11
$\begingroup$

Recently I've been noticing a trend in puzzles, where they're either answered within 2 or 3 hours, or left to sit unanswered for days on end until a massive slew of hints is provided that finally points someone towards the answer. I find this to be an issue with how puzzles are presented on this site, and I was wondering if there's a solution.

This and this are examples of puzzles that were solved basically on the spot. Both of them strike me as pretty good puzzles, and I'm much less worried about this end of the spectrum since puzzling has the "many brilliant minds" effect. However, I think that if the site were populated with only puzzles like this, there wouldn't be as much allure, as either you solve the puzzle immediately and feel minimal reward, or someone else does.

On the other hand, this and this are examples of the other end of the spectrum, where they've been stewing for days and/or weeks waiting for someone to find the answer amidst a thick veil of obscurity. Multiple hints have been added, and yet still no one has made any progress. This type of puzzle is even more troubling, since if every puzzle was like this no one would ever get the satisfaction of solving anything.

I'm a firm believer that puzzles requiring 5 hints are probably just too hard (even though some of my favorite puzzles, and some of my puzzles as well, have well crossed that threshold). I like puzzles that are solvable on their own, and become moderately easy when 3 or so hints are provided. Is there a way to quality check puzzles to make them fit that guideline? Surely I can't be the only who ends up frustrated over a puzzle that sits unsolved for months.


P.S. At the end of my monologue, it became clear to me that too-easy puzzles aren't an issue here. Still, I like to be able to make puzzles that aren't too easy without being intensely difficult as well, and I'd like to find the threshold for that.

$\endgroup$
  • $\begingroup$ It certainly depends on the sort of puzzle - "fuzzy" puzzles (like riddles) tend to sit longer than "rigid" ones (like math puzzles). I tend to post leaning heavily into the second category and I've not yet found the "too hard" barrier there (but not for lack of trying) - I've seen a few questions of that category that hung around a while, but not many. $\endgroup$ – Milo Brandt Jul 22 '15 at 1:01
  • $\begingroup$ See this interesting comment that may be relevant to the conversation. $\endgroup$ – GentlePurpleRain Jul 22 '15 at 20:44
6
$\begingroup$

Right from the start we should differentiate between original hints and additional hints. The first are part of the puzzle and are rightfully so. They may or may not be in spoiler-tags, but they should belong to the puzzle as such.

The 2nd category of hints is the problem. I think such hints "happen" because a puzzle poster usually doesn't get a lot of feedback while people are "thinking" on the problem. If he then gets impatient, he adds a hint or another, or another... Which is generally bad. (And I'm possibly guilty of this as well...)

Hints are a bad tool to rise attention and up motivation for a puzzle.

So how to avoid this situation?

I think we - as a community - can help by using the commenting section more actively. In particular, if we are stuck with a puzzle for a while (but have been thinking about it), we should request hints.

Hints should only be added if requested.

A puzzle poster should not set out with the idea of "I'm going to add hints over time..." . Instead he should rather be patient and don't add hints. If he thinks the puzzle is too hard without hints, then it should potentially be posted in a different form right from the start. (So the hints are part of the original puzzle.)

Of course one may misjudge the difficulty of one owns puzzle easily - and hints are a valid way of adjusting the misjudged difficulty-level or reign in a puzzle which turned out to be too-broad.

Hints are fixes to a puzzle's difficulty. Ideally, you should not need them.

I think there should be some self-imposed rules for puzzle authors on hints. (And I should start applying them myself from now on ;c) )

  • Never start a puzzle which will automatically get a set of "added hints" over time. Any puzzle should be solvable by the first, initial posting. Hints are only patchwork fixes similar to edits. They might be needed, but they should not be part of the original plan.

  • Do not only consider readers "now" but also "later". A puzzle-post is a good puzzle if any newcomer in a year's time (or later) can find it, read it, attempt to solve it, and enjoy it. Good puzzles can be killed by too many hints, even if in spoiler tags. Maybe we should even close/delete those (as they do not serve any purpose in future.)

  • Give amble waiting time before deciding your puzzle is too hard. 24hrs are absolute minimum due to time-zones, but I would say 48hrs should be the minimum.

  • If a puzzle seems to be ignored/not worked on. Use comments to ask for feedback. (Anybody working on this?) If some motivation boost is needed: Offer a bounty

  • Only add hints, if there is feedback (comments, partial answers etc.) that the puzzle is too hard/needs guidance. Do not add them, just because time has passed with no reaction.


Having said this, I think as far as hints are concerned there is the golden rule: less is more

Some puzzle may linger for weeks if not months before being solved - IMHO, that is okay! Don't let impatience spoil your fun - not our fun as author and not your fun as solver. (Nothing can be more discouraging by working hours on a difficult puzzle, making good progress and then suddenly see 'unwanted' hints which spoil it for you...)

$\endgroup$
  • $\begingroup$ These are all good points, and I agree with much, if not all, of this. I think one of the reasons (again, I'm guilty of this) people post hints to puzzles is to "bump" them back up to the top of PSE, to make sure people are seeing them. Is there any way we can get around this? I know it's disheartening to have a puzzle just stewing with no responses for a week, lost somewhere in the depths of the network. $\endgroup$ – Bailey M Jul 21 '15 at 13:37
  • 1
    $\begingroup$ @BaileyM Yes! This is exactly my point. Bumping up by hints is bad. The proper way would be to initiate a (modest) bounty instead. The 2nd best option would be to add a comment asking if anybody is still interested in this and hope for reactions. The 3rd best, but somewhat sneaky/mean, is an edit to question - just a shameless bumping up. (Don't do that too often or you'll get burnt!). Adding an unrequested hint is just a another way of this shameless bumping up but with additional ill-effects. $\endgroup$ – BmyGuest Jul 21 '15 at 13:41
  • $\begingroup$ @BalleyM Also, while you're not in control of it, many users (like me) sort the thread by "unanswered posts", so having your puzzle linger for a while does not necessarily mean it get's ignored... It all boils down to patience a bit. Don't expect that everybody is "jumping" onto your puzzle just because you are interested in it at the moment. You will have more joy (in the end) if people honestly solved your puzzle and weren't just guided through. $\endgroup$ – BmyGuest Jul 21 '15 at 13:44
  • $\begingroup$ @BmyGuest Are the actions that "bump" a question laid out explicitly somewhere? I ask because I did exactly this behavior for one of my recent questions. I asked it a few weeks ago, then posted a hint a couple days ago to bump it back up the queue. However, I also needed to post feedback to someone's answer (letting them know it was incorrect), but I'm not sure if commenting on an answer bumps the question. If it does, then my hint probably was not necessary (the answer I responded to was very close). If not, though, I can certainly understand why people add hints to bump their question. $\endgroup$ – VictorHenry Jul 21 '15 at 17:03
  • $\begingroup$ @VictorHenry responding to a question in a comment doesn't bump the question. $\endgroup$ – Bailey M Jul 21 '15 at 18:20
  • $\begingroup$ Another thing to consider is the difference between clarifications and hints. For instance, retroactively clarifying something like "this puzzle requires no outside knowledge of chess" on a puzzle related to a chess board is probably a good thing to know you can do, without leaving explicit hints like "study the placement of the rooks on the board in relation to the bishops". $\endgroup$ – Bailey M Jul 21 '15 at 18:21
  • $\begingroup$ Thank you for writing this up! This is everything I've wanted to say about hints and more. $\endgroup$ – xnor Jul 22 '15 at 23:05
2
$\begingroup$

I think, unfortunately, that any puzzle that requires more than a few hours to be solved on this site is probably too hard.

As has been mentioned previously (and in the OP above), there are many brilliant people looking at these puzzles, and one of them is bound to have the leap of intuition required to find the key to unlocking any given puzzle.

If we were to take the best of our puzzles here that were solved in 4-24 hours, and put them into a book, and give them to any average individual, I expect that individual would find them extremely challenging, and would take much longer to solve each individual puzzle.

I have found myself disappointed when I design what I think is a very clever puzzle, and it gets solved within a couple of hours. My inclination then is to make the next puzzle "harder", so it won't be so "easily" solved. I think this inclination is wrong, and harmful to the success of this site. Just because a puzzle is solved quickly by brilliant minds, doesn't mean it is not a good puzzle.

When a puzzle is made "hard" enough that it can't be solved in a reasonable amount of time by all the brilliant minds on this site, that means it isn't "hard"; it's lacking enough information to be solved. Sometimes, someone takes a great leap of intuition and manages to grasp a few straws that get constructed into a feasible answer, but for the most part, the puzzle goes unsolved until further hints are provided.

This makes the puzzle creator feel good: "I'm so smart; I fooled all these people!", but in reality, they're not smart; they're just not providing enough information for the puzzle to be solved.

Realistically, I think the solution is for us to get over ourselves, and stop worrying about how quickly our puzzles will be solved. Put them out there with enough information and hints to be readily solvable, and expect that they will be solved quickly.

If you look at some of the nominations for best puzzles of the last year and a half, most of them did not take a long time to solve, but were well-constructed. They had engaging stories/text, and well-thought-out puzzles that were obviously carefully constructed.

That's what we should be aiming for when we create puzzles: Well-though-out, carefully-designed puzzles that can be reasonably solved by people on this site within a few hours (for the most part). There is the occasional excellent puzzle that still requires a long time to solve, but it's the exception rather than the rule.

I think we should also try to point all users who are creating puzzles to posts like

Especially if a user has never created a puzzle before, this can help weed out a number of bad puzzles before they even get posted.

I don't know if it's possible, but maybe providing one or more of those links right on the "Ask a question" page would help those who are posting their puzzles to think about the quality before actually posting.

This has been a bit of a rambling post. The tl;dr is that good puzzles aren't (necessarily) hard puzzles, and we need to get over ourselves and focus on creating good puzzles instead of trying to stump the Puzzling community.

$\endgroup$
  • 1
    $\begingroup$ An interesting way to appease both sides is with puzzles with multiple steps. If each step takes ~24 hours to solve, and you have five steps to your puzzle, not only does the puzzle takes five full days to solve (which is a lot!), but also many different people get the satisfaction of solving something. This does lead to the issue of only being able to accept one correct answer, but it's definitely better than intentionally leaving out important information just so your puzzle isn't solved in 20 minutes. $\endgroup$ – Bailey M Jul 22 '15 at 15:43

You must log in to answer this question.

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