Despite the declarative nature of what follows the title was (and is) intended to open the floor for others to argue what they think are the qualities of a good question. These are only my thoughts and arguments for them. If you feel they are errant or you simply have a different perspective please share your thoughts.
I contend that culturally we have an over-emphasis on "showing what you have already tried."
I was inspired to finally write this after reading the comments made regarding the recent question Drawing pairs (planar graph, circle packing), but that is only one example. Systematically I see an over-emphasis on demonstration of effort and prior attempts with regard to Questions.
Allow me to back up and explain my reasoning from the beginning.
This site operates on at least two different levels: the mechanical, and the personal. I think the Stack Exchange model actually emphasizes the first; to the organization we are all merely content creators, to the end of fueling search engine traffic and ad clicks. Nevertheless I neither deny nor apologize for the fact that we have a strong personal interaction on our site; in fact I don't think our site would exist without it.
I surely do not put my efforts into this site for the faceless goal of aggregating data on the Mathematica system. Instead I am here foremost to help people as I have explained before. I know from experience that it is frustrating when you feel that an asker ignores an answer you put work into and seemingly learns nothing from it. Nevertheless this site is ostensibly not about the personal interaction of one user with another, but instead the creation of Questions and Answers that benefit not only one reader but many. This is the mechanics of our site. This is not a discussion forum or an enthusiasts club.
While it is tempting to focus on a Question as the initiation of a personal interaction, that personal interaction is in the long term meaningless to the mechanics of our site. Ultimately what is important is the quality of Questions and the Answers they inspire as viewed by future readers with no personal connection to any of the contributors. (Of course in the here-and-now the personal cannot be ignored or there won't be any content for future readers.)
From this foundation I argue that a good question is:
- well written (implying clearly written)
- interesting to potential answerers
- valuable for future readers
- on topic and of reasonable scope
Further I argue that demonstration of effort as in "show what you have already tried" is irrelevant to a good question, and in fact often obfuscates one.
Did a lack of demonstrable effort prevent xkcd-style graphs from becoming popular far beyond our typical readership?
Would the example at the opening of this post be a better Question if it were encumbered by clumsy attempts that didn't work?
Is the same author's earlier and popular question Generating visually pleasing circle packs suddenly of less interest because he has (apparently) made a habit of not showing effort on solving the problems? (But perhaps on writing the Questions?)
Would not my own zero-effort-shown questions: How can I generate this "domain coloring" plot? and How can I calculate a jigsaw puzzle cut path? now be less clear and approachable if my halting efforts were included?
Does chyaong's original code in Generating a Sierpinski carpet serve any purpose? It is irrelevant to both the question and the answers (mine at least).
I previously argued, with notable popularity, that questions are what we make of them. I suggest that we make the best of the questions that are brought to our site, whatever the source. I suggest that we encourage people to post questions conforming the the four basic goals I outlined above, and that we communicate that their efforts are better spent on writing good questions than on demonstrating effort. I believe that personal interaction will (unavoidably) handle itself; people will be inclined or disinclined to answer certain users' questions as they always have been. Let us instead turn our focus to the mechanics of the site and to what makes for good Questions. Let us answer Questions not (only) because we like their author but because we find them of intrinsic value.
That my argument may not be misconstrued let me plainly state:
My argument above applies only to conceptual Questions of enduring interest.
Many technical Questions surely need to include code for clarity if nothing else.
Questions of narrow interest and limited future value need to both show effort and be as focused as possible, as they amount to requests for personalized support; they must therefore motivate responders with the hope of a receptive an appreciative audience, as well as minimize the time needed to diagnose and explain the problem.
I do not mean to discount or marginalize the personal nature of our community; if anything it is the more important. I too am seeking balance here.
I shall elaborate on my reasoning in an additional section tomorrow.