In general I agree with the guidelines of what makes a good question, but my overriding goal is "what questions are useful to learn from".

In general, I think people upvote questions that they have found useful or have learnt from. But that shows a disparity between what we are considering acceptable questions and what members find useful, as many questions with good numbers of upvotes have been closed as not being a good question.

So how can we evolve what is an acceptable question where it makes sense to improve the site?

1 Answer 1


Yes, it evolves, and it's set by community guidelines and the community's interpretation of the general Stack Exchange guidelines, the most basic of which is that this is a question and answer site.

Popularity in and of itself is a bad measure as well; popularity is great, but without a second measure it's pretty useless. Otherwise Justin Bieber is the best singer in the world; no matter how many people agree with it, I think most of us agree we don't want the "Justin Biebers" of questions on this site. To my mind those are the rants, the discussions that go in circles for decades, the surveys that don't really mean anything. To counter the Justin Bieber questions we have deep questions about HCI, Psychology and applying these things to real situations. Things get popular for stupid reasons and great content gets ignored for stupid reasons, so let's keep our eyes on usefulness and constructive content.

Per an earlier discussion in chat, I'm pretty sure you're particularly concerned with this question: Better term for "user"

My view is that this question isn't acceptable because it's really not a question with answers, it's a poll/survey question. There's no right or wrong, nor are there solutions. There's just a list of options and voting indicates popularity, not correctness, usefulness or practicality. That's an abuse of the voting system when votes are supposed to indicate usefulness. Gauging opinions can be useful as a whole, but winning a popularity context should not earn one "usefulness" points.

Even worse, IMO there are some good answers that could be given to the underlying problem. Personally I feel the whole "let's not call them users" thing is a big pile of crap and how you treat, respect and interact with people is far more important; I think a very valuable question/answer set could be created on that idea. But that's not what the question is; I can't answer that question as posed in that way. It begs the question that "user" is a naughty word and doesn't call for anything more than a list of nouns (and that's more or less what it got).

So there's room for community guidelines and changing them, but there's a few classes of "questions" that just don't do well on Stack Exchange, and that's by design:

  • Discussions: Answers here pretty much stand alone, by design. Back-and-forth is explicitly discouraged at almost every level of the system. We are not a forum by-design
  • Surveys: The problem here is mostly when answers are judged based on popularity rather than insight or quality. Sure X is a great book, or maybe Y is a great term, but if it's just a suggestion it's not an answer, nor is "well I use X and X works". When every answer is equally valid and votes just indicate popularity that's just an insult to all the questions where hard work is being rewarded instead of "Oh man this book is cool right?"
  • Rants: X Sucks, am I right? I mean Metro? Skeuomorphism? Left aligned websites?! AM I RIGHT? Well maybe you are, but that's not a question. Good questions instead get down to "why is X still used when it is seemingly bad" or "Can I use X in a helpful way or does it just cause problems?"

All of these things have their place; discussions can be engaging, surveys can be insightful, and rants can be a great read. Heck, they can even be useful. But just because it's useful doesn't mean it fits here. Physical keyboards are useful, right? But they still don't really work on an iPhone.

Good design includes things because they're useful. Great design excludes things because they get in the way. Focus is vital. Sometimes things cause more harm than good; discussions and surveys are the Stack Exchange equivalent of a physical keyboard on a mobile device. It works great on other stuff, but not here. That's not really what we're doing here, and if you tape in on here, it distracts from what we're really doing: Questions with answers.

  • I'm not actually referring to that question. I've noticed a number of useful questions that have been marked as closed lately that I have learn't a lot from. I feel the site is poorer for not including some of them.
    – JohnGB Mod
    Nov 30, 2012 at 14:33
  • I find the idea that questions have to be limited to ones with a right / wrong answer or a solution, a limited perspective. That may apply well on SO where there are objective right / wrong answers, but UX is a far more subjective field and so applying the same criteria that an objective field uses seems like a mistake.
    – JohnGB Mod
    Nov 30, 2012 at 14:39
  • 2
    @JohnGB I never said they have to have a right/wrong answer, though those questions can certainly be on topic. They just need an actual answer, not "try axure" or "I think responsive design is a fad". There needs to be some metric for correctness or usefulness beyond "people agree with this". A good answer is not good because it is popular, it is good because it is useful and explains it's reasoning or important concepts.
    – Ben Brocka
    Nov 30, 2012 at 14:41
  • Thanks, that actually helps clear it up for me.
    – JohnGB Mod
    Nov 30, 2012 at 15:07

You must log in to answer this question.

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