tea.mathoverflow.net - Discussion Feed (Best practice for flawed questions)

Anton Geraschenko comments on "Best practice for flawed questions" (5680)

2010-05-31T18:46:52-07:00

I like the rule of thumb that if an answer has to begin by guessing what the asker wanted and then answering that guess, then the question is vague enough that it should be closed for revision. That way, people can still suggest alternative formulations in the comments, but are prevented from spending too much time answering a question that may not even be what the OP was asking. Also, such flawed questions tend to be edited after an answer or two is given (and turns out to be answering the wrong question), in which case the answers end up looking totally bogus to everyone's annoyance.

Mariano comments on "Best practice for flawed questions" (5641)

2010-05-26T08:25:09-07:00

If badges made sense, there'd be one "Fixer" awarded for turning the non-sensical into the sensible and interesting. And Joel would have been awarded a couple of dozens of them.

François G. Dorais comments on "Best practice for flawed questions" (5631)

2010-05-23T13:30:42-07:00

Is it better to close flawed questions or to leave them open?

The example I have in mind is question 25664, which is not a bad question at the heart but the formulation is flawed in such a way that it is impossible to give a definite answer. The two excellent answers by Joel and Andrej both proceed by outlining the flaws of the question and proposing alternatives formulations. Note that this particular case is a little moot since the author has expressed support for closing.