I've noticed that there are quite a lot of questions on site that are either very difficult to answer because:

  1. they're too general (i.e. don't focus on one particular problem or methodology)
  2. they're unthinking (i.e. are not logically reasoned or carefully thought through / are 'bad' questions)
  3. they're too specific to be generally useful (e.g. soliciting critiques on sites).

This has led me to wonder what could be done on site to raise the general quality of questions. I think there's a danger that the site could suffer (and risk alienating knowledgeable members of the wider UX community) if the number of non-useful questions greatly outnumber the useful.

I realise that it's important to ensure that the community grows, and that new users are made to feel useful - but part of me strongly believes that the situation needs to be improved if the community is going to grow to become an authoritative source of UX knowledge.


3 Answers 3


Your specific complains are all reasons for closure in my opinion, which is part of how Stack Exchange improves question quality.

In my opinion, the specific close reasons are as follows:

  1. Too General: Not A Real Question or Not Constructive; if it's too general it doesn't fit here, and needs to be
  2. Unthinking: Not a Real Question, or just plain downvote worthy. If it's not based in logic so much that it's unanswerable, it's NARQ and we should close the question. If it otherwise doesn't match a close reason but the question is just stupid, downvote the question. Downvotes sort out bad or problematic questions for improvement.
  3. Too Specific: Too Localized. We have a built-in close reason for this. If it only applies to one/very few people it needs to be closed for improvement. The question should be reworked so it's about a general case or asking for approaches to a problem rather than solving a specific problem.

As a user you can help us not only by asking better questions but by voting to close, downvoting or flagging questions where appropriate. Questions that aren't useful should be improved or closed for eventual removal if they can not be improved.


One possibility might be to try to raise the participation level of professional UX practitioners. I think a higher mean level of good questions could 'raise the bar' and increase quality organically.

Perhaps we should post something via IxDA?

  • Feel free to promote the site where you feel appropriate, professionally or socially, just don't be spammy. Community members are welcome to share any way they chose.
    – Ben Brocka
    Apr 6, 2012 at 14:44

I don't think there is a problem with the quality of questions. I think there is a problem with how UXSE forces users to frame questions.

I think part of the difficulty in asking questions and providing answers is that questions revolving around UX are entirely subjective, tacit, fluid, and/or contextual. So many questions are answered with the much maligned "It depends." And it usually does depend on myriad factors. Concise, explicit, and/or discrete answers simply do not exist in the field of user experience. It pains me to see a good question get closed because it's more of a conversation piece than a discrete answerable question.

The reality is, no one can confidently answer any question on UXSE (most of the time). Anyone who says they can either doesn't really understand the craft of user experience or is lying. The only way to find a confident answer or solution to a problem in UX is to test it, to perform research.

As Facebook says, "Code wins arguments." We can argue back and forth about the best approach to a problem but the only way to be sure is to build something, test it, and see which users prefer. And even then it's not 100%. User needs evolve along with the technology that supports those needs.

I think UXSE would be better suited if it did allow "discussion" type questions that are open ended. These types of questions can allow for insights and solutions to emerge. They can become unwieldy but that is a design challenge for UXSE.

  • We allow open ended, constructive questions as per the Good Subjective, Bad Subjective guidelines. If you want to discuss things User Experience Chat is always available, but the Question and Answer format doesn't work for them. We can't do everything well, but we do somethings very well. There are some very good, very answerable questions out there.
    – Ben Brocka
    Apr 6, 2012 at 14:43
  • Consider this post:ux.stackexchange.com/questions/3412/…. It has an answer and it was edited to "remove the subjective part". Unfortunately, I don't know what that consisted of but this question still remains very subjective. Which is more efficient? It depends! It depends on your users, how they're going to use it, where, when, etc. The answers are completely personal; they're based on what they think is best. But that might not be what's best for the OP. To only way to really know is to performance research. Apr 6, 2012 at 16:06
  • If you check the revision history the subjective bit was "Which one please you or annoys you the most?". That's a pure discussiony bit and all answers are equally valid; it's okay to ask that in chat but it's not fit for our Q&A. Answers need to involve facts, research, examples or expertise, not just "what do you like". The answers on that Q are still pretty meh, but the accepted one shows it's work and explains why.
    – Ben Brocka
    Apr 6, 2012 at 16:28
  • What we don't want are questions like "Should I use icons or text"? with answers that are "I like text" and "I like icons". At best it's a popularity poll, which are off topic.
    – Ben Brocka
    Apr 6, 2012 at 16:29

You must log in to answer this question.

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