10

I occasionally fail audits for the Reopen review queue. My problem is not that I click without any thoughts, but that I honestly don't think the questions should have been reopened (had they been closed in the first place).

Today's example is this

https://stackoverflow.com/review/reopen/1503501

where the question is

Why can a string literal be implicitly converted to char* only in certain case?

void f(char* p)
{}

int main()
{
    f("Hello"); // OK

    auto p = "Hello";

    f(p); // error C2664: 'void f(char *)' : cannot convert parameter 1 
          // from 'const char *' to 'char *'
} 

Now C++ happens to be my specialty, and I can immediately see that there is no string literal being converted here. The p in f(p) is a pointer variable, and not a literal.

To me this looks like a rather bad question that really deserves to stay closed. The "Not a real question" close reason looks entirely reasonable.

Review failed?!

Why do we only get an "I understand" button in the review screen, and not an "I disagree" button? Because I really don't understand what I did wrong!

8
  • If you feel strongly about this being a NC close candidate, do not forget to vote on that question. Commented Feb 12, 2013 at 10:55
  • According to Tim's answer it's based on flags. It's not really duplicate of this other question however the answers there pretty much answer this question as well. Commented Feb 12, 2013 at 10:55
  • 2
    +1; I am having the exact same issue! I'm getting a little click-shy on the Reopen Queue! I really do try to be very careful when reviewing. I fail a honeypot now and again; but the Reopen queue gets me an average of once every other day or more. Commented Feb 12, 2013 at 10:58
  • 3
    I'd love to see an "I disagree" button; would love to see how many people failing audits are going to click that one. Can we detect controversial audits with that feedback and rotate them out perhaps? Can you make that a feature request? Commented Feb 12, 2013 at 10:58
  • 2
    The question is open and upvoted. That's the problem, not the audit.
    – yannis
    Commented Feb 12, 2013 at 11:03
  • I don't feel strongly about closing the question (partly because there are special rules for string literals, just not in this example), but I do feel strongly against reopening lower-quality questions. We have seen that upvotes is not a sure indicator of quality, sometimes it just indicates fun or bike-shed issues.
    – Bo Persson
    Commented Feb 12, 2013 at 11:52
  • 2
    I can also see a problem with an "I disagree" button, because the users the audit is supposed to catch are perhaps likely to disagree anyway.
    – Bo Persson
    Commented Feb 12, 2013 at 11:55
  • I have a feature request to choose audits differently and allow some reviewers to disagree with them: meta.stackexchange.com/questions/168374/… Commented Feb 20, 2013 at 19:59

1 Answer 1

4

I describe the criteria used to select these - and some of the problems with it - here.

That said, the goal here is to get you to look at a closure with fresh eyes. The whole point of having re-open reviews is to examine closures and determine if the question actually needs to be closed.

Is the title on that question misleading? Sure - the error message itself tells the author he's converting pointers and not strings. But if anything, that's a good reason to fix the title - the question itself clearly describes a point of confusion encountered by the author, and does not seem at all "ambiguous, vague, incomplete, overly broad, rhetorical or unanswerable" - in fact, it attracted several answers that explain the discrepancy without much trouble.

To date, no one has felt the need to close-vote or even down-vote that question, even with the additional attention this meta post brought it. That makes it a pretty good audit.

2
  • 2
    Maybe the problem here is that I didn't ask myself "Should this question have been closed?" but "Should this question be reopened?". So, presumably the question is closed, it's not a very good question, and no important edits made. From my point of view, do I need to take some kind of action here? No, all is well - "No action needed" or "Leave closed" was the obvious choice. Of is "Skip" perhaps the equivalent of "No action needed" for this queue?
    – Bo Persson
    Commented Feb 20, 2013 at 20:45
  • You make a good point - the interface does ask, "should this be re-opened". While that is the decision you're making, in cases where no edits have occurred the implicit question does become, "should this have been closed".
    – Shog9
    Commented Feb 20, 2013 at 20:50

You must log in to answer this question.

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