12

Currently, if a moderator tries to close a question with a bounty on it, they are presented with the following nag screen:

nag screen presented to moderators when trying to close a question with a bounty

At this point, a moderator has to go to the mod tools, remove the bounty, and then close the question.

This should be streamlined so that when a moderator closes a question with a bounty, the bounty is automatically removed and refunded, and the question is then closed.

Right now, the screen is nothing more than a nag (for moderators) and doesn't serve any practical purpose.

Improve this question
12
  • Never had to do it before, but I agree. +1 By the way, currently, if you remove the bounty through the mod tools, is it refunded or lost?
    – Alenanno
    Commented Feb 5, 2012 at 22:00
  • 3
    @Alenanno It's refunded.
    – casperOne Mod
    Commented Feb 5, 2012 at 22:00
  • 6
    <insert snarky comment from comm team about how this is a case in 1/10000 flags and so you should suck it up and go the long route without whining> There! Now you can status-declined this.
    – Lorem Ipsum
    Commented Feb 5, 2012 at 22:08
  • @yoda <equally snarky reply comment about how taking five minutes and a few lines of code would make the aggregate time savings worth it, as we don't just measure the effect of the impact of doing it **just once**> and then get status-declined.
    – casperOne Mod
    Commented Feb 5, 2012 at 22:18
  • 6
    <insert low-rep user's snarky comment thinking he has any place commenting on such a post>
    – Lix
    Commented Feb 5, 2012 at 22:24
  • @yoda see the future much?
    – casperOne Mod
    Commented Feb 6, 2012 at 1:52
  • @casperOne Impossible to see, the future is; but this was pretty easy...
    – Lorem Ipsum
    Commented Feb 6, 2012 at 2:28
  • @yoda thanks for cherry picking =)
    – casperOne Mod
    Commented Feb 6, 2012 at 2:32
  • 3
    my issue is that this request is suggesting fixing the wrong problem. the real problem is that the community can not participate in closing bounties stuff.
    – waffles
    Commented Feb 6, 2012 at 4:42
  • 1
    @waffles Those are two separate issues. The community not being able to remove bounties is different from streamlining the process that a moderator performs when encountering this case. It doesn't fix the wrong thing, the wrong thing that you mention is something else. Not everything a moderator does has to have roots in an ability that non-moderators have (as evidenced by many of the current abilities mods have).
    – casperOne Mod
    Commented Feb 6, 2012 at 4:47
  • @casperOne by making it easier we would be increasing the chance of misuse. I looked through the bounties refunded in the last 30 days and was not perfectly happy with all the handling.
    – waffles
    Commented Feb 6, 2012 at 4:57
  • Related.
    – user1228
    Commented Feb 6, 2012 at 15:28

2 Answers 2

Reset to default
16

I think you should be required to consider the bounty removal as an extra step.

You may make the counter-argument that it doesn't matter whether a question has a bounty on it or not if it's still close-worthy, but that argument assumes that close decisions are black-and-white, when in fact they are not (especially if a mod gets involved in the closing).

Removing a bounty is not a reversible operation, so it should never be done silently.

Improve this answer
4
  • 1
    The proposal is not to do it silently, it's to automate the steps. There's a difference. There's no request for a loss of information. Additionally, there's no mention that a close decision is black-or-white, this is about a hindrance that occurs after that decision has been made. It's not about making the decision, it's about enabling the moderator when they've made a commitment to a course of action.
    – casperOne Mod
    Commented Feb 5, 2012 at 22:40
  • 3
    I think the point Robert is making is that when you're performing an action that is outright irreversible, it's good to have an extra step that gets in the way to make you think one more time, "do I really need to do this?" That's what it means for a close decision not to be black-and-white: the initial decision is not guaranteed to be the correct one (as opposed to purely logical questions which have single fixed answers that don't change no matter how many times you do them).
    – David Z
    Commented Feb 5, 2012 at 22:47
  • 1
    @DavidZaslavsky That's based on the premise that it's irreversible, which I believe is flawed. The user can post the same question (or a better one), and apply the bounty to that. Semantically, they are identical. Something like destroying a user account, now I would say that is much closer to irreversible.
    – casperOne Mod
    Commented Feb 5, 2012 at 22:50
  • 6
    @casperOne Irreversible doesn't mean "it's impossible to fix", it means "it's impossible for the mod to undo". You can't start bounties owned by arbitrary users, so removing one is irreversible
    – Michael Mrozek
    Commented Feb 5, 2012 at 22:58
12

Sorry, I am declining this.

Totally appreciate that you see your fair share of bounty refunds due to Stack Overflow's scale and amount of moderation you partake in. However, the manual step is there for a reason. We do not want to make this an easy task.

I just looked through the stats, we are refunding about 1 question a day on Stack Overflow. A tiny amount. Of these refunds the vast majority are to assist in closure. Of the closures some should probably have been migrated/closed by the community.

I don't see any reason to streamline this process. There are 2 distinct actions your are taking:

  1. refund bounty
  2. close question

I am open to amending some of our close logic to allow for casting close votes on bountied questions (even if we require mod intervention for the final closing). A close sends a much more powerful message when multiple users are involved.

As this is such an edge case that only introduces a minor amount of work I see no reason to change it.

My bigger concern is the final outcome which I find not that satisfying in most cases, cause the system forces unilateral closures (nonetheless, no matter how you look at it, it is a huge edge case)

Improve this answer
1

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .