15
$\begingroup$

I don't understand how a community defines its norms given the impact of those who are not part of the community. For instance, the most upvoted question on the OR.SE site is about finding open source software. This is currently closed due to the votes of five people who, near as I can tell, have not asked a question on OR.SE or answered a question on the site. They are, however, experienced in other sites on SE.

Similarly, yesterday, my review queue had a half-dozen suggested closures by someone who is active in 135 SE sites, but has, near as I can tell, no interest in Operations Research. They certainly would not qualify as "experts" in OR. But they also don't seem to be official SE people. Some questions were put on hold due to this.

In no case were there any comments that might educate or inform us on why these actions were being taken.

I thought part of the purpose of the Beta is to define appropriate questions: these "drive by" closures don't seem to helping in that.

So who defines what are reasonable questions in OR.SE? OR.SE users or these highly experienced, but not OR.SE, people? And, if the latter, how can we avoid the "hit and run" aspects of their actions? And meta- meta-, will this be closed due to having 2 (now 3) questions?

$\endgroup$
5
  • 5
    $\begingroup$ And we get i n these idiotic tug of wars between closers and re-openers. How stupid and non-productive is that? The re-openers have to wait for a question to be closed before they can vote to re-open. It would be better if they could vote to keep open and neutralize close votes before the question is closed. $\endgroup$ – Mark L. Stone Jun 12 '19 at 2:18
  • 5
    $\begingroup$ My opinion: if a question needs clarification or improvement before answering, then leave a constructive comment, don't vote to close it. If someone posts their take-home exam, then close the damn thing. $\endgroup$ – Mark L. Stone Jun 12 '19 at 2:30
  • 1
    $\begingroup$ My comment on someone else's now deleted answer: Computational Science scicomp.stackexchange.com has apparently been Beta since Nov 2011. I only started looking at it in 2015, but I don't recall people running wild with ridiculous close votes and downvoting legitimate questions as has been happening on O.R. SE so far. $\endgroup$ – Mark L. Stone Jun 12 '19 at 2:42
  • $\begingroup$ @Mark Stone: with enough reputation you can vote to keep open, but only from the review queues. $\endgroup$ – kjetil b halvorsen Jun 13 '19 at 9:55
  • 2
    $\begingroup$ My question or.stackexchange.com/questions/57/… was closed as too broad for the 2nd time. It was closed, reopened, and now closed again. I can not vote to reopen now because I already did the first time it was closed. This is <>!! $\endgroup$ – Mark L. Stone Jun 13 '19 at 17:59
8
$\begingroup$

It seems to me that this community has expressed an opinion so far that list-type questions, which are unwelcome on many other SE sites and would be closed as too broad, are OK here. Some examples:

We also have a few questions that could be interpreted as opinion-based, but the community seems to be OK with them, such as:

Some of the questions linked to above are the ones that have been voted for closure by some of the SE veterans. I suspect they voted to close because the default SE position is that these questions are too broad/too opinion-based, and the SE veterans are trying to help make us aware of what does and doesn't usually work, by casting close votes. I take your point about the voters not leaving comments; I think on more established SE sites, a close vote is just interpreted as implicitly having the comment "I think this question is too broad" or whatever, but many of us are new to SE, so this implicit message is not clear. It just reads like "we don't want your question, go away," but that is not the intent.

If the OR.SE community wants to allow these questions, I think we can and should do so. The SE veterans each get a single vote, just like we do. We should all be voting to open/close when we think it's appropriate. It might also make sense to write a meta post that explicitly asks, "are list questions allowed?" If the consensus is yes, then that will provide guidance to the SE veterans about how we are shaping the site.

(My personal feeling about the specific question that you linked to is that it is, indeed, too broad/hard to tell what you're asking, but it's really just a phrasing issue. I am planning to post a comment about that, and will probably vote to reopen once the question is tightened up a little.)

$\endgroup$
5
$\begingroup$

The questions are closed following the Stack Exchange guidelines that applies to every Stack Exchange sites, a question that is off topic, too board or primarily opinion based will always be closed because that's how the Stack Exchange network works.

What exactly is off topic, too broad, or primarily opinion based maybe different from site to site as each site might do some tweaks to allowed certain things, but usually it's:

  • Questions asking for too much in one post

  • Questions asking about something that will have a whole lot of different answers.

  • Asking about something that can take up a whole book to answer it.

  • Questions asking stuff like "What do you think about X?" or polling questions.

  • Product recommendations, etc.

See: https://or.stackexchange.com/help/dont-ask and look at that same page on any other Stack Exchange site. Notice it's all the same no matter what site you are reading it from? That's because those rules apply for all sites.

Currently the reputation required to close and reopen questions is low—in private beta only 1 rep, which allows anyone to cast those votes. When the site goes public you will need to have at least 500 reps to cast close and reopen votes. From this you can see that after a while a point will come where you have to be active on the site to close and reopen questions, and along with that the community can come together on the site's child meta to discuss matters on how the site should operate. A page can also be made to tell what type of questions should be asked, but we don't have that yet because the community still isn't sure exactly about what to put in it as of now.

So you guys will be moderating your own site along with your elected pro tempore mods.

$\endgroup$
5
  • 1
    $\begingroup$ Thank you. I have looked at that, and the question does not fit the form of the "don't ask" questions. But my main question, which obviously was not sufficiently clear, is the SE norm that the interpretation of "Don't Ask" will be primarily done by those who otherwise have no interest in the topic at hand? Perhaps the answer is "Yes, SE norms are run primarily by those outside the area, particularly for Beta sites". Given that sites can be in Beta for upwards of 10 years, it is important to know this. $\endgroup$ – Michael Trick Jun 12 '19 at 1:44
  • 1
    $\begingroup$ @Michael Trick Computational Science scicomp.stackexchange.com has apparently been Beta since Nov 2011. I only started looking at it in 2015, but I don't recall people running wild with ridiculous close votes and downvoting legitimate questions as has been happening on O.R. SE so far. $\endgroup$ – Mark L. Stone Jun 12 '19 at 2:26
  • $\begingroup$ My previous comment was for a slightly different wording. The clarification you have made on the rep required clarifies how this works, and is very helpful. $\endgroup$ – Michael Trick Jun 12 '19 at 10:17
  • $\begingroup$ The rep requirement for closure on private betas is 15, not 1, but this doesn't make any difference for users with the 100 rep association bonus from another SE site. However, it seems that it is intended to be 1 and it is 15 only due to technical reasons $\endgroup$ – Discrete lizard Jun 13 '19 at 13:57
  • $\begingroup$ This is so true, while there's a group whom likes exploring they're the very ones whom don't properly curate. Defining what's open (what subjects we want to include) is useful, and our purpose during early beta. Revisiting what (categorically, not specifically) to close is not as productive. $\endgroup$ – Rob Jun 18 '19 at 3:38

You must log in to answer this question.

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