-4
$\begingroup$

Today I was given a link to an example of a good Affix riddle, and I agree that it is good. However, it was solved in 10 minutes, so maybe it is a bit easy?

But the puzzle's difficulty is not my point.

My point is that the puzzle is a perfect example of what typically happens here on puzzling se, and which, in my opinion, needs to be addressed and corrected if possible (as much as it can be) in order to provide good and fair incentive to puzzle creators to create and post quality puzzles.

What occurs:
A great puzzle is posted and solved quickly. Then people read the puzzle, like it, then read the answer and upvote that answer, either forgetting to upvote the puzzle, or subconsciously thinking that their upvote of the answer is an upvote for the puzzle or covers both puzzle and answer. This dynamic obviously favors answerers significantly more than creators/askers, yet the former couldn't exist without the latter. Look at the example link: the answer has nearly double the upvotes as the puzzle itself. I doubt many of those voters would hesitate to upvote the puzzle also, but people just don't think to do both.

In other words, near 100% of the answer upvoters (for the example above and most other questions) would not hesitate to say yes to a prompt asking if they want to upvote the puzzle also. Therefore, we have the rep disparity i am talking about.

My idea for a solution...

Before giving my idea, let me be clear that there are definitely both pros and cons to this idea, but I think its existence would be better than what we have now by equalizing question/answer upvotes and promoting better quality puzzles overall.

The idea:
My idea is that downvote functionality remains unchanged, but that whenever any answer on a question/puzzle is upvoted, the puzzle automatically gets an upvote also (1 max upvote per user per puzzle, of course.) (Actually, most ideal is a yes/no prompt asking if the voter wants to also upvote the puzzle. In this ideal case, I see only pros and no cons.)

However, assuming that that "most ideal" fix cannot happen, I will address the "automatic" idea.

Even in the case where it is automatic, the only issue is that sometimes the puzzle is not that good, so one doesn't want to upvote it, but does want to upvote the answer. However, I'd argue that if this idea was implemented, the result would cause what SHOULD already happen to actually happen, which is that a bad puzzle should not be answered, so that the creator is not encouraged to make more like it.

The net effect would be better puzzles and more fair voting of both questions and answers.

Comments, critiques and other ideas are welcome and encouraged.

Edit: Here is a common example of how question upvotes can be forgotten: this new pacman-themed puzzle looks very cool, and i want to upvote it, but only after i see the answer and make sure it's good. Others act this same way. This practice, while great in my opinion, contributes to the disparity also, no explanation needed.

$\endgroup$
27
  • $\begingroup$ Of course, for some puzzles, one does not know how good or bad they are until the intended answer is made known. $\endgroup$
    – JLee
    Commented Jun 14, 2022 at 15:01
  • $\begingroup$ Lol. Care to say why, downvoter? $\endgroup$
    – JLee
    Commented Jun 14, 2022 at 17:51
  • 5
    $\begingroup$ (not the downvoter) - I think the automatic idea may face the same problem as the ideal one; to implement this, you would actually have to make changes to the core behavior of PSE (rather than organizational/community) - This is not an easy feat in the sense that mods can't just customize PSE to their liking, and those who could are generally opposed because they want the SE network to work more or less the same everywhere. $\endgroup$ Commented Jun 14, 2022 at 18:15
  • $\begingroup$ @LukasRotter good point, but even if there is no immediate solution, i think it's helpful to post this to define the issue. The power of dialogue is awesome to iron out bugs and present ideas and cause aha moments. $\endgroup$
    – JLee
    Commented Jun 14, 2022 at 18:27
  • $\begingroup$ There's already a prompt for new users to upvote questions, afaik. I think this suggestion goes too far and would clash across all the other network sites $\endgroup$ Commented Jun 14, 2022 at 19:15
  • $\begingroup$ @PuzzlingFerret If there is a prompt, if it is not question-specific i don't see how it helps. Not liking my solution suggestion is fine, but agreeing that this is an issue is an entirely different thing. $\endgroup$
    – JLee
    Commented Jun 14, 2022 at 19:44
  • $\begingroup$ A downvote on this meta says to me that the downvoter thinks that this isn't an issue, and therefore is unworthy of discussion. If you see it as an issue, but instead have a better idea, or will admit that you don't, or at least not yet, then that is a different thing. $\endgroup$
    – JLee
    Commented Jun 14, 2022 at 21:02
  • 3
    $\begingroup$ Do you have any evidence that your perceived problem really exists (that people are actually forgetting to upvote a question after upvoting an answer). I upvote (and occasionally downvote) questions and answers independently, based on their merits. Just because an answer is worthy of an upvote, doesn't mean the question is, and vice-versa. There are plenty of well-written, comprehensive answers to distinctly mediocre questions. I would strongly object to my upvotes of those answers being taken as implicit approval for the question. $\endgroup$
    – fljx
    Commented Jun 28, 2022 at 13:39
  • $\begingroup$ @fljx Would you also object to a pop-up when you upvote an answer asking if you'd also like to upvote the question/puzzle? My evidence is from observation of the posts on the site, both mine and others. My evidence is also from knowing human nature. $\endgroup$
    – JLee
    Commented Jun 28, 2022 at 13:49
  • $\begingroup$ @JLee like this one? meta.stackexchange.com/a/89059/175002 $\endgroup$ Commented Jun 28, 2022 at 14:47
  • 5
    $\begingroup$ @JLee Yes, I would object to a pop-up every time I upvote an answer. That's really poor UI/UX. If I wanted to upvote the question I would have done so already, before scrolling down to read the answers. Or, (rarely) if the answers reveal that the question has hidden depths that weren't obvious on the surface, I'll scroll back up to give it an upvote. A nag-box is not the answer. $\endgroup$
    – fljx
    Commented Jun 29, 2022 at 8:37
  • 4
    $\begingroup$ @JLee The simplest answer is leave things as they are. Is there an imbalance between question and answer voting? Maybe. But even if there is, your assumption that it is caused solely by people forgetting to upvote a question after upvoting an answer is impossible to verify. $\endgroup$
    – fljx
    Commented Jun 29, 2022 at 9:59
  • 2
    $\begingroup$ @JLee I think your assumptions in your question are wrong, and your understanding of how and why people currently vote is fundamentally flawed (i.e. "either forgetting to upvote the puzzle, or subconsciously thinking that their upvote of the answer is an upvote for the puzzle or covers both puzzle and answer." where is your evidence?). And the conclusions you draw are opaque (i.e. "there will be many more great puzzles and fewer "bad" puzzles" how are you coming to this?). If anything I would deliberately downvote on questions more often under this system to 'counteract' the autovote. $\endgroup$ Commented Jun 29, 2022 at 14:40
  • 1
    $\begingroup$ @JLee I have read and re-read your post trying to understand the leaps of logic your are calling 'intuition' and cannot see any critical thinking. Here are some places where you make assumptions: "I doubt many of those voters would hesitate to upvote the puzzle also, but people just don't think to do both.", "near 100% of the answer upvoters would not hesitate to say yes to a prompt asking if they want to upvote the puzzle also" $\endgroup$ Commented Jun 29, 2022 at 14:51
  • 1
    $\begingroup$ More: "the only issue is that sometimes the puzzle is not that good, so one doesn't want to upvote it, but does want to upvote the answer ... a bad puzzle should not be answered, so that the creator is not encouraged to make more like it." Bad puzzles should be answered so long as they're not close-worthy! Also, why would people not answer questions if someone voting on their answer would upvote the question? $\endgroup$ Commented Jun 29, 2022 at 14:53

2 Answers 2

4
$\begingroup$

The automatic version of your fix encourages either upvoting puzzles that may be low quality, or not answering bad puzzles as you mentioned.

Sure, some puzzles are bad, don't show any research, are simple or very hard code puzzles, but when you don't have much to solve, you might as well give them a try. And when that happens, if the solver writes a high-quality answer on a low-quality question, it's not that they shouldn't have answered, it's just that they solved a puzzle, like they would for any other on the site.

The "ideal" fix, to me, is not necessary. Good questions get rewarded with high votes, good answers on those questions, even fast answers are also rewarded high votes. I don't think users need a reminder to upvote the question, either they already did it, or they don't see it befitting an upvote.

Questions that see more upvotes than their accepted answer often show work put into them, they stay unsolved for some time and attract curiosity, and attention. Am I saying that the linked question doesn't show work put into it? Absolutely not. I'm not saying a puzzle solved in a short time doesn't have work put into it neither. However, the quicker you solve puzzles, the less you had to think about, the less you are puzzled, and by extension, the less time you got to spend doing one of your favorite activities, being puzzled, and searching for answers.

In summary: It shouldn't be about the upvotes or the reputation, it should be about enjoying puzzles, sharing puzzles, and being passionate about all that is puzzling.

$\endgroup$
9
  • $\begingroup$ So in a nutshell, you think the issue i brought up is not an issue. $\endgroup$
    – JLee
    Commented Jun 15, 2022 at 16:33
  • $\begingroup$ To me, if you feel like a puzzle is "worth more", you can share it, after votes, views are also a nice metric $\endgroup$
    – Auribouros
    Commented Jun 15, 2022 at 22:24
  • $\begingroup$ "but when you don't have much to solve, you might as well give them a try." Yes, and this happens mostly because there is much more rep incentive to answer than to create/ask, so the majority are waiting for others to ask, resulting in long lapses of sparse, sometimes low quality, puzzles. $\endgroup$
    – JLee
    Commented Jun 16, 2022 at 12:49
  • $\begingroup$ It shouldn't be about the upvotes or the reputation" yep, so by that logic it's probably better without rep at all. HOWEVER, it is there, and should therefore be as accurate as possible. $\endgroup$
    – JLee
    Commented Jun 16, 2022 at 13:03
  • $\begingroup$ I think there was a meta post saying "Reputation shows your familiarity with the subject, how appriciated you are by your peers..." and I think that's true. Stack Exchange has often been rewarding answers more, and that's fair for most sites. And even though this site is different, the behavior of the ones on it is too, puzzles (or questions in general) already receive more votes than on some other sites. I don't think reputation is a problem, and it shouldn't be either. Also as Stiv mentioned, that system would bring a lot of "mid-tier" puzzles to higher vote counts. $\endgroup$
    – Auribouros
    Commented Jun 16, 2022 at 13:11
  • $\begingroup$ Think deeper. If there were more good puzzles, people wouldnt feel compelled to mess with those bad ones, and they would definitely be more ignored. $\endgroup$
    – JLee
    Commented Jun 16, 2022 at 13:14
  • $\begingroup$ I like making puzzles, as well as looking and deciphering them as much as the next gal, but you also have to appreciate a well-structured answer, with a bit of author flair and a nicely placed joke at the end. Yeah, the puzzle was great and all, I gave it an upvote, but that answer is why I'm here, and even though the solving process might not have been thrilling or anything brand new, the added personality of the answer makes it worth the upvote. $\endgroup$
    – Auribouros
    Commented Jun 16, 2022 at 13:14
  • $\begingroup$ Yes, but answer upvotes are not an issue. Overall i mean. One can always cherrypick. $\endgroup$
    – JLee
    Commented Jun 16, 2022 at 13:15
  • $\begingroup$ As a side note, I have often accepted answers that I have not upvoted, some answers feel robotic, and even though they do show the gimmick of the puzzle and the answer, they feel like automated robotic answers, and don't give off the "community" vibe that Puzzling (or SE in general) wants to promote. $\endgroup$
    – Auribouros
    Commented Jun 16, 2022 at 13:15
3
$\begingroup$

Further to the points made by @Auribouros, this auto-fix would merely introduce another disparity:

If implemented, an answer could never score as much as a question. (So is that really fairer?)

Furthermore:

Questions with many highly-upvoted answers would end up with many, many more upvotes. This is another problem for quality, and a fairly big one, since this would give a heavy bias towards questions that end up on Stack Exchange's network HNQ's for technical algorithm-related reasons rather than what someone might consider true merit (therefore potentially generating yet more views and auto-upvotes), and those not-fully-thought-through puzzles (regularly, matchsticks-based) where multiple solvers post solutions not intended by the OP and are often rewarded by the community with upvotes for their creative thinking.

This latter point I feel is particularly key - here, far from being a disincentive to posting 'bad' puzzles, the OP of a poorly thought-through question would be rewarded by this auto-upvote system precisely for creating a poorly thought-through question, propped up by several creative answerers! Ultimately, truly excellent and highly upvoted puzzles would get lost in the noise generated by these multi-answer-auto-upvote questions; our site quality would be significantly diluted, and 'high upvotes' becomes a less useful metric for determining quality (albeit not a perfect system as it stands - I find it's normally more useful within a particular tag rather than across all puzzles overall).

Yes, I admit it is a shame when a good question does not garner the upvotes it probably deserves, but a 'bad' puzzle will usually always find a willing answerer, for whatever reason - whether for the thrill of answering a question, the possibility of rep, a keen new site user, or out of a completionist mindset (leaving no puzzle unsolved), among others. I don't believe this proposal would have the mindset shift you're really hoping for, and 'bad' puzzles would still be posted and answered. (Besides, sometimes setters only learn what makes a 'bad' puzzle - which can be subjective - by trying it out, posting something, and then receiving feedback...)

Plus, I feel there are many legitimate reasons why a question might not merit an upvote while an answer would - it is, after all, possible to write a very good, creative, upvote-worthy answer to a very bland, uninteresting question.

(For another example, think of 'puzzle fatigue', when a user repeatedly posts variations of the same puzzle - whether what-am-I riddles, affix-riddles, number sequences, or one of the many other fairly-easy-to-churn-out-another-one puzzle types - in a short space of time. The first one may be worth an upvote for creativity, but it's clear that people soon tire of this type of behaviour, feel that later posts show no real creativity above and beyond the first one, and begin to abstain from voting - or even downvote - to express this feeling...)

Conclusion:

It seems to me that all we can really do is keep promoting the notion that people should upvote good, useful, clever, or well thought-through puzzles. Other 'fixes' or tinkering behind the scenes is likely just to introduce different biases or create new unforeseen incentives for bad question design.

$\endgroup$
15
  • $\begingroup$ "If implemented, an answer could never score as much as a question. So is that really fairer?" Yes, IMO it is definitely fairer, especially if we want to maximize puzzle quality over time, instead of creating an environment that rewards point grabbing. As for, "Questions with many highly-upvoted answers would end up with many, many more upvotes." I respond with, "Yes!, That is the point. A question that has attracted all that attention IS WORTHY of those upvotes. If it wasn't interesting, then it wouldn't have all those answers." $\endgroup$
    – JLee
    Commented Jun 16, 2022 at 9:41
  • $\begingroup$ It is interesting that those playing devil's advocate usually don't mention the ideal solution that I mentioned, and if so, don't go into its merits and any drawbacks. Instead, they settle for, "It can't be done." But all things, people, all things CAN BE DONE. Do you, or anyone else reading this, also think that a prompt or pop-up (nothing automatic) right after an answer upvote (if the question was not already upvoted) asking if the user would also like to upvote the puzzle would have more cons than pros? $\endgroup$
    – JLee
    Commented Jun 16, 2022 at 9:49
  • $\begingroup$ "It seems to me that all we can really do is keep promoting the notion that people should upvote good, useful, clever, or well thought-through puzzles." That leads to a site like we have now, and a point-grabbing mentality like we have now. Is that really better? $\endgroup$
    – JLee
    Commented Jun 16, 2022 at 9:51
  • $\begingroup$ In summary, you're saying that the issue I brought up might be there, but that the "automatic" idea won't fix it and will likely have more cons than pros, or maybe even 100% cons and zero pros. (but this is assuming that people won't learn the new system and modify their behavior accordingly. I think they would, and quickly.) $\endgroup$
    – JLee
    Commented Jun 16, 2022 at 9:53
  • $\begingroup$ @JLee A pop-up to prompt the consideration of a question upvote would be a better solution than the auto-upvote you propose. In fact, that suggestion got kinda lost in your original post, appearing as it did only in parentheses. I would strongly dispute your claim that it is fairer for an answer to receive fewer upvotes than a question - in fact in my answer I already present several counterexamples: it is entirely circumstantial. There are many complex factors - conscious and subconscious - that go into the decision whether to upvote or not. $\endgroup$
    – Stiv
    Commented Jun 16, 2022 at 10:12
  • 1
    $\begingroup$ And while 'point grabbing' may be an issue in some cases, you appear to be hugely undervaluing the contributions of many users whose answer quality is frankly superb, going above and beyond to ensure readers understand the solution. A user who clearly explains the solution process, with step-by-step logic, images, good formatting, clarity - to me at least - is absolutely worth their weight in gold, or upvotes. $\endgroup$
    – Stiv
    Commented Jun 16, 2022 at 10:12
  • $\begingroup$ Since the answer is always read after the question, it is human nature to upvote and move on, forgetting to go back to the question itself. That produces the disparity we have now. Is it a problem, and if so, what is a solution that would fix it (as much as possible) without making things worse? $\endgroup$
    – JLee
    Commented Jun 16, 2022 at 10:15
  • $\begingroup$ Let's say I want to build my rep on puzzling. By far my most efficient path to do that is to answer as many puzzles as possible and create very few, knowing that the time-to-point payoff is orders of magnitude lower and often bears no reflection of the time spent. My idea to correct this thing that I see as an issue is built off of most (not all) people's general tendency to strive for MIPs (meaningless internet points). If answering and asking were both about the same efficiency in gaining MIPs, that would improve puzzle quality over time. $\endgroup$
    – JLee
    Commented Jun 16, 2022 at 10:22
  • 3
    $\begingroup$ @JLee To be honest, I have often thought completely the opposite also - that there are many highly voted questions where the answer deserves more upvotes than it got. This may be as a result of people upvoting the question when it first comes out, then never returning to it when answers appear. Or answers provided to long-time unanswered questions after a flash of insight. Or reader biases against high-ranking answerers who 'don't need any more rep' - just examples, not comprehensive. There are disparities both ways. By all means, try to resolve it - but I don't have the answers right now! $\endgroup$
    – Stiv
    Commented Jun 16, 2022 at 10:24
  • $\begingroup$ Ultimately, there is a huge element of individual taste behind upvotes/downvotes and it will be very hard to do anything to counteract that. I for one consider it VERY unfair that Jafe's cryptic crosswords don't get far more upvotes than they do, because there is clearly BUNDLES of effort behind these and they are ALWAYS excellent, high-quality and entertaining. But at the end of the day, people who don't like - or don't understand - cryptic crosswords will not be leaving an upvote necessarily, as an artefact of their internal tastes and preferences. $\endgroup$
    – Stiv
    Commented Jun 16, 2022 at 10:27
  • 1
    $\begingroup$ A counter-example can always be found, but I'm talking in general, overall. but thanks for all your points and replies, and for saying that you don't have the answers right now. That's the spirit, I think. Well I don't want to do away with people upvoting or not as a result of their tastes. That is crucial to know what is popular. I am one of those who doesn't upvote cryptic crosswords, for example. $\endgroup$
    – JLee
    Commented Jun 16, 2022 at 10:28
  • $\begingroup$ "but a 'bad' puzzle will usually always find a willing answerer, for whatever reason" The main reason is because that is all that is left unsolved often times, and that is due to the lack of proper incentive to create compared to the incentive to answer. $\endgroup$
    – JLee
    Commented Jun 16, 2022 at 13:06
  • 1
    $\begingroup$ @JLee Another problem with how rep gaining works: If you want to maximize the rep of your questions, you should make them easy and simple, since questions without answers can not end up on HNQ. So I guess the best formula there would be "some interesting/appealing visual puzzle that will be solved in a few hours". $\endgroup$ Commented Jun 16, 2022 at 17:13
  • $\begingroup$ @LukasRotter good point. I don't know how HNQ works, but i do agree with Stiv's observation that most any puzzle going to HNQ gets a couple downvotes bc of it, regardless of puzzle quality- even his massive and intricate 99 problems puzzle which took over 3 months to create. $\endgroup$
    – JLee
    Commented Jun 16, 2022 at 17:16
  • $\begingroup$ @LukasRotter more importantly, my point with this post is not to maximize my question rep, but to construct (or correct) the system itself to equally (or close) incentivize both puzzles and answers, resulting in way better and more puzzles. $\endgroup$
    – JLee
    Commented Jun 16, 2022 at 17:20

You must log in to answer this question.

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