9

(Forked from Learning the art of the review queue)

For those who often "Skip" in reviewing close-vote items because of lack of confidence / experience / training, is there:

  • A more detailed guide to how to be a good reviewer (than the brief text in (more))
  • A way to view some historical examples which would train a someone in best practices?
  • A list of recommended meta.se reading
13

While the following outline breaks down the various close types, the heart of the matter is really "is this question answerable (here)?". I found a relevant quote from jasonwryan (one of the first 20 users registered on the site, over 6 years ago!) while researching this:

I make a judgement call about the overall quality of the question and whether it meets the threshold; once I have made that decision, I then start looking around for the correct justification.

I'd refer you back to the earlier review question where I suggest simply paying attention to the site's questions and answers to get a feel for the types of questions that get answers vs the ones that get closed.

One way to tell if a question is answerable is if you think that you could answer it. The trickier way to tell if a question is answerable is if you think that someone on the site could answer it.

A question in the close-vote review queue has received one or more close votes; they may all be the same type or of varying types. You get to decide whether you want to leave the question open, close it (for a particular reason), or edit it.

Note that choosing the Edit button -- instead of choosing the Close button -- implies a vote to leave open! Related link: Editing before vote to close . If you want to edit a post and close it, open the post in a separate tab/window before VTC.

Here's how I broke down the process and some thought processes along the way.

  1. Enter the close-vote review queue.

  2. Read the question; notice the specific close-type at the top; the close type listed is the majority of the vote types given, which may often be unanimous, but not always. Get a sense for what the question is about. Note if the asker is a new user or not; they may not be familiar with the site yet. Consider their experience level with Stack Exchange in any comments you leave.

  3. If it’s been flagged as too broad, see if you agree. Can the question be answered by someone who has knowledge of the question’s topic (see also the tags)? Can it be answered in a reasonable length post? Does it lack specific details that would lend itself to a specific answer? Has the question been edited recently to address concerns in the comments?

    • It is too broad; VTC. If there's a specific part of the question that makes it too broad, or you have thoughts on how to productively narrow the question down to an answerable level, and there's not already a comment to that effect, consider leaving such a comment.
    • It is answerable; Vote to Leave Open.
    • Not too broad, but VTC for a different reason.
    • Skip.
  4. If it’s been flagged as a duplicate, read the duplicate question(s) and see if you agree. Has the question been edited recently to address differences from the duplicate(s)?

    • VTC as a duplicate.
    • The duplicate does not answer this question. Vote to Leave Open.
    • Skip.
  5. If it’s been flagged as Off-Topic, problem that can't be reproduced, read the Comments (and any self-answers) to see if you agree. An easy case is clear indication from the OP that the problem no longer exists. A “simple” typo could be cause for closure -- assuming the OP knew better, and it was a mistake. Typos from ignorance could be a teaching opportunity, and could be left open if it’s apparent from the question and/or comment string that they don’t understand the typo in question.

    • The problem can’t be reproduced; VTC.
    • There is an instructive typo here; Vote to Leave Open.
    • VTC for a different reason.
    • Skip.
  6. If it’s been flagged as Off-Topic, posted on multiple sites, look to see if there was a link left in the comments to the cross-post; otherwise, click on the user link, then their Network Profile, then on the “activity” tab, and optionally on the “posts” sub-tab. See if a similar (or the same) question was posted elsewhere. Consider that the cross-post may have been deleted in-between the OP being made aware and your review.

    • If a cross-post exists, VTC. If you had to find the cross-post, leave a comment with a link for the next reviewer.
    • If it had been posted, but is now gone/deleted, optionally leave a comment to that effect and Vote to Leave Open.
    • No evidence of a cross-post, but VTC for a different reason.
    • Skip.
  7. If it’s been flagged as Off-Topic, requests for learning materials, see if you agree. Meta links: How do we feel about requests for learning materials? (from 2014), Add specific close reason and explicit point in the help pages for learning material requests (also from 2014), and Are we abusing "Request for learning materials" as a reason to close? (from 2016). Questions should have a fairly specific problem to them, and not be of the form “what are the steps to set up an mail server?”. To me, these are a specific variation of “too broad”, where the “right” answer is a potentially ever-changing tutorial from the vendor or elsewhere. See if the question been edited recently to bring it back on-topic.

    • VTC.
    • Vote to Leave Open.
    • VTC for some other reason.
    • Skip.
  8. If it’s been flagged as Off-Topic, migrate to Meta, see if it’s asking about the site instead of on the site about an actual Unix & Linux topic. Rare.

    • VTC, migrate to Meta.
    • Vote to Leave Open.
    • Skip.
  9. If it’s been flagged as Off-Topic, migrate to SO, is the question on-topic at SO and answerable (clear) there? (Meta link for the creation of the SO migration path). Also: Meta discussion on programming questions, and the link from our tour page for UNIX C API and System Interfaces being on-topic.

    • VTC, migrate to SO.
    • VTC because it’s unclear or not answerable as-is; optionally leave a comment that a better Q could be on-topic at SO (or elsewhere). “We don’t migrate crap”.
    • It is on-topic at U&L and is answerable; Vote to Leave Open.
    • It is on-topic at U&L, but VTC for a different reason vote to leave open; otherwise, the question could get incorrectly migrated to SO (since there are other votes to migrate).
    • Skip.
  10. If it’s been flagged as Off-Topic, other custom reason, see if you agree with that reason. Has the question been edited recently?

    • VTC with the existing reason.
    • VTC for a different reason.
    • Vote to Leave Open.
    • Skip.
  11. If it’s been flagged as unclear, see if you think it’s answerable as-is (seeing, but ignoring the comments for the sake of closure). Has the question been edited recently? If OP comments would clear it up, leave a comment to suggest editing the post.

    • If answerable, Vote to Leave Open; optionally leave a clarifying comment -- something that you think the OP may have meant but didn’t make explicit, that would make the question clearer. If it’s an obvious or extreme case, consider making (or proposing) an edit (as well as a comment, for future reviewers).
    • If not clear/answerable, VTC as unclear.
    • VTC for some other reason.
    • Skip.
  12. If it’s been flagged as being primarily opinion-based, see if you agree. See if the question been edited recently.

    • VTC as opinion-based.
    • A slight adjustment could save it: additionally leave a comment.
    • VTC for different reason.
    • It is answerable with facts -- Vote to Leave Open.
    • Skip.

close-q flowchart

  • I'm not sure if this is the best format, but I'd welcome any feedback or clarifications on any of the points! – Jeff Schaller May 29 '17 at 3:32
  • I have the graphviz source for the image, if anyone has improvements for it. – Jeff Schaller May 29 '17 at 13:02
  • "each close-vote only comes through as one type" - not necessarily. You might also need to choose which incoming reason is appropriate. That's probably most common in POB/too broad, but it happens for other pairs too. – Michael Homer Jun 4 '17 at 21:35
  • 2
    I would like migration to SO to be less eager: is it a good programming question not about shell, sed, or awk scripting or POSIX APIs, and not about obtaining or using software on Unix system? Some questions are on-topic in both places and certainly shouldn't be migrated (but, too often, are). We also end up migrating too many questions that aren't on-topic at SO at all. Because migration goes by majority, if it shouldn't be migrated and there are votes for it you need to leave open and comment - giving another close reason isn't good enough. – Michael Homer Jun 4 '17 at 21:50
  • I've updated the post to address SO migration and the close-types. – Jeff Schaller Jul 9 '17 at 18:11
  • "One way to tell if a question is answerable is if you think that you could answer it." As the question scope increases it becomes harder to know whether a question is unanswerable even with domain knowledge. At that point you're left with only two options: Leave Open or Skip. (Case in point, without reference, I saw a question recently that was perfectly reasonable within the scope of a shell on a QNAP. It was closed as unclear, probably because in the scope of a shell on a full Linux/UNIX distribution it didn't make sense.) – roaima Feb 12 at 19:58

You must log in to answer this question.

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