YARAFP: Yet Another Review Audit Failed Post [closed]

Question

I just failed a close-vote audit by voting to close this question.

I genuinely believe there is a good case for closing it, as the question has a number of issues. Yes, it has five upvotes, an accepted answer, and 400+ views, but I don't believe the popularity of a question is a necessary indicator of its quality.

• It's a duplicate of this earlier question, which was downvoted to -2 and closed as "too broad." It could also be considered a duplicate of this question.
• The user doesn't state what they've tried so far. There's no clear effort put into solving the problem.
• It's only marginally on-topic for SO -- I've seen people go both ways on "hardware or software primarily used for programming" depending on how they're feeling that day. I'd consider it a more appropriate question for Apple Support. I might not agree with an off-topic close vote, but could see the case.
• Little context is given and the grammar is poor.

The best case for keeping it that I can see, and (as best I can tell) the reason it garnered the upvotes and views that it did, is that a number of people all had the same problem at the same time and found this post had an answer. But that doesn't mean it's a good SO question.

So is it an obvious guaranteed close? No. But I can certainly see the case for it, and that to me is why close votes are votes. Coming across it in the new-questions queue, with no upvotes and no answers, I'd have flagged/voted to close it.

I wish the suggestion here were implemented, and concur with jball's statement in their answer:

[...F]ailed audits are very invalidating. Fundamentally I can not click "I understand" on a completely incorrect warning.

To which I'd add: especially when these tests are being generated by algorithm.

I'd love to hear your thoughts and comments on this.

Answer 1

The way that you've evaluated the audit question can only be done by a human. There's no practical way to program a computer so that it can understand and evaluate semantics. Consequently, there are always going to be people who fail audits for reasons having to do with human factors.

I'm a moderator, and I still fail the occasional audit; in fact, I probably fail about 20 percent of them, for reasons that have partly to do with my particular workflow. Example: sometimes I will click vote to close to see how other community members have voted. That is a failing action on some audits.

As long as we have a non-sentient computer program choosing these audits, well-meaning folks will fail some of them. Honestly, I wouldn't worry too much about it. The people we're trying to stop are the robo-voters, not the folks who fail the occasional audit for reasons that a computer cannot understand.

Answer 2

FWIW, I disagree that it is a duplicate of the first question you linked - while possibly related (or caused by the same issue), the earlier question was exceedingly unclear and ultimately just requested speculation / psychic debugging from readers ("sometimes it works" problems are often really hard to nail down unless information is provided on what's special about some of those times).

The second question you linked to probably is a duplicate. Since it went unanswered, I closed it as a duplicate of the later question - closing in the other direction would just waste the time of future readers with the same problem.

Not stating what he's tried isn't ideal, but... I don't think it's a show-stopper here. The asker is doing something fairly straightforward that should just work - in a sense, "what he's tried" is summed up by the problem statement itself: "when I connect with mac mini, iPad doesn't show in Xcode".

The grammar could use some work. I've edited it - you could've done the same.

But I can certainly see the case for it, and that to me is why close votes are votes.

No, close votes are votes because different people have different criteria for closing things and closing is supposed to reflect a consensus. If you don't feel strongly about whether or not something should be closed, then by voting you're implicitly giving additional weight to the opinions of others, voluntarily throwing away your privilege as a voting-eligible member of the community. Vote to close when you believe something should be closed, not when you're undecided as to whether it should stay open. If the question is useful and on-topic, then leave it open; if you can't picture yourself defending your decision (either way) before a critical audience, then you should probably be hitting "skip".

Coming across it in the new-questions queue, with no upvotes and no answers, I'd have flagged/voted to close it.

But you didn't come across it as a new question with no votes or answers. We already have automated systems in place to detect poor-quality questions without regard for context - as a human reviewer with the ability to identify relevant context and the intelligence to interpret it, you should be able to go beyond simple checks and evaluate questions on their usefulness - this is why the existence of answers, their score and accepted state are exposed in the review system!

The primary purpose of audits is to help reviewers become better at what they're doing. If you fail an audit, try to identify ways in which you can review differently in the future: in this case, a few seconds fixing grammar and integrating the context from answers and comments would've made the question much more useful to others and passed the audit.