8
\$\begingroup\$

All too often I look at questions proposed here and I think: "The proposer clearly doesn't have a running solution to this question."

Should we make it an expectation that before posting a question, whether code-golf or code-challenge, the poster should have running code that implements the problem. While I don't think they should have to reveal their code at start, perhaps they should reveal what language their implementation is in, and how long it is.

I imagine for code-golf, it isn't important that their answer is golf'd. But, for code-challenge, especially ones with execution time or memory requirements, the poster's solution should conform.

Thoughts?

\$\endgroup\$
7
\$\begingroup\$

In a similar vein Joey has been pushing to have a tool to validate the results.

I regard both suggests in a positive light, and would strongly encourage people posting non-trivial questions to consider doing both.

That said, I've been trying to avoid too many hard and fast "You must jump through this hoop in order to post" type rules. We have pretty low traffic, and I hesitate to make it worse.

Granted, I think the quality of our questions is less than stellar and that may be contributing to the dearth of traffic.

Please feel free to encourage others to have a working solution before posting. IN particular feel free to suggest that the effort of producing a working model would have allowed the user to identify some of the corner cases left unspecified.

\$\endgroup\$
  • 1
    \$\begingroup\$ Thanks for mentioning me :-). And completely agreed. Given that I currently prepare another CG contest at our uni I notice that having one or two reference solutions to tasks helps finding corner cases and parts of the spec that are unclear. And it eases generating test cases by a few orders of magnitude. \$\endgroup\$ – Joey Jun 13 '11 at 19:13

You must log in to answer this question.

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