19
$\begingroup$

I see a concerning trend where puzzles get upvoted and downvoted on first impressions before any solution attempt is posted. And, if a solution is posted that makes the votes seem unwarranted, there's not enough votes after that to swing it the other way, and by then it's too late to for the votes to have an impact.

It's often hard to tell if a puzzle is good from its statement because so much depends on the solution. It might be clever or tedious, well-fitting or arbitrary, satisfying or disappointing. A good puzzle is enjoyable to solve, not just to read. By the very nature of puzzles, the crucial element, the solution, is hidden. Yes, it is important that the puzzle be intriguing and be well-written and seem like it would have an interesting solution, but voting only on that is judging a book by its cover.

This is a particular problem for enigmatic or cryptic puzzles, whose surface content is liable to look like gibberish.

I believe this causes a number of problems:

Searchability: It's harder for future readers to decide if a puzzle is probably good. The long-term goal of the site is to make a well-organized repository of high-quality puzzles, so it's important that well-received puzzles stand out.

Feedback: Puzzle-writers get worse feedback via votes, making it harder to judge what people liked. A new-ish user who posts a great puzzle that looks like nonsense might get discouraged with the flood of rash downvotes and leave. A writer who writes a flawed puzzle that looks pretty might not get the message that it's a poor puzzle because of all the upvotes.

Reward: Votes give encouragement for puzzle-writers to make good puzzles. But, skewing the reward towards surface appearances might condition puzzle-writers to post shallow but immediately-appealing puzzles over deep but satisfying ones.

(A related problem is that votes are largely proportional to views, and a well-received +10/-0 looks just like a poorly-received +25/-15 one to anyone without 750 rep.)

In part this is a technological problem: once you vote, you can't change it unless the puzzle is edited. So, those whose minds were changed by a solution can't change their votes. I sometimes refrain from vote to wait for a solution, but then I forget the post and never vote.

Is this indeed a problem? Is there something we can do about it?

$\endgroup$
  • 2
    $\begingroup$ This is, in my opinion, merely a side effect of the real problem that it is difficult to judge the quality of a puzzle that has not yet been solved. I think it may be more productive to first find a solution to the underlying problem, which may well solve this problem too. $\endgroup$ – Doorknob Mar 29 '15 at 23:05
  • 3
    $\begingroup$ Here are two fairly extreme examples: a puzzle that started off with loads of DVs and then got loads of UVs after being solved (admittedly it was edited a lot) and a puzzle that started off with loads of UVs and then got loads of DVs after an answer was accepted (an accepted answer with a score of -21 !!!) $\endgroup$ – Rand al'Thor Mar 29 '15 at 23:39
  • 3
    $\begingroup$ One idea would be to discourage voting until a puzzle is solved, but this might not be something we want to do. $\endgroup$ – Rand al'Thor Mar 29 '15 at 23:39
  • $\begingroup$ @randal'thor, those seem (to me) like examples of where the current system has actually worked well - in the end. I'm more concerned about questions where that revision of votes doesn't take place but maybe should do. $\endgroup$ – A E Apr 1 '15 at 11:44
  • $\begingroup$ My puzzling.stackexchange.com/questions/24030/… is now [On Hold] so it is going to be closed soon. It is marked like that just after 1 person answered. How can I expect from other visitors now? I changed and edited several times but it is still [On Hold].... I guess that's not stupid puzzle that much to be said "too many possible answers, or good answers would be too long for this format. Please add details to narrow the answer" Have a look! :'( $\endgroup$ – Nai Nov 19 '15 at 11:15
6
$\begingroup$

I agree, and I think it would be helpful if one were able to change one's vote on a question without the question having been edited.

Because sometimes it turns out that I haven't properly understood a puzzle until I see the answer.

$\endgroup$
  • 3
    $\begingroup$ Mods, is there any chance that this could be enabled on Puzzling? $\endgroup$ – xnor Apr 2 '15 at 21:28
  • $\begingroup$ I think it would help a little, but is this really our best solution? I hope not. $\endgroup$ – JLee Apr 2 '15 at 21:35
  • $\begingroup$ @xnor It's unlikely. SE rarely makes custom modifications to beta sites. $\endgroup$ – Aza Apr 4 '15 at 5:13
  • $\begingroup$ We could decide to permit making a 'null edit' to a question just to allow change of votes. No technical change needed. $\endgroup$ – A E Apr 4 '15 at 7:43
  • $\begingroup$ @AE I'd find it awkward to do so to change to a downvote since it makes it basically announces that I did so. $\endgroup$ – xnor Apr 5 '15 at 10:11
  • $\begingroup$ @xnor If someone else edited it though, you could change your vote. $\endgroup$ – A E Apr 5 '15 at 11:11

You must log in to answer this question.

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