11

To get credit for participating on the review queue, you need to take some kind of action. Either you need to downvote or flag (which is only appropriate if there is something wrong with the answer), comment or edit (which is only appropriate if you have something to contribute) or upvote, which is only appropriate if the answer is - to quote the tooltip - useful.

To determine that an answer is useful, I need to know that it is correct.

I do not know whether most answers to randomly-selected answers are correct, even within my own tags, and I imagine that I am typical in this regard. Perhaps this is simply because I am not a very experienced programmer and I'm projecting my lack of knowledge onto my image of the typical SO user; if others find that they reliably know the answers to most of the questions within their tags, then please set me straight. But I am assuming that this is not so for most people. It is certainly not for me.

Therefore I click Skip for most good-seeming, correct-looking answers I see - which accounts for most answers on the review queue.

However, I am conscious of the fact that this just means that somebody else will have to rereview the answers that I have spent time reading over only to Skip. I also suspect that if an answer doesn't have any grossly obvious flaws, then even if it is wrong it will always be upvoted and pushed off the review queue before anyone with the expertise to point out the error comes along and reviews it. Consequently I often feel like I am wasting my time and others' by skipping.

As such, my question is this:

  • Should I simply Upvote any answer on the review queue that I can't see anything wrong with to help get it off the queue?
  • Or should I continue to click Skip for 80% or more of the answers I'm asked to review, as I do now?
  • Or are there a significant number of supremely knowledgeable wizards on StackOverflow who know the answers to the majority of questions they see, who I should leave the review queue to instead?

I have similar doubts about reviewing questions, but wanted to restrict this question to the reviewing of answers since whether an answer deserves an upvote is much more objective than whether a question does. (To deserve an upvote, it must be right.)

  • 9
    "Should I simply Upvote any answer on the review queue that I can't see anything wrong with to help get it off the queue?" - NO!!! – animuson Dec 7 '12 at 20:18
  • You've skipped editing in your possible actions. Editing is probably the single most valuable action you can take, and I find that a significant percentage of posts in these queues are worth editing. – Servy Dec 7 '12 at 20:19
  • 5
    This is the reason why I don't even bother with the late answers/first posts queues – Sam I am says Reinstate Monica Dec 7 '12 at 20:19
  • 2
    @Servy True that I missed that. Edited my first para. I still think most posts fall into the "Looks good, but how the heck am I supposed to know for sure?" camp, though. – Mark Amery Dec 7 '12 at 20:21
  • @MarkAmery You might be interested in my answer to another question, where I complain about this very issue. – animuson Dec 7 '12 at 20:27
  • @animuson I just clicked through to it and saw that I'd already upvoted it on some prior occasion. :) I think we have fairly similar views on the current structure of the review queues. – Mark Amery Dec 7 '12 at 20:29
  • This exact reason is why I think it would be better if a user had to have some experience in a category in order to be allowed to upvote in the review queue, see my post on this related question: meta.stackexchange.com/questions/149621/… – WebChemist Dec 7 '12 at 20:37
  • Instead of upvoting every Late Answer you came across, you checked here first? I want to give you a hug. Seriously. – LittleBobbyTables Dec 8 '12 at 2:21
  • if memory serves, of 600+ posts I reviewed in Late Answers so far, I upvoted 2 or maybe 3 ("hidden gems" as instruction there calls them). Most of the rest are flags, 500 or probably more. Late answers are great for flagging, few weeks ago there was a day when I submitted about 50 helpful flags using this queue – gnat Dec 8 '12 at 8:38
9

I also suspect that if an answer doesn't have any grossly obvious flaws, then even if it is wrong it will always be upvoted and pushed off the review queue before anyone with the expertise to point out the error comes along and reviews it.

Yes, that's true, and it's a major problem with the current state of the review queues. Please do not contribute to this problem further by partaking in this inappropriate behavior.

should I continue to click Skip for 80% or more of the answers I'm asked to review, as I do now?

When you are in the situation you describe you should be skipping the items. Upvoting an answer when you have no knowledge of its correctness is harmful to the site. If you know that the answer is factually correct, and it is also formulated well enough to properly convey that correctness to the reader, then it's appropriate to upvote.

You should try your best (as hard as it is) to not worry about what else happens. Ask yourself, "in what way can I make the site better with respect to this post?" If you can make the site better by downvoting bad content, adding more information in a comment, improving grammar/formatting in an edit, flagging to remove content that definitely needs moderator attention, or indicating correctness through an upvote, then do so. If you can't think of any way of actually helping then you should skip. If other people choose to perform inappropriate actions...well...that's unfortunate, and when discussing changes to the queue as a whole here on meta, it makes sense to theorize about what other people are doing and why, but in the moment, when reviewing, that shouldn't be your concern.

Performing an action that you wouldn't normally do just for the sake of making your review count go up is harmful. Even if someone else might do that same harmful action, it's still possible the next reviewer won't, and that they have some helpful edit to make to the post.

  • In response to this, though, I have a question: what would happen if nobody partook in what you describe here as 'inappropriate behavior'? Is the answer not that we would have an enormous backlog of good-looking answers to specialized questions in the review queue that would get each get presented to, and skipped by, scores of users before being cleared, thereby stopping people from finding the proper bad posts that (I think) review is intended to filter out? That's why I wonder if I should just upvote, and also why I've argued for a "Look's fine" button in the past. – Mark Amery Dec 7 '12 at 20:37
  • 3
    @MarkAmery I think we might as well just do away with the entire queue, for exactly the reason brought up here and elsewhere. You're a lot more fit to interact with someone's post if you've come to the question intending maybe to answer, or get an answer than if you've just come across the post in some queue – Sam I am says Reinstate Monica Dec 7 '12 at 20:41
  • 3
    @MarkAmery If the only actions that could be taken within the queue were to upvote or downvote, then I'd agree with you, but that's not true. I find that for the vast majority of posts in either queue there's something that can be edited, or some aspect of SO guidelines that can be discussed in a comment. In any case, rather than theorizing about a problem that's not there (and is easily solved by just removing items from the queue after X skips) I would focus on addressing the problems we know we have. – Servy Dec 7 '12 at 20:48

You must log in to answer this question.

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