2
$\begingroup$

I see a number of questions (https://robotics.stackexchange.com/questions/10313/design-of-scara-joint-problem) where a combination of english as a second language and a question from a user new to the stackexchange world (https://robotics.stackexchange.com/users/14287/alaa-momen) causes the question to be put on hold rather quickly.

Some of these people might have contributed to robotics.stackexchange in the future had they not been turned away so quickly.

Robotics straddles a lot of disciplines. It can be difficult to articulate a question well when you are asking about an area you are weak in, more so in a second language.

Does it really hurt the site if you leave the vague questions open for a couple weeks; give them a chance to be clarified? Give the new community members a chance to participate and learn?

For the question above, the user does not have the experience to know that the kickstarter project they saw was clearly created by amateurs and might not be a good one use as an example; but on the other hand it's clear that they are looking for a low cost joint design that would work for hobby type maker machines with SCARA kinematics. As far as questions about mechanism design go, that's not too bad for someone that's not a mechanical engineer. A small amount of shepherding could turn this into an interesting question and hopefully a couple good answers.

This question is just one example of the broader issue though.

What happens if you aren't so quick to put questions on hold?

Here's a vague question that could have been closed as too broad, but in the end was nicely illustrated. Preventing leaks in motor shafts for underwater bots

Here's a shopping questions that should have been closed but apparently is one of the best robotics.stackexchange has to offer given the upvotes and views. How to choose the right propeller/motor combination for a quadcopter?

My main point is that I think that putting questions on hold as quickly as you do hurts the growth of the site. Sometimes the only way to clarify a question in a topic as broad as robotics is to propose an answer and then refine both the question and answers iteratively.

$\endgroup$
2
$\begingroup$

I feel like I beat myself up every single time I close a question, because I ask myself this question every time.

In generally, personally, the criteria I'm looking for in a question are:

  1. Is there a correct answer? If I think that three people could post three different things, and all three would be equally valid, then probably the question is not well defined. This generally means it's a broad/life question ("What books do you suggest for a beginner like me?") or it's an unbounded design question ("Mounting a gimbal BLDC motor"). I base this criterion on the first bullet point of the Don't Ask page, "avoid asking subjective questions where every answer is equally valid: “What’s your favorite ______?”
  2. Is the question a poll? Similar to the above, if the gist of a question is, "I want everyone to tell me their opinion on my books to read/choice/method/implementation" I'll close the question. This is basically a restatement of the above.
  3. That's about it, actually.

Generally, if the question is, "What's the best ___ I should use?" I'll close the question. If the question is (similar, but notably different), "How do I determine the specifications for ____?" I'll leave it open.

The first question means the OP already has some set of solutions in mind and wants our opinion on what the best one is. However, generally not included in the question - specifications. OP would like for us to say, "___ is the best," but without including any background on the particular application. Sometimes I'll close those questions with the "show your working" reason, which basically asks OP, "Well, what did you try, and what did you get?"

Every time I close a question I try to add a message in addition to the canned text notifying the OP or other people what action I think they should take. This is usually something along the lines of, "Edit your question to include ___ and I'll reopen it," or, "Come to chat and we can discuss your application."

Why I don't feel so terrible about closing questions

  1. Regarding chat - it looks like three people have come to discuss problems since the beginning of April - that's three people in 5 months. It's a resource, you can ping me (@Chuck) or MarkBooth, and I generally try to check chat at least once a (week)day even if I haven't been pinged.
  2. Regarding OPs editing their own question - they don't. Those two are the most recent ones I closed where I said, "Edit your question/be more specific" and then OP never came back to edit it. There are more, but I'm not going to go digging at the moment.
  3. The biggest reason - I count on the community to keep my closures in check.

Regarding that last point, do you know how many questions currently have a vote to re-open them? None. Not one. There are zero questions with even one vote to re-open. Generally, with almost every question I close, I'm torn as to whether or not I should close it, so the first vote to re-open that question gets I'll usually re-open it.

Now, I'm not advocating that you go through and vote to open everything, because I closed them all for a reason, too. But, if you think a question could be re-worded and would then be a great question, then by all means, edit the question and then vote to re-open it! I'll approve the open vote and the edit, and then I won't close the question again.

There are some questions I'll probably never re-open, but those are the ones like, "Guys! I saw a robot on TV last night and now I want to build robots. Where do I start?" Other than that, if you think a question should be re-opened, edit it to an acceptable state and then vote to re-open it.

I generally check the site at least 3-4 times M-F and usually at least once Sat/Sun, so it should really only be a day or so (if that) before I take action. Also, MarkBooth is capable of re-opening a question if I'm not around or you all (non-mods) can just re-open the question yourselves with enough votes.

tl;dr

I close it because I think it violates the rules. If you think the question can be interpreted in such a way that it doesn't, then please edit the question to make that more clear and then vote to re-open. The question will be re-opened.

$\endgroup$
  • $\begingroup$ This is very insightful and pretty reasonable, but I'm not asking why questions are being closed, nor what the rules are. That's clear elsewhere. I'm asking if it might be better for the moderators to let the community self regulate a bit more before they step in. To let questions that look a bit unclear or too much like opinion or shopping questions, have a chance to improve before they get put on hold. $\endgroup$ – hauptmech Jul 22 '16 at 2:10
  • $\begingroup$ Some notes on your numbered points: $\endgroup$ – hauptmech Jul 22 '16 at 2:47
  • $\begingroup$ Some notes on your numbered points: 1) Perhaps people don't know about chat. 2) If an OP does not participate, that's fine. Their question will die a natural death without moderator help won't it? 3) stackexchange is community regulated and moderators only step in for flagged or exceptional cases. Why would a community member challenge a moderators action unless it was particularly grievous? $\endgroup$ – hauptmech Jul 22 '16 at 2:57
  • 1
    $\begingroup$ @hauptmech - I think if the Robotics community were bigger then I could sit back and allow it to self-regulate. A poorly worded question at Electronics SE is closed by votes within minutes. Robotics is a beta site, but our community is small enough that, even with a reduced number of votes required to close, a question may linger for a long time before it gets closed. This is compounded by the fact that the majority of our users have relatively little reputation (are relatively new). $\endgroup$ – Chuck Jul 22 '16 at 12:11
  • $\begingroup$ So I view my role as needing to be more hands-on. At one point, a question I wound up closing was commented on by a relatively new user, who was involved in other questions I had closed. The new user commented that the question was too broad and liable to get closed. Because I set an example, a new user was able to identify and comment on a poor question. $\endgroup$ – Chuck Jul 22 '16 at 12:15
  • $\begingroup$ Regarding your question, then, I would say that the community isn't acting fast enough to close low quality questions. Those then stay open, people answer, and then everyone gets the impression that low quality questions (polls, open-ended design questions, etc.) are acceptable. The problem with that is those usually only help OP, which means that they probably won't help future visitors. Letting a bad question get answers doesn't make it a good question. The only way that happens is to edit the question. $\endgroup$ – Chuck Jul 22 '16 at 12:15
  • $\begingroup$ So, last comment from me for now. I do get your point, and I will try to take a more active position on encouraging OP to rephrase or restate the question such that they still get an answer and that it meets quality criteria. I am a mod, but I'm also just some guy. If you disagree with any of my actions, vote to undo them! I certainly welcome the feedback. Also, FYI, I've been struggling with this same topic in moderator chat. Ultimately I think I've expressed my conclusion clearly above - close poor questions, however many that might be. When the community catches on they'll take over. $\endgroup$ – Chuck Jul 22 '16 at 12:29
2
$\begingroup$

Summary:

  • Leaving poor questions from new users open creates too much risk that they will set a poor example for other new users (see broken window theory).

  • There are not enough active community members with close privileges to leave questions to be community moderated.

  • Moderators are always willing to re-open questions which have been improved, so if you see a question which you think should be re-opened, either vote to re-open if you can, or flag it for moderator attention if you can't.

  • For the good of the site as a whole though, moderators also want to know about poor quality questions, so if you see a question which you think should be closed, either vote to close if you can, or flag it for moderator attention if you can't.

$\endgroup$
  • $\begingroup$ Good idea @hauptmech, feel free to accept this answer to push it up to the top. $\endgroup$ – Mark Booth Jul 25 '16 at 10:31
0
$\begingroup$

It is always a fine balance to strike between being welcoming and willing to help new users, and letting them get into bad habits which you know they will never shake off.

Questions such as the first one you mention, which tick multiple boxes on the list of close reasons are unlikely to be of use to anyone without significant revision, so they have to be closed as soon as possible. If they then get edited to make them into suitable questions, they can be re-opened, and we try to make it clear that this is what is needed.

Using Stack Exchange effectively is a skill which has to be learned. Anyone who is not willing to learn how to write an effective question, or useful answer, is not going to get very far with Stack Exchange.

Does it really hurt the site if you leave the vague questions open for a couple weeks; give them a chance to be clarified? Give the new community members a chance to participate and learn?

Yes, it can cause significant damage to the site. This is related to Broken windows theory, the more poor quality questions remain open on the site, the more poor quality questions are written. The lower quality the site appears to be, the less likely people are to trust the content of the site and the less likely they will come back here to have their own questions answered.

Compared to many sites, we have quite a relaxed attitude here on Robotics, perhaps too relaxed. I don't go hunting for questions to close, I tend to wait for people to flag or vote to close questions, so they appear in my moderator or review queue. Most of the time, a question which has been brought to my attention does require closing, and although it is rare, sometimes subsequent revisions will get a question re-opened.

My main point is that I think that putting questions on hold as quickly as you do hurts the growth of the site.

Looking at the analytics, the number of questions closed is insignificant compared to the number of new users, but the growth in page views is less than the growth in users, which suggests many people are signing up and then not really using the site.

So what you are suggesting may well encourage the small number of people who get their questions closed, while making it less likely that a far more significant number of people will be turned away due to low quality questions they find when they get here.

† Analytics are a moderator tool, so the specifics can't be shared, but I try to share what information I can within the bounds of what is allowed.

$\endgroup$
  • $\begingroup$ I don't have a grasp of the analytics (nor any time to do other than raise the question above), but on the surface it looks like 2 of 4 Moderators are AWOL and there are less than 50 active community members for the last year. It does not feel like there is a critical mass of active members (though there are some very good ones active and lurking). $\endgroup$ – hauptmech Jul 23 '16 at 3:19
  • $\begingroup$ While some of us are more visible than others, looking at the stats, three moderators have been active this month, while the other was last active in March. We currently have a fairly light workload though, over the last three months we have averaged a flag every other day. Even when adding in the review queues, the workload isn't high. On larger sites, moderators spend more time on mod queues and let the community take care of reviews, since those site have enough users with close privileges. $\endgroup$ – Mark Booth Jul 24 '16 at 15:52

You must log in to answer this question.

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