17
$\begingroup$

I am aware that this issue has been rehashed on Meta numerous times, so I was surprised to discover today that the guidance given to posters on the "Ask question" page is still far from optimal. This makes me wonder whether some of our persistent problems could be eliminated simply by providing more explicit guidance to new users.

Today I came upon a short post by a new user that needed substantial improvement in order to be acceptable for the site, and I wanted to point them to the "How to ask" page. Not remembering the link, I thought that surely it must be prominently displayed on the "Ask question" page, so I went there. Sure enough, in the right sidebar, at the bottom of the page, I found the link "how to ask a good question here" and quickly passed the link on to the new user.

Pursuing this further, however, I discovered that that link leads to this page, which, in my opinion, contains fairly generic advice that doesn't do much to help users avoid the contentious, site-specific issues that arise most frequently on math.stackexchange. I did remember a Meta post containing detailed site-specific guidance, and, after a quick search on Meta, I added that link in my comment to the new user.

Exploring further, I found that the "Ask question" page actually does contain links to the "How to ask a good question" post on Meta, but that they are presented in such a way that they are likely to be overlooked. At the top of the right sidebar, under "Step 1: Draft your question" there are links "Provide details" and "Write an interesting, specific title", both of which lead to answers to the Meta post. I think, however that many users will ignore those links, especially if they feel that their post and title already contain all needed information.

How about something like the following:

Dozens of questions are closed every day for failing to adhere to community guidance. Pay particular attention to the following issues.

  1. Do not ask questions that have been asked before.
  2. Avoid problem-statement-questions and "no clue" questions. Provide context.
  3. Other issues (question must be about mathematics, be detailed, have a good title, not be opinion-based or discussion oriented, not be about the operation of math.stackexchange, ...)

While we might not all be happy with current community norms, I think we can at least all agree that if new users aren't warned about them, we are setting them up to fail. If I were a new user, I'm pretty sure that if I wanted guidance I would click on "how to ask a good question" or "help center" or "asking help". But none of these, at present, lead to the most helpful information.

$\endgroup$
8
  • $\begingroup$ Do you prefer this post and this post? I feel like these posts are direct in their demands, quite comprehensive, and implementable. What I do is this : If I'm referring to the former post I usually say "focus on points 6,7". Likewise, with the latter post, it's points 2,3,4. $\endgroup$
    – Sarvesh Ravichandran Iyer
    Dec 3, 2021 at 16:17
  • 1
    $\begingroup$ @TeresaLisbon I like the long one (the one I linked to) better. A user doesn't have to read it all. A nice thing about it is that there's a summary of the main points at the top of the post, with links to the more detailed answers below. This gives the use a quick overview of the main issues. $\endgroup$
    – Will Orrick
    Dec 3, 2021 at 16:23
  • $\begingroup$ I see what you mean @WillOrrick. I think the kind of comment you suggest makes far more sense on the "Ask Question" page. We will have to trial it and see how it works out. It's effectively a bit negative in the sense "this is what you shouldn't ask", but maybe this could work, after all posters will then be informed that they can't post anything. $\endgroup$
    – Sarvesh Ravichandran Iyer
    Dec 3, 2021 at 16:30
  • $\begingroup$ @TeresaLisbon If someone wants to try to phrase the guidance in a more positive way, we can consider that. In the long run, however, I think it's kinder not to lead users to believe we're more welcoming than we actually are. $\endgroup$
    – Will Orrick
    Dec 3, 2021 at 16:42
  • $\begingroup$ You're right, saying it straight up positively or negatively before letting it come into the system might be more beneficial than inviting it and then closing it. I prefer the negative connotation, not because I want to scare new users but because I'd want new users to know that not everything is ok, and at the moment I don't think the question page says that at all in any degree of explicitness. $\endgroup$
    – Sarvesh Ravichandran Iyer
    Dec 3, 2021 at 16:51
  • 4
    $\begingroup$ Regarding point 1 in your suggestion, searching this site for anything involving symbols is a specialised skill in its own right, and decent MathJax knowledge is basically a prerequisite. That said, I don't see why we should treat closing as a duplicate as something to warn people about! If it's a good dupe target (i.e. it is reasonable to expect the dupe target to answer the asker's question), then we have likely provided exactly what the asker was wanting from us. Plus, reasonable dupe questions don't tend to glean as many down-votes or unpleasant comments as other closure reasons. $\endgroup$
    – Theo Bendit
    Dec 5, 2021 at 4:51
  • 1
    $\begingroup$ @TheoBendit Those are good points. In the context of duplicates, the answerers sometimes get taken to task, but usually not the askers. Ironically, the question I mention in this post that motivated the whole thing got closed as a duplicate, and not as a "no-clue" question, which is what I was trying to warn the poster against. It was obviously a homework question, and the closure as a duplicate provided the poster with what they needed to complete their assignment. One thing I would add is that dupe targets often show up in the "Related" list, so searching isn't always necessary. $\endgroup$
    – Will Orrick
    Dec 5, 2021 at 6:14
  • 1
    $\begingroup$ @TeresaLisbon: I proposed this improvement before, and you can read my reasons there. $\endgroup$
    – user21820
    Dec 13, 2021 at 18:31

0

Reset to default

You must log in to answer this question.

Browse other questions tagged .