@Brandon_J I responded to that already

@Brandon_J And I think that's a problem. Chat rooms get vastly less traffic and attention than Meta. Meta is the place to air those sorts of issues. Burying it in chat means getting very little feedback from the general user base.

(For an issue with this much potential impact, I'd argue it ought to be pinned to the sidebar.)

@Brandon_J It's worth noting that the reason Code Golf doesn't insist on a single correct answer is because the purpose of the site is not to collect knowledge about a particular topic. The purpose of Code Golf is to play a game for the enjoyment of its users. Puzzling shares this purpose: it is not here to collect knowledge about puzzles, but rather for its users to engage in solving puzzles for their own enjoyment.

This purpose was establish during the discussion about "challenge" questions. While more abstract questions about approaches to constructing or solving puzzles are certainly on topic, the vast majority of posts here exist for participants to entertain themselves.

My point being that I think that having to worry about ensuring every post has a single, demonstrably (I won't say provably, since that's probably too high a standard for puzzles that are not in the logic or math categories) might significantly detract from people's enjoyment.

@jpmc26 @jpmc26 One of the issues I see with allowing open-ended "puzzles" on this site is that they invariably attract a lot of low-quality answers. The goal of PSE is to be a repository of good quality puzzles, and having questions that inspire these kinds of low-effort solutions runs counter to that goal.

@jpmc26 I disagree with this claim. Most of the best puzzles on this site have a single, demonstrably correct solution, in your words, while most open-ended questions tend to receive at most a lukewarm response.

And for a real world example example, take a look at the kinds of puzzles presented in puzzlehunts, such as the MIT Mystery Hunt. Again, you'll see the vast majority of these puzzles have objectively correct solutions, and I don't see complaints from people that they are too broad.

(Sorry for tagging you twice btw, didn't know that replying to a message would automatically tag)

Crap, I ran out of time to edit that post. I meant to say "too restrictive," not "too broad."

@PiIsNot3 Reception to open-ended looks similar to all questions: data.stackexchange.com/puzzling/query/1039061

@PiIsNot3 More detailed query: data.stackexchange.com/puzzling/query/1039061 There are more negatively scoring questions. The group that's significantly less is the 10-20 range. The >20 range is slightly higher in open-ended percentage-wise. I'd argue this is still pretty similar, even if there's maybe a few more percentage points negatively voted questions. It doesn't seem like there's a significantly greater dislike of open-ended than of the general site, though.

In particular, there's certainly not a complete dearth of >10 questions.

I still find 10% of all the questions with the tag having a negative score to be worrisome

Dang it. I thought it would fork the query. First one should link to this version: data.stackexchange.com/puzzling/revision/1039061/1283229/…

Feel free, but your claim was that questions in open-ended "receive at most a lukewarm response." The data doesn't seem to support that assertion.

OK, maybe I was being a little harsh with my judgement

And keep in mind that open-ended is hardly the only kind of puzzle that doesn't necessarily have a definitive solution.

Yeah, formation-of-numbers type puzzles can also be counted as open-ended

which is why I think a Meta discussion dedicated to that specific point is in order, as it doesn't seem to have been discussed in the open.

@PiIsNot3 More than that. What about an optimization problem that turns out to be NP-hard? Are those going to have provably optimal solutions? Are they automatically bad puzzles as a result?

But again, I don't believe that "popularity" should be the only reason why a low-quality puzzle that attracts low-quality or troll answers should remain. At best they should be closed, in my opinion, and there are plenty of other puzzles that don't require popularity for them to be considered good

@PiIsNot3 "I don't believe that 'popularity' should be the only reason why..." Nor do I, as I've re-emphasized. This is a misrepresentation of the point I'm making. Please stop repeating it.

@jpmc26 I believe Deusovi addressed this issue in the meta post when they wrote that "questions about whether something is provably optimizable" should be considered on-topic

@jpmc26 Then I'm still not sure what stance you're taking on this issue. Are you arguing in favor or against removing the open-ended tag? Is it purely an issue with Deusovi's meta post, as you've stated before? Or is it something else entirely now?

(I don't want this to get too confrontational, so I apologize for misinterpreting your claims)

@PiIsNot3 Which post is this? I'm not seeing it in the chat history.

@jpmc26 The original meta post that this chat is tied to

@PiIsNot3 The stance I'm taking has morphed only slightly. It's still essentially an objection to the point that puzzles must have a single, demonstrably correct answer as one particular argument, but what I'm saying at this point is that there seems to be a missing Meta discussion about site scope and whether that's a requirement the community wants or thinks is appropriate.

@jpmc26 OK, I see now, thanks for clarifying, I was starting to get confused by the direction that this was taking

One more clarification: For that discussion, about how the community sees the site's scope and whether requiring a single demonstrably correct solution for all questions in any tag, I think popularity is definitely a relevant consideration.

I believe there are several meta posts that have reached consensus on whether certain puzzle types are on topic or not. I've linked several here and in the comments of the meta post, but the most applicable to this situation would probably be this one, which is referred to a lot when questions are closed as off-topic

If problem questions arise more easily in questions with no single provably correct solution, there may be other ways of addressing that than to ban them wholesale, as well.

@PiIsNot3 (My apologies for responding harshly. I've had some recent unpleasant experiences with being misrepresented here on SE, so I may have been overly sensitively.)

It also seems like this very issue was brought up when the site was still in public beta. I guess having a SE site that acts nothing like an SE site presents quite a few challenges

@jpmc26 No worries, thanks for clarifying your position!

I believe, based on the consensus that the original meta post drew, that the community now considers open-ended questions to be off-topic, except for the specific types of questions that Deusovi identified at the end of their post

Whether that'll cause problems down the stretch can only be speculated on, since to my knowledge the site hasn't had to deal with this issue ever