A problem I have encountered on ux.stackexchange.com is the subject of how broad is too broad for a question. I think this is compounded by the definition of 'too broad' being itself too broad to really be helpful. From the question guidelines:
"Your questions should be reasonably scoped. If you can imagine an entire book that answers your question, you’re asking too much."
'Reasonably scoped' is a very vague definition and 'an entire book' is an example taken from the most extreme end of the scale.
When it comes to Stack Overflow I think this is helpful. Programming problems are very pointed: 'Declaration X is not working under conditions Y ... here is my code' but I think with UX questions a clearer definition of 'too broad' is required.
If we are to look at the extremes of the scale between broad and specific we could see the following two example questions:
"Are things easy to use?"
"Is pinch and zoom a good method for zooming when viewing photos in apps using a capacitive touch screen on a smartphone, when considering elderly users who are missing a hand and are walking down the street?"
The first question would require a library to answer whereas the second could be answered but would probably only solve one problem for one designer and not be of much use for anyone beyond that as their situation would differ slightly. In fact, we could ask a million questions like that.
I feel there is likely to be a sweet spot between the two, a level of specificity which makes the question easily answerable in a few paragraphs and examples but broad enough to be transposable to other problems, to elicit some transferable truth and therefore be more useful to more people.
For example, each question on UX has several components that define the situation that the user is in which the asker requires an answer to. Such components, or variables, would include:
- the tool that is being used (both it's software and hardware components);
- the goals of the user;
- the condition of the user;
- the context in which the user is making use of the tool;
- the interaction method;
- a specific thing that the asker needs to be measured;
- (I am sure there will be more ...)
Each variable offers an opportunity to specify something and move from the broadest example question above to the most specific.
Perhaps these variables should be defined and a question expected to specify at least n of them to be passed.
How many variables about any given situation to we need to specify in order to create a reasonably scoped question?