7
$\begingroup$

Some days ago a new user posed a problem and sketched a sweeping line algorithm combined with a tree data structure to solve it. Here the problem, as far as I remember:

There is a set of rectangles with sides parallel to x and y axes and a set of points. find the point that is enclosed by the largest number of rectangles.

This was the link to the question before it was deleted: https://cs.stackexchange.com/questions/79534/point-contained-in-the-most-rectangles

The user also tried to give an estimation of the running time. My impression was that both the algorithm and the running time estimation were correct. He wrote that he wants to know if the algorithm is correct because he is not familiar with this kind of algorithm. A user told him in a comment that such kind of questions (check my algorithm and maybe check the estimation) are not welcome here. When I came back to this page some time later it was removed.

As far as I can see question of how to solve a problem algorithmically are allowed on this site. Also I see that posters should try as much as possible to solve there problem and also post their efforts here and tell what is there problem ("What have you tried and where did you get stuck?") So im very puzzled by the fact that reviewing an algorithm and its running time is not in scope off this site. Does the user post too much? should he only post the first three lines of his algorithm and write "I do not know how to continue and do not know how to estimate the running time"? The seems to be a little bit ridiculous. But it also seems to be the consequence of such a "we do not review" policy.

$\endgroup$
3
  • 1
    $\begingroup$ See some pertinent discussions linked here. Such questions are indeed problematic, but mostly for themselves: few users read long posts (this one here isn't too bad) and think through everything, only to confirm for the asker that there does not seem to be a mistake. Also, a yes or no doesn't make for an answer that would help anybody else. We can certainly discuss again how to deal with such questions, thanks for bringing it up! $\endgroup$
    – Raphael Mod
    Aug 4, 2017 at 12:07
  • 1
    $\begingroup$ I agree with @Raphael, the questions/answers are problematic mostly to the user asking the question. I have gone through that same rabbit hole spending significant time trying to answer my own question before only to be left a little confused at the end and needing someone to check it. In my opinion, when it comes to other users viewing the question, typically "check my answer" is not as helpful as "what's the best approach for this problem? Here's what I've tried...". $\endgroup$
    – ryan
    Aug 4, 2017 at 18:49
  • $\begingroup$ After reading some posts @Raphael pointed to I think posting a problem with a solution does not necessarily mean that the user has put any effort in solving the problem. For example in cs.stackexchange.com/questions/9831/… the user states the problems (a formula tha defines a language), notes some production rules and asked "what do you think?" From this I can't deduce that the user has put any effort in the solution of a problem. The solution can be copied from anywhere. $\endgroup$
    – miracle173
    Aug 8, 2017 at 6:01

1 Answer 1

Reset to default
4
$\begingroup$

The two extremes are questions that are essentially "Please solve this exercise for me" and those that are essentially "Please grade my solution to this exercise." Neither of these is a good question for this site: the former because we're not in the business of doing people's homework for them; the latter because questions are expected to be generally useful, and it's unlikely that anyone will ever want that answer graded again.

In particular, the now-deleted linked question essentially just says, "Here's my answer. Could you check it, please? I'm not very confident in it." If they'd asked concrete questions about their solution (e.g., I've assumed that every widget is green but I can't prove that – is it true? Can I avoid that assumption?) then the question would have been fine. Indeed, in that case, the question could have been asked as a question about the colour of widgets and only mentioned the exercise and attempted answer in passing as motivation. Surely people will want to know about the colour of widgets in the future, so the question is generally useful.

To specifically address the question of whether somebody can show too much effort, I'd say generally no. But we have to be clear about what the problem actually is. In the case of "Please grade my answer", the problem isn't the exercise the asker was trying to solve but, rather, the unstated thing that's causing them not to be confident about their solution. So the problem with this kind of question is that the asker has shown too much effort on something that isn't their problem, and hasn't even said what their problem is.

$\endgroup$
6
  • $\begingroup$ "the latter because questions are expected to be generally useful, and it's unlikely that anyone will ever want that answer graded again." That is true. However, take my case into consideration. I'm a student, so seeing questions on solutions for exercises (as long as they don't fall into the "solve this for me, please" category) helps me understand notions and the whole thinking process. $\endgroup$
    – theSongbird
    Aug 6, 2017 at 11:14
  • 3
    $\begingroup$ @theSongbird I agree that worked examples can be useful. But that's not within the remit of our site: we're a question and answer site, not a worked examples site. $\endgroup$
    – David Richerby
    Aug 6, 2017 at 12:15
  • $\begingroup$ Would a separate work/answer site do the trick in keeping the types of question brought by you out the scope of the main site? $\endgroup$
    – theSongbird
    Aug 6, 2017 at 12:46
  • $\begingroup$ Sorry for the tautology but it'd work if it worked. :-) If it became as well-known and widely used as this site then, sure; if it didn't, then it probably wouldn't have much impact. I should emphasize, though, that it's not really a problem that people post "Can you check my answer" questions here. We usually don't answer them but we do encourage the asker to identify a specific area they're uncertain about and ask questions about that. It's not a problem if a question gets closed, edited and reopened. $\endgroup$
    – David Richerby
    Aug 6, 2017 at 12:53
  • $\begingroup$ No need to apologize, I enjoyed thy pun. ;) Although, if we have many "have I solved it right" questions, doesn't it mean that three is quite a high demand for answers? (hope I'm not too annoying in asking this) $\endgroup$
    – theSongbird
    Aug 6, 2017 at 13:07
  • $\begingroup$ There's a reasonable demand for answers to "have I solved it right". But I'm not sure if that would translate into a successful site on their own: they're a definite minority of questions here (less than 10%, I'd say). They also tend not to get answers before being closed, whereas a "please do my homework for me" question is more likely to get an answer. Here's the thing: grading people's homework is pretty boring. I help people here with CS because I enjoy doing it; I grade people's homework at work because it's one of my responsibilities. I dont want to make that responsibility a hobby. :-) $\endgroup$
    – David Richerby
    Aug 6, 2017 at 13:46

You must log in to answer this question.

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