12
$\begingroup$

I know my proposal is extreme! Please read it as objectively as you can and imagine the community that would be created instead of dismissing it purely for being extreme. Additionally I wrote the post in a bit of a roundabout way exactly for this reason, if it's too long to read, just jump to the proposal below.

The context

So, I was just thinking about the quality problem this site has, and wanted to post an answer to How do we improve the quality of content produced on Puzzling.SE?, but my thoughts started to deviate so far away from the original question that I decided to post it as a separate post so that it wouldn't clutter that one.

My thought process

My thought process went a bit like this:

  1. How can one objectively draw a line between good questions and bad questions?
  2. Oh wait, this is only an issue with puzzles, everybody agrees questions about making puzzles are great.
  3. Either way, drawing an objective line seems impossible
  4. What if we would ban all those questions then... and change them into questions about making puzzles

So, the proposal

Any current puzzle question can actually be quite easily be reworded and reworked into questions focused on the puzzle solving and puzzle making process rather than the answer to the puzzle. In other words, it would become somewhat of a puzzle review Q&A site instead of just the puzzle solving forum/chatbox it is now. Now, as everybody loves the puzzle solving element I definitely wouldn't propose taking it out, I would just propose that instead of stating

The answer is this and this

questions would be reworded to get answers like

I believe this and this in your puzzle implies this (the answer the answerer believes was intended), however this might confuse puzzle solvers and give answers like

  • Exotic answer 1
  • Exotic answer 2

Personally I would advice rewording the puzzle as such:

[...]

The great thing about this is that this allows drawing a hard objective line between what a good and a bad question is. The entire rule change can be reflected with a one liner like "All questions should be either about the puzzle making or puzzle solving process".

Example list of on topic (NOT necessarily good) questions

  • In this password puzzle I made, how can I determine whether there is just one unique solution?
  • I made this mathematical puzzle, however I am unhappy with the transparency of the math parts. How can I hide the mathematics better?
  • I wrote this lateral thinking puzzle, however I wish to word it in such a way that "this and this answer" will be given over "this and this answer". (e.g. top two answers on https://puzzling.stackexchange.com/questions/3911/a-man-pushes-his-car)
  • I came across this Sudoku puzzle, however I am stuck at this and this point. I tried this and this, but I am not getting anywhere. How to approach solving this puzzle.
  • I found this this word puzzle including this unexplained solution. How can I translate this puzzle into a mathematical set of solvable equations.

The problems this solves

  • Answer quality Answer quality necessarily goes up
  • Too localized Questions stop being 'too localized', fitting the SE mentality where a question is not only valuable for the person asking the question, but also for future readers.
  • Unique puzzling.SE would find an unique place in the internet puzzling community (a lot of users have been complaining that it now is nothing more than another Yahoo Answer sub section).
  • Question quality Questions about making puzzles would become more popular again, as that aspect would be highlighted more.
  • SE network It would make the site fit well within the mentality of the SE network. Whether this has advantages for the site is hard to figure out, but at the very least it decreases the chances of the site being shut down and increases the chances of the site graduating.

How to achieve this

Now, I am far from sure that people would be actually in favor of this proposal, but just in case people will object that it's not realistic let me outline how this could be achieved

  • (one time) First of all all unpopular questions would be outright deleted
  • (one time) Salvageable questions would be edited
  • (only during beta) Questions that are bad get deleted
  • (after beta) Questions that are bad get downvoted (by that time the community has hopefully figured out the new model)

Additionally it has been done before. Check the programmers.SE history for example.

Last thoughts

Now, I don't think I am perfect and this question is just the bare base on which a lot more work would need to be done, however I do think that I would definitely love being part of the kind of community this would create. Yes, a lot of users would be lost, but all experts would probably stay and a substantial group of the current users might just make the jump to the 'expert level'.

Oh and, do feel more than free to edit this question. Removing ambiguities, improving the language and improving the sample questions would be all much appreciated.

$\endgroup$
  • 2
    $\begingroup$ So love the fact I got a downvote within a minute before anybody could have had the time to read this. Go puzzling.SE... $\endgroup$ – David Mulder Nov 19 '14 at 0:20
  • 1
    $\begingroup$ This should in principle be an answer to the earlier discussion on the topic, but given how much the site has changed in the last few weeks, I guess it's better to start a fresh discussion. $\endgroup$ – Gilles 'SO- stop being evil' Nov 19 '14 at 0:34
  • $\begingroup$ @Gilles: I am well aware there are a lot of (unenforced) meta discussions hanging around, my reason for posting this separately was that I was trying to propose a framework around which a solution could be build for all problems, rather than solving one problem at a time which for one reason or another clearly isn't working. $\endgroup$ – David Mulder Nov 19 '14 at 0:36
  • 3
    $\begingroup$ The way I might see this working is a little different than posted, though I don't necessarily disagree with the content of your post. In short, we would want to create a single descriptive close reason, and close every question that fits those criteria. If those questions can be reworded, then they would be reopened. Otherwise, we don't need to delete them; the auto-delete scripts will take care of what needs to be taken care of, and the close markers will say the rest. $\endgroup$ – Aza Nov 19 '14 at 0:41
  • $\begingroup$ @Emrakul: As far as I care, feel free to edit that stuff if you think it would work better that way. It's just that answers can't be closed, so that's why I was thinking of deleting. Over on skeptics.SO IMHO there is a huge problem where the moderators are deleting a lot of answers (despite these answers not even being necessarily bad per meta voting) and my point was mainly that that's fine during beta, but let's not start doing that here after beta. $\endgroup$ – David Mulder Nov 19 '14 at 0:47
  • 1
    $\begingroup$ My issue with this is that it drops the PPCG model of challenges - which people find fun. I don't totally disagree and I think it's in good spirit. But, I think this goes a bit too harshly against what the bulk of the community here wants. $\endgroup$ – d'alar'cop Nov 19 '14 at 0:51
  • 3
    $\begingroup$ @d'alar'cop: I know, there is just this one nagging problem: It has become excessively clear that it is not possible to get the level of quality as high as on PPCG.SE for the simple reason that code by it's very nature narrows down questions a lot. We on the other hand have to work with entirely open questions which has resulted in a huge quality problem. Mind you, if you are able to think of a set of rules that do allow high quality PPCG.se-like puzzles, yet disallow everything else I would love to hear it. $\endgroup$ – David Mulder Nov 19 '14 at 0:56
  • 2
    $\begingroup$ @DavidMulder Actually I've been thinking about this, and discussing with Emrakul too. I think it can be viable. What needs to happen is the aggressive discouraging of wild guesses which under no interpretation fit the clues. We can help with this by also aggressively encouraging full explanations of answers. Frankly, if we can do that, I think we've got it sorted. I've noticed that people do follow trends. For example, people started titling better recently. People have started using a standard explanation style. People are getting the message. I love this SE and I want to see this sorted. $\endgroup$ – d'alar'cop Nov 19 '14 at 1:00
  • 3
    $\begingroup$ I don't think it's fair to compare to PPCG, where getting a solution isn't usually the end (since you can always golf it down more/make it better). I like the idea of getting people to point out how puzzles could be better, but we'd have to be careful of low-quality brainteasers disguised by titles like "I just made this puzzle - opinions please?". $\endgroup$ – Sp3000 Nov 19 '14 at 1:06
  • 2
    $\begingroup$ @Sp3000 We could look at codereview.SO for inspiration for that, look at the rules they set up, etc. Maybe even outright ask them for advice on their meta. Still though, even "I just made this brainteaser, could you guys review it please?" will get better answers than "Solve this brainteaser". $\endgroup$ – David Mulder Nov 19 '14 at 1:08
  • 2
    $\begingroup$ @DavidMulder Is this the sort of thing you mean? I thought I'd try it out, if you don't mind :) $\endgroup$ – Sp3000 Nov 19 '14 at 1:41
  • 1
    $\begingroup$ @Sp3000: Well, seeing you just got one of the best answers on puzzling.SE and an incredible up vote ratio (10 upvotes on 44 views) I would consider that quite the success :D . $\endgroup$ – David Mulder Nov 19 '14 at 4:50
  • 1
    $\begingroup$ "I know my proposal is extreme!" - I don't think it is all that extreme. We should seriously consider it IMO. That's the sort of thing I was trying to do here: puzzling.stackexchange.com/questions/4977/… $\endgroup$ – A E Nov 19 '14 at 10:25
  • 1
    $\begingroup$ @DavidMulder I proposed here that perhaps only puzzles with a mathematical abstraction be allowed. $\endgroup$ – Joe Z. Nov 19 '14 at 19:18
  • 1
    $\begingroup$ I get the impression I'd be one of the people lost. I come here to view and solve novel puzzles/brainteasers. The questions I post are invariably of the "here is my puzzle; solve it" variety. I've supplied advice on a few occasions to puzzlers looking to come up with a novel game/puzzle, but it isn't an exaggeration to say that 95% of all questions on the site aren't of that format and can't be bludgeoned into it. Or perhaps they can, but only in such a way that any "here it is; solve it" puzzle can be similarly bludgeoned, in which case the process is moot. $\endgroup$ – COTO Nov 22 '14 at 13:52
9
$\begingroup$

I think what we really need is to edit the questions we have to make them better and ultimately closing the ones that are not up to the right standard. New users need to see lots of examples of good questions .

Also, there seems to be a culture of putting guesses as answers. This should be discouraged and people should use the comments for that.

Also, a sandbox where new questions need to sit for a couple of days before being posted is a great way to get feedback and to improve your question.

$\endgroup$
  • $\begingroup$ +1 for the idea about a sandbox. PPCG uses question sandboxes to great effect. $\endgroup$ – Kevin - Reinstate Monica Nov 19 '14 at 22:44
  • 1
    $\begingroup$ A difficulty that Puzzling has and not PPCG is that a code golf problem can be judged largely by reading it. But, a puzzle must be judged by solving it, or at least by knowing the solution. Still, a Sandbox seems like a good idea. $\endgroup$ – xnor Nov 20 '14 at 0:46
  • $\begingroup$ Some good suggestions, but doesn't really address the OP's question. Are you suggesting we make these changes but keep the "solve my puzzle and that's it" questions? $\endgroup$ – Julian Rosen Nov 20 '14 at 20:47
  • $\begingroup$ @JulianRosen I am not 100% sure what a "solve my puzzle and that's it" question looks like but math.stackexchange.com/questions/205953/… for example, should be on topic. Is that what you had in mind? $\endgroup$ – Lembik Nov 20 '14 at 21:04
  • $\begingroup$ Yes, this was what I had in mind. I interpret the OP's proposal as labeling this question off-topic, so it sounds like you're in favor of being more permissive (for the moment I'm keeping an open mind) $\endgroup$ – Julian Rosen Nov 20 '14 at 21:40
  • $\begingroup$ I like the 'Sandbox' idea for a "stage the puzzle" step - is it technically doable? If so, I would propose... (well, I propose below as an answer to vote on.) $\endgroup$ – BmyGuest Dec 4 '14 at 7:18
8
$\begingroup$

A proposal to vote on:

Perhaps we could ban the questions that ask "what is the answer to this riddle/problem" but keep questions that ask "how is this riddle solved".

For example, difficult maths questions are often on topic on maths stackexchange, but this is because the answers don't just state the solution, but how the solution is obtained.

Therefore I believe questions that require working out, or require the answerer to show their methodology for solving should be on topic, because seeing different methods for solving enhances our ability to understand the art of puzzle solving.

$\endgroup$
  • 2
    $\begingroup$ Yep, that's definitely what I was thinking of when I said that all questions should be about the "puzzle making or puzzle solving process" :) . "How is this riddle solved" would in my book be a poor, but on-topic question. $\endgroup$ – David Mulder Nov 19 '14 at 1:10
  • 2
    $\begingroup$ I believe what you're aiming at is actually very similar to what has been implemented on multiple sites across the network: if you have a question about how to solve a problem, show us how you got there, and tell us specifically what you need help with. $\endgroup$ – Aza Nov 19 '14 at 1:13
  • 3
    $\begingroup$ I'd just like to say that it's really the riddles/challenges that are more like playground jokes that are problematic regarding quality (they are the ones that are often cited as being low quality). $\endgroup$ – d'alar'cop Nov 19 '14 at 1:32
  • $\begingroup$ An issue is that it's not clear whether a downvote means the rules should be more permissive, or less. $\endgroup$ – Julian Rosen Nov 19 '14 at 1:42
  • 3
    $\begingroup$ Mew, it sounds like your proposal isn't really saying that questions should be changed. It's really saying that answerers should stop providing one-line answers and fully explain how they arrived at their answer. $\endgroup$ – pacoverflow Nov 19 '14 at 3:19
  • 1
    $\begingroup$ I'm really unclear on the proposal that's being being voted on. It seems to be up to answers to say how they got solution. $\endgroup$ – xnor Nov 19 '14 at 8:31
  • $\begingroup$ @xnor: If you agree with the proposal outlined in the opening post, vote on that, if you disagree, write an answer and let users vote for that. Not sure either what voting on these answers would additionally mean :) $\endgroup$ – David Mulder Nov 20 '14 at 1:47
  • 1
    $\begingroup$ I agree that an answer without an explanation should not get upvotes or be accepted and normally someone should tell the answerer that explanations are expected on puzzling.SE. $\endgroup$ – Lembik Nov 20 '14 at 7:49
  • $\begingroup$ I'm in favour of this. They way I read it is: Encourage puzzle-posters to phrase their question in a way which makes it clear, answers should provide the path to the solution, not the solution only. (Where appropriate.) I don't see any harm in this, and it could make a very good step forward in quality. $\endgroup$ – BmyGuest Dec 4 '14 at 7:15
1
$\begingroup$

I disagree. Your question is well-asked and deserves upvotes, but the concensus is not with you. Instead, I propose that we take the core of your idea, suggesting improvements to puzzles, and add them into the culture of question answering. Giving the right answer is insufficient, giving the answer with reasoning is good, and giving an answer with reasoning and good suggestions of improvements is awesome. That way, both sides of the challenge debate can be happy with everything on the site.

$\endgroup$
  • $\begingroup$ You might want to check stackoverflow.com/help/whats-meta second paragraph of "Voting is different on meta." . And regarding consensus, I wholeheartedly agree that right now there are simply not enough votes to determine what the community wants and believes. If you want to propose an alternative implementation I would however suggest wording it more fully then this (either here or seperatedly). $\endgroup$ – David Mulder Dec 4 '14 at 5:05
  • $\begingroup$ I edited my answer so that you can change your vote if you so wish to. $\endgroup$ – David Mulder Dec 4 '14 at 5:09

You must log in to answer this question.

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