You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Description: There are times when queues can get inordinately long. Combined with the fact that requests will time out after a specified number of hours (this used to be hard-coded to 3, but is now configurable per course), this can wreak havoc on a course right before a big deadline. We'd like to investigate other ideas on how to handle this.
Who would benefit: Everybody.
The text was updated successfully, but these errors were encountered:
The reason the Request Time-To-Live (TTL) exists as it does now is because every so often we see people putting themselves on the queue right when they wake up, so that when they eventually make their way down to Halligan they're still sitting at the first spot. We'd really like to discourage this behavior, as it's not fair to the other students in the course. I figured that if, after 3 hours, a request hadn't been marked as resolved, it's very likely that the requestor wasn't in Halligan anymore. As has been shown by the past few days, this is clearly not the case.
Two ideas for solutions:
Lock the queue until a TA is actually on duty.
Implement a X-strikes-and-you're-out system, where if a TA attempts to find the student to help and can't, they'll press a "Can't find student" button. The student will receive a notification that they need to update the location on their request. If the "Can't find student" button is pressed X times, the student receives a notification that they've hit their limit and their request is removed from the queue. X will be configurable by course.
We could potentially combine these solutions as well, but I want to get feedback from people that have run in to this problem.
Feature Request
The text was updated successfully, but these errors were encountered: