open, Needs TriagePublic

Description

At present, Edit Check presents people who decline to adjust the edit(s) they're making with a multiple choice survey.

This decline survey, typically includes Check-specific reasons (e.g. I'm not sure what citation to add, I wrote this content and it's not published elsewhere) as well as a generic None of the above applies option if someone deems the existing reasons insufficient.

This task involves the work of introducing a way for people to express, in their own words, why they are declining to act on a Check.

Story

  • As someone choosing to decline acting on an Edit Check and who does not see the reason why they are deciding to do so represented by the pre-filled options the decline survey presents, I'd value a way to express, in my own words, my thinking so that the people/person reviewing the edit can consider my point of view when deciding how/if to act on the change(s) I've made
  • As someone reviewing an edit in which someone declined to act on an Edit Check I'd value understanding why they specifically decided to make this choice so that I can consider this context in deciding how/if to undo/revise what they've done. Seeing what people have written could also help me develop a sense for the intention(s) that prompted them to make this change in the first place.

Ideas

  • What if the reason someone enters via Edit Check is automatically published as a new topic on the article's talk page and the link to said discussion gets automatically included within the edit summary for the main namespace edit.

Open questions

As @DLynch helpfully documented in T397889#10952002, any approach will need to address the following questions...

  • 1. Where would the text someone enters be stored?
  • 2. Who would be allowed to see the stored answers?
  • 3. If the answers are publicly displayed anywhere, what's the moderation story for them? (abuse, personal identification, etc)

Background

  • Currently, decline survey answers are stored by logging an event to the VisualEditorFeatureUse schema, which lacks a way to store any sort of free-text input. In the add-reference check we used to add revision tags as well (see: T342460 for storage discussions), but don't any more.
  • Respondents with <100 edits were more likely to have never used, but have seen, an article talk page (34% compared to 14%) | [English Wikipedia 2025 Successful Newcomers Survey

](https://docs.google.com/presentation/d/18ULw0UWzr6Hu_6X2FV3w89ammgICqN7WttqJpPSgnu8/edit?slide=id.g39c2ce30276_5_5#slide=id.g39c2ce30276_5_5)


An anonymous editor raised this idea on en.wiki here.