34

Few arguments against it:

  1. Pressing the flag button doesn't mean actually submitting it (as in every test/form).
  2. There might be an unknown duplicate, if you assume it's the first post (as you were told) it might be worthy to flag it.
  3. You might want to look if someone else thought it was duplicate by "making half way" to flagging it.

Why pressing the flag button makes you fail an audit? I think it shouldn't.

Edit:

The implementation difficulty (fake flagging UI) and the tips about how avoid failing (and reviewing!) by skipping the answer adds some insight, but still doesn't answer the question why it's not a valid review.

Moreover, I would like to add one more, this question describes a reviewer that failed an audit while trying to flag a question as off topic. The question was clearly off topic, and still after 1300 views no one flagged it.

Improve this question
27
  • 7
    Having a meaningful discussion about this will be difficult without knowing what the message said. Presumably, the audit system thought that there was nothing wrong with the question; therefore, casting a flag would be a disputed action.
    – user102937
    Commented Nov 7, 2013 at 19:27
  • 5
    OK. So if the question is OK, why would a flagging attempt be considered valid?
    – user102937
    Commented Nov 7, 2013 at 19:31
  • 6
    I'd agree here - I'm not personally at the level where I can hit the queues yet, but I often use the helpful flag interface when I'm not sure. I see a question that I'm a bit dubious about, I head into flags to poke around and see if there are any that seem to fit. If there are, I'll flag them, and if there aren't, I'll back out and think about something else. I don't think that's an unreasonable use of the flagging system (it seems to have been set up to support that specifically) and I don't think it's unreasonable behavior for someone who's looking through the queues.
    – Ben Barden
    Commented Nov 7, 2013 at 19:33
  • 15
    Pro tip: nobody fails an audit by clicking the "skip" button.
    – user102937
    Commented Nov 7, 2013 at 19:35
  • 3
    @RobertHarvey no, you fail the audit by actually attempting to review the post.
    – Sam I am says Reinstate Monica
    Commented Nov 7, 2013 at 19:36
  • 7
    My question is, Why can't you wait until the user clicks done to tell them whether they've passed or failed the audit? Is there a logistics issue that i'm not understanding?
    – Sam I am says Reinstate Monica
    Commented Nov 7, 2013 at 19:38
  • 8
    Reviewing audits is like walking through a land mines field: you only proceed if you know exactly what to do.
    – Szymon
    Commented Nov 7, 2013 at 19:39
  • 2
    @Szymon: That might be true if you were punished for every failed audit. You're not. SE doesn't care about the folks who fail the occasional audit; they only care about the robo-reviewers.
    – user102937
    Commented Nov 7, 2013 at 19:39
  • 10
    @RobertHarvey Not really. You're punished if you make too many mistakes. The system only knows you made a mistake. It doesn't care if you spent 10 minutes trying to review and putting effort or if you clicked the button in 2 seconds.
    – Szymon
    Commented Nov 7, 2013 at 19:52
  • 3
    @Robert, there might be different opinions on what punishment is, and different levels of punishment down the road, but a stern warning demanding that I agree that I'm a failure sure feels like punishment to me.
    – jball
    Commented Nov 7, 2013 at 20:30
  • 6
    @RobertHarvey that would be a pretty good reason, as often ranted about, to turn down the nastiness level of the message.
    – Rosinante
    Commented Nov 7, 2013 at 21:31
  • 6
    @RobertHarvey I, and others, find it really offensive. Why not save the stern wording for people who have failed four warnings in short order, and say something milder for isolated failures?
    – Rosinante
    Commented Nov 7, 2013 at 21:34
  • 4
    @RobertHarvey "Consider this from the programmer's point of view." I'd rather not. I'd rather consider it from a user's point of view. On this site, I'm a user, not a programmer. The current UI, while easy to implement, does the wrong thing. Correctness beats ease of implementation!
    – John Kugelman
    Commented Nov 11, 2013 at 22:13
  • 2
    @JohnKugelman: And staying in business beats correctness. Regardless of what you might think about how important your particular issue is, there are always dozens of other issues waiting for attention. Every software company has to decide which issues come first, and which issues get resolved later, if ever. See stackoverflow.com/a/2806990
    – user102937
    Commented Nov 11, 2013 at 22:38
  • 2
    FWIW: if someone has flagged or voted to close a post as a duplicate, there'll be a comment to that effect below the post. There'll also be a comment if someone has pointed out a duplicate without voting or flagging. If you're using the flag dialog to look for duplicates, you're doing it wrong.
    – Shog9 Mod
    Commented Nov 11, 2013 at 23:11

2 Answers 2

Reset to default
8
+100

I don't want the software developers spending time on redesigning the review dialog just because some users want to "peek" at other community members' review actions before deciding on their own. Down that path lies bad reviews; if the first reviewer gets it wrong, the subsequent reviewers will follow him over the cliff. I've seen it happen, many times before.

If you really want to see what actions the other community members have taken before deciding on your own review action, open the post directly, so that you will not be subject to a review audit.

Improve this answer
4
  • This answers 1/4 of the cases.. (third bullet)
    – Yosi Dahari
    Commented Nov 11, 2013 at 23:15
  • It answers all three.
    – user102937
    Commented Nov 12, 2013 at 7:04
  • There might a valid reason to flag as duplicate (bullet 2) and a valid reason to flag to close as off topic (in the edit - bullet 4).
    – Yosi Dahari
    Commented Nov 12, 2013 at 7:15
  • 1
    The review dialog already asks you if the question should be closed as a duplicate (if someone else has voted to do so). Pressing the flag button indicates intent to flag; if your intention is anything else (like peeking at other users' flags), go to the actual post, and push the flag button there.
    – user102937
    Commented Nov 12, 2013 at 18:23
0

The review system is not the place for thinking:

Hm, I wonder what other people have done with this post? Are there any flags? Has someone found a duplicate?

You should be using these queues only if you're confident of your ability to review first posts, late answers, low quality posts, etc. Read and react. If you need to see what others think, open the post in a new tab and see the missing comments or whatnot, but actions in the review pane should be reviews, not attempts to get more information in order to complete your review.

The audit system assumes you clicked flag because you intended to flag it. In the review queues, flagging is a negative action and indicates that you believe there is something wrong with the post. The fact that you have reasons for clicking flag and not completing the action when you're outside the review queue does not make this a good course of action in the review queue. Stop doing that, and click flag only when you intend to flag the post.

Improve this answer
6
  • 7
    "The audit system assumes you clicked flag because you intended to flag it." -- The audit system assumes wrong. Quit assuming! Problem solved.
    – John Kugelman
    Commented Nov 11, 2013 at 23:46
  • 3
    What would be the benefit of that? Reviewers should review with confidence. If they want to open the post in a new tab then ok, but to redo the audit mechanism to support tentative reviewers who have a workaround seems unnecessary.
    – Kate Gregory
    Commented Nov 12, 2013 at 0:07
  • 1
    See @BenBarden's comment: "I see a question that I'm a bit dubious about, I head into flags to poke around and see if there are any that seem to fit. If there are, I'll flag them, and if there aren't, I'll back out and think about something else."
    – John Kugelman
    Commented Nov 12, 2013 at 0:08
  • 2
    @JohnKugelman that's fine out in the wild. But questions are in review because someone has flagged them, or they've been identified by assorted heuristics (length, first post, late post etc) so you're at dubious before you even start to read. If you read it and it's no good, deal with it. If you read it and it seems fine, why should the system support "well it looks ok to me but maybe it's not ok, I wonder what other people think?" That's not what a reviewer should be doing.
    – Kate Gregory
    Commented Nov 12, 2013 at 0:13
  • 2
    Ben didn't say anything about seeing what other people think.
    – John Kugelman
    Commented Nov 12, 2013 at 0:19
  • 2
    If you are in "something is wrong, I wonder what it is that I should flag it as" mode, then you've decided something is wrong. So in that case, the audit system is not wrong to interpret your click that way. And if you're so unfamiliar with flag reasons that you might click flag, then read the reasons and decide "oh never mind this post is perfect" then I still feel you shouldn't be reviewing.
    – Kate Gregory
    Commented Nov 12, 2013 at 0:54

You must log in to answer this question.

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