Note: This task is not about (semi)automatically asking task authors to re-test an issue after a while and in the same action (semi)automatically setting the "stalled" status. That would be a different topic and ticket, maybe T248034.
Stalled is defined as "If a report is waiting for further input (e.g. from its reporter or a third party) and can currently not be acted on, the Stalled status is temporarily given."
This can mean several things:
- "stalled" task status due to being blocked on an external upstream task outside of Wikimedia. Hence should be tagged Upstream.
- "stalled" task status due to a subtask which needs to be resolved first inside of Wikimedia. Hence should have a subtask.
- "stalled" task status due to feedback required from the task author. (Which in itself is also a blurry concept as the task author could also be set as assignee instead or in addition?)
- "stalled" task status due to misuse in order to just hide a task on a workboard or in a list of tasks.
- (maybe something else that is not coming to my mind?)
I would not want to add more statūs / having more field values to differentiate between #1, #2 and #3. Keep it simple.
I'd rather want to clean up #4. When identifying a list of old "stalled" tasks without updates to potentially (semi)automatically decline, I'd also want to exclude #1 and #2 (DB query?).
In any case, this will require a person who sets the "stalled" status to also always explicitly ask for resetting the task status and explain how to reset the task status. Hence this would require being more aggressive in resetting the task status whenever "stalled" status is set without a clear explanation who (person or upstream project) to provide which information or perform which action.