30
$\begingroup$

I know the review audits aren't without their flaws, but with this review:

https://math.stackexchange.com/review/first-posts/109542

it should be clear that something is not right. As usual, the vote tally, acceptance mark, and username were hidden.

Since I liked the answer, I was going to praise the author in a comment. Or, so I thought. Instead, I got a "STOP! Look and listen" by clicking "add comment".

This is ridiculous. Ab-so-lutely ridiculous. There are a myriad of reasons why one could decide to comment that have nothing to do with assessing overall quality -- just to name a few: praise, clarification, or mentioning a typo.

I know only one of and should apply. I hope it's , for that means that the audit failure wasn't "by design".

Bump. (A.K.)

Is there an actual reason for this? Even a year later, when the review audits have been at least somewhat modified, that adding a comment still causes an automatic fail?

$\endgroup$
7
  • 5
    $\begingroup$ It seems to me that the review system was written by someone with a particular workflow in mind, and so they removed a lot of things (or didn't implement them) that would be present in the normal workflow when a question is visited. Another example: it is not possible to vote on a question from the "suggested edit" review page. But, when any of the black action buttons is pressed, the screen changes, and there is no longer a link to the question to make it possible to go vote on it. It seems they expect that, when you are reviewing, you forget all the other aspects of the site. $\endgroup$
    – Carl Mummert
    Commented Oct 16, 2013 at 10:52
  • 1
    $\begingroup$ @CarlMummert If not for you, then for future readers: the "history" tab of the relevant review queue will show the questions you reviewed. I agree that the reviewing system requires a bit of practice to circumvent all of its confusing properties. $\endgroup$
    – Lord_Farin
    Commented Oct 16, 2013 at 17:04
  • $\begingroup$ If the post is a "bad" one then, I discovered, commenting will mean you pass the test. Perhaps it is just so that people don't comment as default in order for them to always pass reviews. However, I don't think commenting as default is a bad thing... $\endgroup$
    – user1729
    Commented Nov 8, 2013 at 10:25
  • 1
    $\begingroup$ @user1729 Commenting should simply constitute neither a pass nor a fail. $\endgroup$
    – Lord_Farin
    Commented Nov 8, 2013 at 10:48
  • 2
    $\begingroup$ If failing a review audit is the worst thing that happens to me today, then I'm having a good day. Try to learn to accept the fact that the review audit procedure is full of goofiness. $\endgroup$
    – Jay
    Commented Jun 10, 2014 at 21:48
  • $\begingroup$ Comments are now disabled on review audits. I tried to comment on a review audit and got a pop-up telling me that I couldn't comment because it was an audit: math.stackexchange.com/review/reopen/803358 $\endgroup$
    – Stella Biderman
    Commented Apr 27, 2017 at 16:24
  • $\begingroup$ Happened to me just now in the First Posts queue. The answer was good and I wanted to suggest how to improve it further. math.stackexchange.com/review/first-posts/941836 $\endgroup$
    – wlad
    Commented Jan 22, 2018 at 0:39

2 Answers 2

Reset to default
6
$\begingroup$

This just happened to me: https://math.stackexchange.com/review/first-posts/223879

The post was good, and I wanted to add a comment suggesting how to solve it.

I'm disappointed that after nearly a year this bug report has still not been investigated by the moderators.

$\endgroup$
3
  • 5
    $\begingroup$ Let's be clear that the site moderators can do nothing about such things. It's up to SE employees to change the review algorithm. They are certainly aware of this issue but it seems they consider the current state of events the lesser of the evils. (The greater one being robo-reviewers clicking "add comment" on every first post.) $\endgroup$
    – user147263
    Commented Jun 9, 2014 at 23:14
  • 1
    $\begingroup$ Oy. Sounds like the SE employees need to read up on perverse incentives, and then fix or outright remove the reviewer badges... $\endgroup$
    – user7530
    Commented Jun 10, 2014 at 0:03
  • $\begingroup$ As you can imagine, this has been proposed. But on the other hand, there aren't enough reviewers on SO as is. So the audits got introduced and were tweaked many times since: Meta.SE is full of review-abuse posts. $\endgroup$
    – user147263
    Commented Jun 10, 2014 at 0:12
3
$\begingroup$

In some kinds of review queues (such as Review: Close Votes), attempting to leave a Comment results in an error message (on submission of the Comment, "An error occurred during comment submission."). This is fairly diagnostic of a review audit.

However when I tried to leave a Comment on this First Posts review:

https://math.stackexchange.com/review/first-posts/565488

it was an immediate Fail, as this meta-Math.SE thread notes and decries.

It is somewhat ironic that this post was selected to be disguised as a First Post, since it had been mentioned here by the Original Poster in connection with limitation of "unilateral" Moderator action.

So I was intending (with some suspicion) to note that the problem had already been asked a few weeks back.

Although the Community came around to accepting the particular post, it would have been legitimate to Comment on the apparent duplicate (as the First Post review audits portray the posts as new and unanswered).

Added (2/2/16): By the same token today I "passed" a First Posts Review audit by trying to add a comment (on a post that had already been removed). It seems to me the spirit of the First Posts Review tilts toward constructive advice, and whether one should be "audited" on this is dubious.

$\endgroup$

You must log in to answer this question.

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