9

We've had some confusion in the past about our site scope, but we now have a clear description all in one place, with community consensus (+7/-0 is pretty substantial for our meta). So, what should we do about questions that don't meet these guidelines (and can't be edited to fix that)? Should we attempt to review extant questions for scope and close where needed, or should we leave things alone until they naturally pop to the front page and deal with them then?

Having open questions that violate our current guidelines can confuse people ("why was my question closed if this similar question is ok?"). Reviewing all the questions would require a substantial effort from the community. There may be a middle ground.

4
  • 1
    Since this question got upvotes but no answers, I've written up the two primary options I'm seeing. Have at it!
    – Standback
    Aug 31, 2014 at 17:53
  • I linked this in an answer below, but I guess it's pertinent to the entire discussion: meta.stackoverflow.com/questions/254402/…
    – Standback
    Aug 31, 2014 at 18:14
  • "can't be edited to fix that" includes two classes of questions: questions which can be salvaged without great effort but whose salvaging would make their answers inappropriate and questions which would be difficult to salvage. SE does not really have a good mechanism for handling answers that were good answers before an edit of the question. This is particularly significant if critique questions become off-topic (which seems likely), though if reference-request "research questions" remain on-topic (while "Questions asking us to do research are off-topic") they have similar issues ...
    – user5232
    Sep 8, 2014 at 22:34
  • ... a "do research" question may be answered by suggestions while the question could be converted into a reference-request question (invalidating the existing answer).
    – user5232
    Sep 8, 2014 at 22:38

4 Answers 4

Reset to default
4

Make a community effort.

Let's have a concentrated push to clean up the site and make it what we want to be. As a coordinated community event, we'll go back and try to close questions which violate our guidelines. This will:

  • clean up the site,
  • be a great opportunity for community participation,
  • sharpen community familiarity with the guidelines (and maybe sharpen the guidelines themselves).
7
  • I downed this (and upped the other) because it seems unfair to penalize old posters who were within the guidelines when they posted. It would be (sort-of) like getting a speed-camera ticket for going 35 in a 35 zone last year, because now the limit has been reduced to 25.
    – dmm
    Sep 3, 2014 at 18:45
  • @dmm I don't think it would penalize anybody; closed questions remain on the site and there's no reputation loss. It's just that they can't receive new answers. As sites evolve there are sometimes questions that were once on topic that no longer are; there's no shame in that. More about what closing means.
    – Monica Cellio
    Sep 3, 2014 at 23:26
  • 3
    We can also choose to apply a historical lock to borderline questions, particularly very popular ones that may have been linked to from elsewhere.
    – Goodbye Stack Exchange
    Sep 4, 2014 at 23:00
  • @MonicaCellio Closing does make it take a little longer to earn the Curious/Inquisitive/Socratic series of badges. Badges are generally lesser rewards, so making three badges more difficult to get for some people (those who have asked critique questions) may be a lesser penalty. Closing also has an emotional impact (perhaps less so for closings due to the on-topic definition changing); emotional impact is not necessarily proportional to reputation/badge impact. Closing may be the best option for the community, but it is not entirely without penalty to those who asked closed questions.
    – user5232
    Sep 7, 2014 at 19:20
  • 1
    @PaulA.Clayton fair point. If we close this questions, my intent was to leave a comment saying something to the effect that site scope has changed and while this question was on-topic when asked, it's not now. I don't want posters to think they did anything wrong.
    – Monica Cellio
    Sep 7, 2014 at 19:25
  • Should a thread be started for questions viewed as likely appropriate for closing? I do not trust my own judgment. I realize that a moderator or four other votes would be needed to close, but without editing an inappropriate close vote cannot be removed. Such a thread might also help clarify the off-topic standard and when questions can be salvaged.
    – user5232
    Sep 8, 2014 at 22:39
  • @PaulA.Clayton that's a good idea. Please feel free to start a new meta question ("should these questions be closed?" or some such). Discussing specific cases might help us reach a better (community) understanding.
    – Monica Cellio
    Sep 8, 2014 at 23:34
3

I believe a good compromise would be to close only the no-longer-relevant questions that appear on the very first page of the "frequent" or "highly voted" questions. This based on the truism that suggests that almost no one clicks to the second page of search results, and so closing only the most popular (and highly-linked) questions would eliminate the majority of potential for discord.

2
  • I like this a lot.
    – Standback
    Sep 9, 2014 at 8:06
  • These questions come up every so often, and people generally find them through searches (or the tags). What's proposed here is essentially what we were doing before.
    – Goodbye Stack Exchange
    Oct 1, 2014 at 21:06
2

Do a partial cleanup

Reviewing everything would be a big task, but we don't need to review everything in order to get some benefit. We should choose some specific areas (tag-based, ideally) to curate, areas where we will get the most "bang for the buck". Which are those? I would say that areas where we are getting the most new or recent off-topic questions are the areas most in need of clarification. If we are seeing problems with proofreading requests, for example, let's find and fix the existing, open questions that might be giving the reader the idea that those are ok. If we decide that critiques are not on-topic (please express your opinion there one way or the other!), then we should address existing critique questions because the reality will not match the policy otherwise. But old, off-topic, open questions that aren't popping up aren't doing much harm, so we can leave those alone.

Note that when I say we should address questions, I don't mean we should automatically just close them. Some questions can be edited to fit our scope better. If so, we should do that and give these questions a shot at some new answers.

I think selectively rehabilitating off-topic questions would be good for community involvement (editing, answering, and maybe thinking of related questions to ask). And I think it would be good for the site, by aligning site content with our guidelines in the areas that are causing the most confusion.

1

Let old questions lie.

Don't waste effort on old questions; just close new ones that violate guidelines. If an old question resurfaces (e.g. with a new answer), or is brought up as a reference, we could go and close it. But avoid closing bunches of old questions at once; it brings little value and mostly floats old, poor questions back up to the front page.

cf.: Is closing old questions a gigantic waste of time and effort?

3
  • I don't disagree with the sentiment, but I'm pretty sure closure doesn't bump a question to the front page. (I think reopening does, but I'm not certain.)
    – Monica Cellio
    Aug 31, 2014 at 17:50
  • 1
    I don't think the site should close old questions that don't meet new guidelines. What's wrong with saying, "Our guidelines changed"?
    – dmm
    Sep 3, 2014 at 18:48
  • @dmm That's a good question. Most sites eventually encounter what's called the "broken windows" problem: People tend to see these questions and ask more like them, figuring they're on-topic here. I think a historical lock would make the most sense for borderline cases.
    – Goodbye Stack Exchange
    Sep 4, 2014 at 22:59

You must log in to answer this question.

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