Recently, I have seen several questions closed for the close reason

"When seeking help to debug/write/improve code always provide the desired behavior, a specific problem/error and the shortest code (as formatted text, not pictures) needed to reproduce it in the question body. Providing a clear problem statement and a code attempt helps others to help you."

when this does not seem appropriate to me. Specifically, these questions are not asking to resolve an error but rather asking how to do a thing at all. In my opinion, asking these users to write a code sample does not benefit anyone, because such a code sample necessarily can't be any better than

myData = [1, 2, 3];
myAnswer = /* I don't know what goes here */;

Here are some recent example questions. In each case, what is being asked is fairly clearly stated (in my opinion), there is no way for the author to provide a broken-code-sample that is not trivial in the fashion I described above — and an answer would not depend on any particular details that might have been omitted from the question by omitting code.

To be clear, I'm not asking for special treatment of Earth Engine questions (they're just the ones I care about); I intend to ask this as a Stack Exchange user who wants to make sure that questions are closed at the right times and for the right reasons, and askers are given clear and actionable feedback.

Perhaps what's going on here is that GIS Stack Exchange does not actually want "how do I do this" questions, but it seems odd to me that debugging questions would be allowed but not these, which could be of much broader applicability.

Perhaps what's going on is just that the close reason is badly worded: it says "debug/write/improve code", but the requests it makes ("problem/error", "code to reproduce it", "attempt") are specific to debugging or code review questions.

  • “write” there does not necessarily mean write from scratch. Not disagreeing with you though. Will read again more carefully later. – Andre Silva Aug 25 '19 at 21:15

Would you be happier if they were closed "as unclear what you are asking" instead?

Basically in many of these questions there is insufficient detail to write an answer. But Google Earth Engine is not my area of expertise so may be they are answerable without any sign of effort by the asker.

  • 2
    All of these questions are unambiguously stated — they do not require guessing by the answerer (and therefore "Unclear what you are asking" doesn't fit). I agree that some of these questions could do with more “research effort” — but the thing we do with low-effort questions (short of e.g. "do my homework") is downvote them, no? – Kevin Reid Aug 26 '19 at 13:54

The questions you cite are not about errors but they are about how to write code.

For such questions we ask that a code attempt is always provided so that:

  • any potential answerers (who volunteer their time here) can see what has been tried and where the asker is stuck
  • the Q&A that results can be a learning exercise for the asker and all viewers alike rather than just a code request

The clear and actionable feedback is:

Providing a clear problem statement and a code attempt helps others to help you.

I think that our community welcomes "Doing X?" questions except when:

  • they show no research effort, and are often downvoted as a result
  • they are about writing code with no code attempt being presented, and are often close voted using the close reason that you are asking about.
  • 1
    Do you have a recommendation as to how one should "attempt" when one does not know what the shape of the solution is? For example, in the question which I described as about geocoding, what is an attempt at geocoding? Should it be a program that has coordinate input and text display but calls an undefined hypothetical function? – Kevin Reid Aug 25 '19 at 22:15
  • @KevinReid They say that they are already working on a GEE application and that they have the lat/long of their area so it sounds like they already have some code but are not presenting it. – PolyGeo Aug 25 '19 at 23:36
  • 2
    My perspective would be that the source of the lat/long is irrelevant — it is clutter that makes the question longer to read and/or less generally applicable — but fair enough. – Kevin Reid Aug 25 '19 at 23:50

You must log in to answer this question.

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