Having now spent some time on the review queues I thought it might be apposite to propose one or two refinements. I note that the current review system is just over a year old, so perhaps it's a good time to look at how it's working.

Three common categories I see for new user questions are:

  1. Eminently closable questions, no redeeming features. Career advice, choosing tools or methods, opinions, lists, etc. The question cannot be fixed, but it's a brutal way to treat a newbie.
  2. Horrible formatting, incomprehensible English, unformatted code. The user should be told to clean up the question and resubmit, and I don't want to do it for them.
  3. Potentially good questions that are far too short. The user should be told to expand the question the question and resubmit, and I can't do it for them.

So I'm regularly faced with questions that I can't or don't want to fix, but that badly need fixing. I can add a comment or I can flag for deletion, but I don't have a satisfactory way to push it back on the user to read the help and fix the question (or delete it).

[Meanwhile it's just as likely someone else will OK the question while I'm still thinking about it. This can be pretty frustrating, but that's another story.]

A solution could be a pre-canned comment similar to "Oh Hold", that blocked the question until some other reviewer cleared it, and with a link to relevant help text. I'm sure others could think up better ways.

This question is related: How to help hapless newbies become better SO users.

[There are similar problems with reviewing answers and approving edits, but that's enough for now.]

In response to comments: The choices I get to see when I flag a question are strongly oriented towards closing, not fixing. I am reluctant to flag for deletion a question that just needs some work by the OP.

What I think might help are options that are oriented towards saving the question rather than discarding it. That might be some different wording, and/or a lower threshold for putting the question into that state and getting it out again. Maybe 2 votes instead of 5; maybe rep 2K instead of 3K+.

If this is heading toward a suggestion for a new feature, then this is what I would suggest.

  • A new status for either a question or answer called "under review"
  • Can be switched on or off by anyone during a review (so one vote by 2K+ either way, but only during review)
  • Otherwise similar rules to "on hold": no votes or answers, recycles through review queue.
  • Final deletion should probably still require voting, but I leave that up to the high reps.

The purpose is to catch bad questions very quickly and push the job of fixing (where possible) back on the author, not on the reviewer, hopefully with some helpful comments.

  • 6
    One good way of putting it "on hold" is to actually put it on hold. If the question meets these criteria you've outlined, why is flagging/voting to close not the best option?
    – animuson StaffMod
    Apr 13, 2014 at 15:15
  • So you'd want to have an "on hold" state which would be different than the current "on hold" state (which is in fact the "closed" state but under a different name), right? And this "on hold" state would be attainable after one person with the necessary rep votes to put the question on hold? Or am I misunderstanding??
    – Louis
    Apr 13, 2014 at 15:19
  • 2
    ok... I wouldn't like to put the power to close a question singlehandedly to the hands of our biggest repwhores. I agree with your categorisation, however. Apr 13, 2014 at 15:24
  • 2
    your activity at different sites indicates that you 1) review from LQ queue and 2) focus on questions more on smaller site (Programmers) than at SO. If this is so, consider editing to reflect that - thing is, at SO closing of questions recently became much faster, and it's unlikely that feature like this will have much effect there
    – gnat
    Apr 13, 2014 at 16:07
  • ...when working in SO close queue, I frequently stumble upon questions gaining 5 close votes in 20-30 minutes - stuff like that is exceptionally rare at Programmers
    – gnat
    Apr 13, 2014 at 16:13
  • @gnat: I review what I find. I have equal rep on SO and Programmers and I really don't focus on one or the other. Yes they're different, but the newbies look much the same. I have imperfect understanding of the various features and nuances, and I'm not sure I can do much better.
    – david.pfx
    Apr 14, 2014 at 3:49
  • @animuson: See edit.
    – david.pfx
    Apr 14, 2014 at 3:52
  • @Louis: Yes, I think you've put your finger on it. See edit.
    – david.pfx
    Apr 14, 2014 at 4:56
  • 1
    The idea of closing a question is to prevent wrong answers and give the OP (or another user) the chance to improve the question. If it is edited it will show up in the reopen queue for review. Apr 14, 2014 at 7:12

1 Answer 1


I understand your point. I think the current vote to close is sufficient to help the OP work on his/her question. It doesn't provide a solution for the people not voting while they actually need to.

I find it frustrating myself very often that high rep users able to vote to close don't vote since they can come up with a reasonable answer and gain high rep with that.

In answer to your edit and comment:

I think closing is the only option. You don't know whether OP is capable of and willing to edit the question to quality. If not, the question will remain closed. Else it can be reopened easily.

  • See edit. I'm reluctant to flag to close when the question is obviously salvageable. I suspect all these choices look quite different to high rep users and I don't actually know what motivates them.
    – david.pfx
    Apr 14, 2014 at 4:58
  • @david.pfx wrt reluctance, see Do you wait for edits before voting to close a question?
    – gnat
    Apr 14, 2014 at 9:23
  • @gnat: Yes, that link is about the same problem and yes, I wait and no, I can only flag, I don't have the rep to vote on close. This is a quest for a middle ground when reviewing. Something more definitive than comment and wait, something less brutal than instant vote/flag to close.
    – david.pfx
    Apr 14, 2014 at 10:21
  • 2
    @david.pfx on 90% okay questions, flagging / voting to close is very far from being brutal. If you feel like a simple edit can salvage it, reopening is as easy as it gets: edit pushes the question into reopen queue from where it gets reopened
    – gnat
    Apr 14, 2014 at 10:36
  • 1
    @david.pfx Adding to what gnat said. If you know that the close queue has >10k questions in it, you might wonder whether a question to be reopened might just get lost in the flood. The reopen queue is most of the time empty. So a question that lands in the reopen queue is guaranteed to get attention quickly.
    – Louis
    Apr 14, 2014 at 12:50
  • @Louis: Yes, 14.2K right now, so how many don't get closed that should? And how many get reopened? I'm suggesting a quicker lighter way to suspend questions that are good candidates for salvage, to make the review process work better.
    – david.pfx
    Apr 14, 2014 at 14:25

You must log in to answer this question.

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