The FAQ describes what makes a good question, but not what makes a good answer. I think at this point most of us know quite well how to distinguish a good answer from a not-so-good answer, so we should be able to add a section to the FAQ helping people out. It'd also be a great place to point people to when we want to help them improve their answers.

How would you describe a great answer? What makes it great? What's compulsory, and what's fantastic but not necessary? What kind of answers do we want to see on UX? What kind of answers would we prefer not to see?

Let's see your suggestions!

share
4  
More Spock, less Kirk? –  Erics Jan 17 '12 at 23:51
1  
This was recently discussed on MSO as well. –  Anna Lear Jan 18 '12 at 0:57
    
A miserable little pile of--oh wait, you didn't phrase it right for that. –  Ben Brocka Jan 18 '12 at 1:11
1  
Oh, that's an easy one. If it has "Roger Atrill" or "Michael Zuschlag" written at the bottom, it's usually great :) –  Vitaly Mijiritsky Jan 18 '12 at 18:27
    
Can we add this to the FAQs? I was just looking to cite something for a well-intentioned new user, and only found this (which doesn't seem super "official"). –  Daniel Newman Mar 23 '12 at 20:59

2 Answers 2

Posting a question is like being on a journey, but stuck at a crossroads. There may be many possible roads for the traveler (that's the OP) to take. They may feel they're lost and not sure they're even on the right road. They may simply not be where they thought they were. Or maybe it's foggy and they can't even see the way forward.

A good answer

A good answer is one which provides the opening poster (OP) with at least enough information to do one of the following.

  • Apply the information directly and proceed accordingly.

  • Continue as they were, having just clarified their position.

  • Decide upon one of several possible directions in which to proceed.

  • Consider backtracking down their path so far and re-plan their route.

A great answer

A great answer is one which not only meets but exceeds expectations.

A great answer provides extra information not just about what to do at these crossroads, but what's further down the road, what considerations may have to be taken into account.

It's not just answering the question - it's like looking at the method of travel. Is the traveller on foot? Take the shorter route. Is the traveller in a car? Consider the route via the filling station because you can see the dial on their dashboard is low (which they hadn't noticed).

A fantastic answer?

Yeah - a great answer - that's cool - the OP's going to love you for it.

But a fantastic answer? A fantastic answer is one which also paints a sign with all this information - and more, and leaves it at the roadside, so that all future travelers, using slightly different modes of transport, can make use of it and see where to go and what to do when you're not around to explain it all again.

Not just any old road sign painted in a thick brush - I mean a solid weatherproof information board with maps; pointers to more information on the subject; publications that might be useful; feedback from other people who have been on the same journey.

Tell a story - tell the whole story.

That's the way we should treat users.

That's the way we should answer questions.

Here's is a rare species of Answerus Fantasticus Exemplum.

share
1  
Inspiring! But quite abstract. Do you have some examples of what makes answers good, great, or fantastic? Point to some existing ones you've found on the site? Is an answer without references good? –  Rahul Jan 18 '12 at 12:14
    
yup - I took the metaphor and ran - just ran. That's how I travel. –  Roger Attrill Jan 18 '12 at 12:49

Research Citations

In programming, opinions are fine because we can test them in milliseconds to see if they "work" correctly.

Testing opinion-based answers is expensive for UX practitioners. A method to reduce the risk/cost of incorrect answers is helpful. Research citations enable practitioners to verify answer correctness without having to conduct experiments themselves.

Wikipedia handles the correctness issue with its Verifiability policy. To quote the linked page:

This page in a nutshell: Other people have to be able to check that you didn't just make things up. This means that all quotations and any material challenged or likely to be challenged must be attributed to a reliable, published source using an inline citation.


Additional References

Also, @Anna Lear referenced a good Stack Overflow answer in a comment above. While I think the author (yoda)'s code-related considerations add unnecessary complexity for UX purposes, he/she also posted another, simpler answer at meta.gardening.

Here's the table from it: enter image description here

share
    
Definitely. Of the questions I've asked, the more highly voted on ones are those where I've asked for studies / research. (ux.stackexchange.com/questions/14696/…) (ux.stackexchange.com/questions/6133/…) and also my top-voted response is one where I gave some hard facts (ux.stackexchange.com/questions/6929/common-screen-resolution/…) IMO this show's there's an appetite her at UX.SE for factual based questions / answers. –  JonW Feb 17 '12 at 16:08

You must log in to answer this question.

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