8
\$\begingroup\$

Please note: the intended tone of this question is an "Are you sure? (Y/N)". It doesn't have the flavor "do anything differently".


The problem emerged with this question, having currently at least one deleted answer(s) (what is invisible for users below 2000).

The most relevant win criteria of the puzzle is the algorithmical complexity of the solution.

An answer using Mathematica was posted, making the algorithmical complexity of the solution depending heavily on the internal algorithms used by this software. These are not enough well known, and practically unknowable, because Mathematica is a closed-source software.

However, the solution seemed a quite interesting one for me, and maybe also for others (it got multiple ups).

Accepting the situation, the poster declared the solution as non-competing.

Despite that, it was deleted by community votes.

I think, also non-competing or undecidable answers improve the site. However, it would be probably not useful to have a large mass of non-competing, unanswerable answers. Furthermore, the primary goal of the site may be to collect mainly competing answers to the puzzles.

\$\endgroup\$
17
\$\begingroup\$

Yes

Non-competing doesn't allow an answer to do what it wants, and not meet the questions specifications. If a question has a specific scoring system, answers whose scores can't be calculated shouldn't be marked non-competing, they should be removed from the question.

In short, if I posted an answer to the Hello, World! question that didn't display Hello, World!, is it a valid answer? No. Can it be marked non-competing? No. To quote Dennis

Marking an answer as non-competing does not exempt it from being invalid.

The Mathematica answer, while being interesting, did not meet the answer specification, and, per our rules it should have been deleted. This also applies to any other answers which, while accomplishing the task at hand, can't be scored.

\$\endgroup\$
  • 1
    \$\begingroup\$ What if it actually prints hello world, and it may fulfill the criteria, but nobody can check it for sure? \$\endgroup\$ – peterh - Reinstate Monica Sep 24 '17 at 16:14
  • \$\begingroup\$ @peterh what criteria? \$\endgroup\$ – caird coinheringaahing Sep 24 '17 at 16:14
  • 1
    \$\begingroup\$ The win criteria of the question. In this specific case, the algorithmic complexity of the solution couldn't be determined, although it solved the problem. \$\endgroup\$ – peterh - Reinstate Monica Sep 24 '17 at 16:15
  • 5
    \$\begingroup\$ @peterh if it can't provably complete the task, and can't have an objective, determinable score, it isn't a valid answer \$\endgroup\$ – caird coinheringaahing Sep 24 '17 at 16:19
  • 4
    \$\begingroup\$ A better analogy would be a 'Hello, World!' program where for some reason you can't possibly determine the byte count (think of a tokenized language like TI-BASIC, but then one that does not reveal how many bytes each token is). It performs the task, but you cannot determine the score. \$\endgroup\$ – Sanchises Sep 25 '17 at 13:35
  • \$\begingroup\$ Of course, if the question explicitly allows non-competing answers, they should be allowed per this meta consensus. \$\endgroup\$ – JungHwan Min Oct 4 '17 at 2:31

You must log in to answer this question.

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