Riddle Sandbox - 2016-08-29

« first day (6 days earlier) ← previous day next day → last day (29 days later) »

12:00 AM

@Alconja Yeah. Point 6 in my answer (people losing interest and momentum) is one of the biggest issues. At least for me personally, but maybe other people don't approach puzzle solving in quite the same way I do.

Point 8 (not enough meta-active people) isn't an issue yet, but it may become one. As is also suggested by the top item on the starboard here ---------------------->

@Deusovi Yeah, I can agree with that. I guess the question is if there's a better way...

For example, I had pretty much stopped browsing the riddle tag because it had turned to junk (though it was once one of my favourites)

So far I haven't been able to think of one. Close reasons wouldn't really work because there's nothing technically wrong with bad riddles that would make them off topic and therefore closeworthy.

Yeah, I ignored all riddles for several months.

So I know we're discussing "losing" people or riddles due to imposing this, but we were losing people by not doing anything (speaking from personal experience)

I'd honestly considered leaving, since it felt like 90% of the front page was just crappy riddles.

I'm sure it wasn't that much, but it was still a lot. The good content was being drowned out.

@Alconja IMO, yes, there are better ways. But I'll get to that in a minute ... I think there are still a few of Deusovi's comments I haven't responded to.

Have you guys ever been to Movies & TV, btw?

12:05 AM

Nope.

cool. :) I gotta disappear for a few mins anyway (got actual work to do.. i'll stick my head back in soon and catch back up)

Well, sometimes I jump around to random SE sites when I'm bored, and I'm sure I've hit Movies & TV once or twice.

But not for any significant length of time.

Right. Because that's where I've learned a lot of the solutions I'm about to propose.

Alright, shoot.

@Emrakul I strongly disagree with this, sorry. What if we get some keen and talented riddler arriving on the site who doesn't care about maths or logic puzzles or any of the other stuff here? Are you saying people aren't allowed to specialise exclusively in riddle? Nobody is going to want to jump through hoops in order to post their riddle when they're not already invested in the site.

I know most of the bad riddles come from new users - but most bad posts in general, on every site, come from new users. The solution is not to block new users. We were all newbies once. If newbies start to feel unwelcome, then eventually the site is going to stagnate.

12:11 AM

Okay, and the solution to improve riddle quality is? Bad riddles consistently got upvotes without the sandbox.

OK. The thing which neither you nor @Alconja seemed to get in the comments on my answer was that with enough community consensus, we can implement a close reason for 'bad riddle'.

Even if they get upvotes from people who don't know any better, people who do know better - people more committed to the community, who are likely to have more rep - can shut them down with close votes as well as downvotes.

And how do you propose we decide what's bad? If new users get their questions closed, they'll think we have a grudge against them.

If it's just because of the sandbox, then they can understand that it's our policy, not specifically against them.

That's the sticking-point, sure. But with enough discussion from a bunch of committed people, I'm sure we can hammer out a definition of 'bad riddle'.

It doesn't have to be completely objective. We can say "here's a bunch of guidelines" (we already have a few old posts about what makes a good riddle - I can dig them out if you want) "and if your riddle doesn't match them well enough, then it gets closed - sorry, but them's the rules".

Wait, let me go find you some precedents.

We'd been tossing that idea around in mod chat for a while. The problem is that the guidelines are hard to nail down, and we'd also have to get significant user support. We tried doing that for just trivia-only riddles, and didn't get any support for a close reason for a while.

On Movies & TV, they have a plague of bad identification questions. "Help, I saw this movie when I was young with a man and a dog in it - what was it called?" (I exaggerate, but some of them really are terrible.) So they established a special close-reason which says:

> Identification questions must contain sufficient detail to meet the site's quality standards and should not be about a commercial or music video. For help writing a good identification question, see: Identify-This-X Questions.

Note that identification questions as a category are on-topic, but 'bad' ones aren't. If they don't meet the site's "quality standards" (which aren't all that clearly defined, really), they can be closed - not just downvoted - for that reason.

And the community are pretty good these days about getting the bad ones shut down.

There are some people, including at least two of the mods, who want to ban them altogether; there are others who occasionally complain on meta about the policy being too draconian; but on the whole I think they've found a good balance.

12:22 AM

Hm, interesting.

I feel like our situation might be slightly more difficult, because "good riddle" is more nebulous than "good ID question".

Presumably you can usually say "give more detail" over on M&TV.

But with riddles I find it difficult to pin down what exactly is wrong with them, and I fear that would lead to complains about non-objectivity.

Actually I've always thought riddles are surprisingly similar to ID questions. Both are basically "here's a bunch of random details about something; now find its name". If there are enough details to be able to identify the thing uniquely, it's a good question; if not, it isn't.

Riddles need more than that, though. There should be some form of misdirection, wordplay, or metaphor - otherwise it's just "identify this thing" which isn't really a puzzle.

For instance, this riddle:

3

Q: I am known for believing my Teddy is real

I am known for believing my Teddy is real I have a girlfriend though she is rarely seen The reliant one is my greatest rival Throughout this all I utter barely a word Who am I? Who you are looking for is a person's name, they are fictional.

@Deusovi Hmm. Not sure if I agree with that. Hang on, let me find an example ...

9

Q: I shrunk the kids

In the United States, You can find me in a camp and in a city, You can find me in a creek and in a grove, You can find me on a hill and on a land. You call my nicknames all the time, You are addicted to me, But if you are careless, I will kill you. You saw me alone in 200...

This doesn't have misdirection or metaphor, and I wouldn't really say it has wordplay either. Yet I still thought it was a fun and challenging riddle.

There's enough clues (e.g. the title) to point you in the general direction of the solution, and once you get it, you know it's correct - as in all the best riddles, the answer is self-confirming.

I'd consider that a form of wordplay - using several different meanings of a word is usually good enough for me.

I'm more referring to things like the one I linked you.

@Deusovi OK, that one's in a style I don't particularly like, but I still think it's a valid riddle. What would you say about this one?

3

Q: Who am I, if not Lewisoff?

I'll help you get across a river. Once unrecognised in Fischer's song, Later I found fame alone And also as a US state. Who am I? Hint 1: Hint 2: To be accepted, an answer must explain all lines including the title. But any serious attempt at a solution will get an upvote from me.

(Ignore the 2nd line, since that was a bit too ridiculously hard. But the rest of it?)

12:31 AM

Also seems fine to me. Again, there's wordplay in it - the "alone"/"solo" thing and the double meaning of "Indiana".

Not my favorite riddle, but it's not bad.

9

Q: The first time I was seen was for Christmas - Who Am I?

First question here, let me know if anything needs improvement The first time I was seen was for Christmas My First name is the same as my mothers surname My job gives me and you power Although I am the weakest I dance better than most Who am I? It is a character in a TV s...

-2

Q: I am a genius. Find out who am I

I am a genius. You all miss me I specialise in Mathematics and crime. A deadly combination, isn't it? People consult me for both. I have only one nemesis of my level of intelligence. He destroyed my reputation. But, Don't worry, I will be back soon. In the mean time, Can you guess Who am I?

riddle literature tv

Those two are better examples of "pure trivia 'riddles' ".

@Deusovi Yeah, nothing really interesting about that one.

In your other two examples, I hadn't actually heard of the solutions, but I assume that to someone who has, they're similarly boring?

They're both fictional characters. Second is Homer's boss in the Simpsons, first is a British sitcom character.

And there's no wordplay or anything - it's just directly describing the character.

This isn't just an issue with fictional characters, but it's most common with them.

This comment just reminded me of another issue related to my point 1 (discouraging new users - sorry to keep coming back to this, but I think it's the most important problem with the sandbox).

Let's say a new user joins the site and posts a bad riddle.

Without the sandbox system, you tell them it's a bad riddle, they learn, they do better next time. But someone still answers it and solves it, so they feel they've had some involvement with the site.

With the sandbox system, they fail, their riddle goes to the bottom and gets deleted. What then? They don't have any connection to the site any more; they've just been failed in their only attempt. They're more likely to just leave rather than work to improve and do better next time.

I think you're making too many assumptions there.

> ...they do better next time

From what we've seen so far, generally they do not do better next time.

They get upvotes, ignore the comments that come in, and continue to post bad riddles.

Generally, we try to be very friendly to new users, even if their content is bad (as long as it's not spam or hateful).

But closing their riddles until they've passed the sandbox test doesn't feel very friendly, even if it's just part of the process.

I saw one user already getting pissed off about it.

12:45 AM

I see where you're coming from, but people typically don't listen to criticism if they get upvotes. And 90-95% of riddles on main get upvotes, regardless of quality.

Yeah, I do worry about that.

SE didn't feel very friendly to me at first, to be honest. Not sure why.

@Emrakul "needs"??? Who's idea was that? I personally think it is up there with the most stupid ideas I've heard of in the stack exchange network. If the riddle (ie question) is good, upvote it. If the riddle is no good, downvote it. If it's really no good, flag it. That's how every other site deals with poor question quality, and I can't see any actual good reason for this site to be the exception. The "sandbox" is just a proxy for the normal process - it adds nothing, but moves the process to a different, concealed, less-community involved place. I played along this time, but won't again. — Bohemian yesterday

Oh yes, that.

@Deusovi Would they listen to criticism if their question gets upvoted and closed?

Nope.

We've had several users post a lot of bad content even after questions get closed.

I'm not going to mention a name here (for obvious reasons) but... do you remember the Security To The Party questions?

@Deusovi Yeah. And I know what name you're thinking of ;-) And I'm pretty sure he'd be fine with us mentioning his name - he's got used to his badge of "guy who asked all the crappy questions" (although he did post good ones too!)

12:51 AM

Nope, you don't. This is a newer user.

I'm just reminding you of that, since the new user posted a very similar question, but with a different "backdrop".

(Not talking about warspyking.)

Oh, OK.

10/15 of their questions by early August were negative. Six of those were deleted. In the comments of nearly all of them there were suggestions on how to improve. We even sent the user a modmail warning them about bad quality questions.

Between then and now, 3/4 of their questions were negative. One was deleted.

And one of the non-deleted questions was very similar to SttP, but with two data points.

How are they not Q-blocked by now?

They have been. Four times.

You can suspend for low-quality contributions, if necessary. warspyking was once. (Not that I'm telling you how to do your job :-) )

12:55 AM

Nah, not planning on it.

Not at the moment, at least.

They're at least actually trying.

But yeah, that's just one user. I can think of three off the top of my head. We keep giving suggestions, none of them learn.

Basically what I'm trying to say is that just criticizing doesn't work.

(Also, all of the users I've thought of have had bad questions upvoted.)

OK, well, that's where custom close reasons can enter the picture.

Have you heard of the 'general reference' close reason SE used to have?

Nope.

It was basically a way of shutting down 'easy' questions: if a question could be answered by general references (Wikipedia, say) rather than specific expertise, it could get closed for that.

They still have a version of that on English Language & Usage. Let me find the text of the close reason ...

> Please include the research you've done, or consider if your question suits our English Language Learners site better. Questions that can be answered using commonly-available references are off-topic.

Again, it's a way of shutting down questions that are just 'bad'.

Dozens of questions get closed on ELU every day for this reason. Questions which are about English language and usage, but can be answered by a dictionary.

Hm, interesting.

Again, I worry about it being more subjective here, but it's nice to know that there's precedent.

To me, as a new user, the riddle sandbox feels like a place where the 'Question Grooming' part of Stack Exchange can happen. I'm not sure I agree that "Too Easy" is a reason not to let things through, but it can help prevent the "Too Broad" or "Poorly Formatted" problems. On other sites, these would be fixed by editing, however here, a riddle can't really be safely edited until it's solved.

Sorry if I'm butting in, just got back and saw the discussion, and wanted to share that, personally, the sandbox seemed to work pretty well for my first riddle.

1:02 AM

@Sconibulus No, please do join in! I'm very interested to hear a new user's perspective on the sandbox system.

My account doesn't actually say I'm new, and I've been around other StackExchange sites for a while, but pretty much all of my contributions here are within the last week or two

I think the most important reason for a sandbox is that a puzzle can't really be "cleaned up" by the community unless they know that certain misspellings, odd word choice, or other miscellanea aren't intentional clues/hints

@Sconibulus Was just writing up a comment to that effect: other SEs have probably avoided "low quality" close reasons, because fundamentally one of the philosophies of SE is that if it's low quality, people can edit it to make it better. But you can't really do that with someone else's puzzle, so a "low quality" close reason actually makes more sense here...

I understand how being told to sit in a sandbox could be frustrating, especially if someone just wants to share something interesting they heard

But I think people who are crafting their own riddles benefit from other hands helping to polish it

It might be better as a chatroom than a meta post, in the long run...

@Sconibulus Yep. And if they have to wait several days for their riddle to actually get published, then the interest wears off and they don't care any more (kinda like point 6 in my answer).

The sandbox isn't supposed to be permanently mandatory, remember.

1:10 AM

Hey @Mithrandir - long time no see! :-)

@Randal'Thor :) Yeah. What's up?

@Randal'Thor If their interest in their own question wanes that fast, is it likely to be of reasonable quality?

@Alconja I'm thinking of the following sort of train of thought: "hey, I've got a cool riddle" -> "oh, here's a cool community I can share it with" -> "wtf? they want me to pass some kind of test" -> waits in sandbox -> "oh soddit, I'll take my cool riddle to Reddit instead" -> leaves

Are they likely to find us because they have a cool riddle?

I dunno. How do people find PSE?

@Deusovi, you got any site analytics which would help?

1:14 AM

I feel like Hot Network Questions is probably a big one

Start with an attempted answer or three, then try contributing themselves?

@Sconibulus Might be the biggest.

Yeah, I absolutely imagine there'll be people who don't bother posting ideas, because the barrier to entry created by the sandbox

@Sconibulus HNQ is how I got here originally. This question, to be exact.

I don't remember, but I probably came here via HNQ too... and to add to that, my first few puzzles weren't particularly high quality either

@Sconibulus That's certainly a good point, and I can see the advantage in having experienced riddlers to help others with improving their riddles while in the sandbox. But it's not for everyone. The idea of banning all riddles that haven't passed through the sandbox just gives me a horrible feeling.

1:17 AM

Sorry, just got back from a fire drill.

Yeah, HNQ is almost certainly it.

Although in the search results that led people here, one of them was - and I am not making this up - "has anyone heard the term tarmac before".

Well, the sandbox is designed as a temporary measure. Once we have firmer guidelines in place, I imagine it will disappear. I hope it doesn't disappear entirely, and there will continue to be a place to get help improving a riddle before posting. Any 'Low Quality' close reason should include a link to this place, in order to help poor posters improve, rather than just being blocked

@Sconibulus Yep, that's the plan!

How temporary is temporary?

Yeah, I quite like that... new close reason -> "low quality" (for all puzzle types, not just riddles), redirects the user to a general puzzle sandbox

2

@Randal'Thor We haven't set any actual dates at the moment.

1:21 AM

plus people could post random (partially complete) ideas direct to the sandbox as a starting place for collaboratively produced puzzles

@Alconja We plan to have separate sandboxes for different puzzle types - one for riddles, one for ciphers...

I think this was started Tuesday?

@Alconja Yeah, that would work perfectly. So people who post good riddles can carry right on (and we won't lose people like Hugh Meyers), but people who post bad riddles get directed to the sandbox to improve them first.

Monday night.

@Deusovi cool, that works too

1:23 AM

Yeah, a close reason would be great - the issue is the subjectivity. We mods wouldn't be able to use it, since it'd seem like we're just being mean for no reason.

And if the regular users used it, it would seem like people were "ganging up on" the newbie.

There is a valid problem with a 'low quality' reason though...

sometimes a riddle only seems bad until you see the solution

"subjectivity" is slightly less of an issue when you require 5 votes (as you say, mods couldn't use it)

and then everything fits remarkably well

I don't recall that ever happening with a riddle. I've seen it with other puzzle types, but not riddles.

@Deusovi Something I saw mentioned just the other day (can't remember where) is that people tend to get much less riled up about five-person closures than about modhammerings. Often five people -> "OK, I'm outnumbered here, maybe something's wrong with my question" whereas one mod -> "mod abuse!!! how dare this person close my question?"

1:27 AM

Fair enough.

(I kinda wish there was a non-unilateral close vote option for mods.)

Just train up the 3k+ rep users in how this new close reason works, and let 'em off the leash. Mods only really need to close questions if they're egregiously bad.

(Except maybe on very small sites where there aren't enough users with close-vote powers.)

@Deusovi YES. I so agree with this. meta.stackexchange.com/questions/41062/…

Yep, I find it weird to not be able to do something that a normal user can.

@Sconibulus I'd say that if a riddle seems bad at first, and then once you get the solution it obviously fits so much better than anything else, then it's probably a good riddle. If the 'correct' solution turns out to fit just as easily as a bunch of others, then that's a bad riddle.

(see: the two examples linked under point 5 of my answer)

Yeah, I can't see one of them, but that's the kind of thing I was thinkning of

Those two examples aren't riddles though. I don't remember that ever happening with a riddle.

At least, not bad --> good.

"seems good, is bad" has happened before.

(cough cough plastic bag cough)

1:38 AM

@Deusovi That rings a bell, but I can't quite place it - linky?

-5

Q: A simple puzzle

You could try taking a snooze with this It won't be pleasant, won't bring you any bliss It's practically everywhere, maybe even on your skin But you wouldn't bring it near the faces of your kin Even though it makes lots of noise, no one pays attention to it Until one collapses and dies from its h...

@Deusovi Huh, that's ... yeah.

It was a meme here for a while.

What made it completely ridiculous was the OP's "I'm not sure" in their own answer.

Here's a riddle which seemed bad and ended up being (probably? I may be biased, since I was the solver) better than it looked.

Eh, I'm not really a fan of that one.

A lot of it seemed to be useless, and other parts were completely unclued.

And it doesn't really make complete sense even after I read the solution.

1:49 AM

So do you reckon that if a riddle looks bad, it probably is bad?

Or at least, it's easier to conclude that way than the other way.

@Randal'Thor You know, first impressions are everything.

Probably, yes. It's possible that it's not bad even if it looks to be.

But it's fairly unlikely. Presentation is a major part of a riddle IMO.

So a "bad riddle" close reason would be unlikely to hit anything that shouldn't be hit.

And if it doesn't hit something that should be hit (a la "plastic bag"), then the question can always be closed after it's solved.

Yeah, I'd say so.

Not sure if I've made this clear, but I do really like the idea of new riddlers having the option to get help polishing up their riddles before they post them. That could bring a great sense of community spirit, as well as a better end product. What I don't like is making it compulsory.

So this suggestion from @Alconja might be (IMO) a perfect solution:

33 mins ago, by Alconja

Yeah, I quite like that... new close reason -> "low quality" (for all puzzle types, not just riddles), redirects the user to a general puzzle sandbox

1:53 AM

Yeah - we're planning on keeping the sandbox around.

44 mins ago, by Mithrandir

The sandbox isn't supposed to be permanently mandatory, remember.

Might've been better to just introduce the sandbox without making it mandatory at first, then make it mandatory later if necessary.

Ain't hindsight grand :)

If only I'd got my answer (+11/-0 already!) posted before the sandbox was actually started :-)

Hey @Emrakul!

2:00 AM

@Randal'Thor though, i'm not sure it would've risen 11/0 so quickly without that hindsight... case in point: the proposal itself rocketed up quickly and unanimously...

Yep.

The proposal went to +25/-0 in what, a day?

True ...

which is a lot on a little meta like m.PSE

(not to say it wouldn't have been a positive contribution to the conversation, just that it's hard to imagine how things will turn out based purely on theory)

It was +27/-0 for a few minutes, I think. Then it was +27/-1.

It depends on whether people upvoted because "yes, this hasn't turned out well" or "yes, these are good points against it".

2:03 AM

Yeah, and I don't think we'll really know.

And at this point it doesn't really matter.

Maybe it's partially disgruntled new users who don't like that their riddles got downvoted in the sandbox.

If we do decide to implement a new close reason, would you want to keep the mandatory sandbox until the details of that close reason are all hammered out?

I'm not sure, to be honest.

I think one or the other by itself would be fine.

2:20 AM

Well, it's really late here. I'm off to bed, but thanks for the chat - I think we've really made progress here :-) See you all again soon!

Catch you later (and welcome back)

See ya! It was nice talking c:

3:20 AM

@Mithrandir @Deusovi @Alconja Was +27/-0. Now is +35/-9. So is +8/-9 in the past few days.

7 hours later…

10:14 AM

0

A: The Riddle Sandbox

My riddle is here, You use me when speaking of the collective. Everyone gathered and put into perspective. Tell me to be silent and there will be a hush. Cal me to you and you may find a rush. Speak of a nation, a multitude endowed. When referring to me, it is always ...

3 hours later…

1:05 PM

2

A: The Riddle Sandbox

Yay, first riddle! I have 256 slots, of which 10 are wasted! I can produce objects lots... ...with just 246 slots!!! My name is long My fame is undisputable I can sing a song with a graphic unconfusable!! Hint 1: Hint 2: Hint 3 (Warning! This is very revealing!):

2 hours later…

3:02 PM

@Deusovi @Alconja @Mithrandir Oh, something I forgot to mention yesterday, which might help with the actual closing of low-quality riddles/questions: how about a dedicated chatroom for crap-catching?

I believe they have such chatrooms at Stack Overflow (the notorious SOCVR), Movies & TV, and English Language & Usage. If we get enough people interested, the discussion over low-quality posts will lead to a lot more of them actually getting closed, without the need for mod intervention.

@Randal'Thor Cool. But I still can't VTC here - I just hit 2k. :/

Yeah, there's a bit of a problem with not enough active users with plenty of rep.

Currently no answers can be deleted here without mod intervention, since we only have two 20k+ users who aren't a) suspended or b) mods.

With you being one of them :P

Yep, me and the mysterious f''.

@Randal'Thor Just BTW, did the 'I'm you' ever make onto TMMoSFF?

3:10 PM

Hello @all

@Mithrandir IDUYA (I don't understand your acronyms) :-P

@Randal'Thor The many memes of SFF

Oh wait, nvm

@Mithrandir Yeah, just got it :-) And no, not as far as I know. In fact, I didn't even remember that was a thing?

I remember the "everyone is Richard" and "Praxis is N_Soong", but not "I'm you".

@Randal'Thor Basically, a room for 3k+ users?

'I'm you' = Mithrandir = rand al'thor

3:12 PM

@IAmInPLS Lower-rep users can still flag to close.

:P

Don't you remember Mythical Rand?

Now I'm getting off-topic, so I'll stop.

If a post has, say, 2 VTCs and 3 VTC-flags, then I don't think anyone would object to a mod hammering it.

@Mithrandir Oh yeah :-D It's starting to come back to me now. Did Axelrod start that meme?

@Randal'Thor Think so. He called me Randy as I left Mos.

@Mithrandir He calls himself Terriblefan these days. I'm chatting with him right now in Mos.

@Randal'Thor Ok, seems a good idea to me. But the main rooms on PSE 'lack' of activity sometimes, do you think many people will join?

3:16 PM

@IAmInPLS There are ways we can draw attention to a new "close vote review" room: a featured meta post pointing to it; let people know about it in comments; get mods to superping people who might be interested.

3

Should be worth a try at least. For the first while it might be more or less the same bunch of people closing many of the questions, but as we attract more and more people to the room and the project, that can change.

@Randal'Thor And as more people reach 3k.

Yes, that too!

offers two large bounties to @Mithrandir :-P

3

@Randal'Thor would accept if that wasn't violating the rules

@Randal'Thor Agreed. A meta post should be enough to notify the ones willing to join this room :)

Yeah, many of the questions I've VTCed had several flags.

And I'm wholly in agreement that the sandbox was not an ideal solution, and probably should be taken away.

3

3:21 PM

Hello @Terriblefan :P

Oh hello, Mythical Rand.

@Deusovi About the sandbox, I think the mistake was to introduce it too fast

@Mithrandir wonders if I should remove the stars on that

I was one of the first upvotes because I was concerned about the quality of the riddles, but it doesn't produce the effect I was waiting for

@Emrakul Hello!

user61230

Hello!

3:22 PM

@Randal'Thor Why, so that it doesn't look like you're breaking rules again? :P

@Mithrandir Have you seen the Rankin and Bass version of The Hobbit?

@Terriblefan I have seen absolutely no version of any Tolkien movies.

@IAmInPLS I still maintain that we can deal with quality without throttling the whole tag through meta in this way.

@Mithrandir Good for you!

So, getting back to the topic of the chat: I am in fullhearted support of that new chatroom for bad posts.

@Randal'Thor ...But I have broken my oath and watched the first Star Wars.

should it be only riddles?

3:25 PM

Eer, bad posts.

or bad puzzles in general?

Did someone say Bad puzzle?

@Randal'Thor Yes, we still need a way to deal with the quality. Maybe the room you are suggesting is a step forward a new, better way than the sandbox :)

@Terriblefan Ouch. remove that atrocity from this room, please.

@IAmInPLS I completely agree. In retrospect, we should've introduced the sandbox first, then made it mandatory only if necessary.

3:28 PM

@Mithrandir Alright

@Deusovi Well, we try, sometimes we fail, sometimes we succeed... But we learn in every way

philosophical thoughts off

@Sconibulus How much of a problem are bad riddles, compared to other bad puzzles?

@Sconibulus I think the room @Randal'Thor is proposing will be for puzzles in general

If we're going to have a custom close reason for bad puzzles in general, it might be harder to word at least semi-objectively than one just for bad riddles.

But yeah, riddles are the first things that come to mind when thinking about low-quality posts (at least, for me)

3:31 PM

number-sequence seems to have troubles

(from my experience of ~2 weeks, may not be overall)

Hello

What about ciphers

@Ankoganit Hello !

@Sconibulus Yeah, it's easy to come up with bad number-sequence puzzles. Ideally there should be more clues dotted around than just "here's a bunch of numbers, find the next one".

@Ankoganit I think this can apply to many tags actually...

Yeah, the most common bad tags I've seen are number-sequence, riddle, and cipher.

3:35 PM

6

A: Getting riddle quality back on track: a proposal for the path forward

I'll add this as an answer: Let's include ciphers. Bad ciphers have long been the bane of this community. Therefore, I propose that ciphers also be sandboxed first.

At +10/-4.

My proposed criterion for what makes a puzzle 'bad': the intended solution fits the puzzle no better than one or more other possible solutions.

Not necessarily a worse fit (as in the "plastic bag" riddle or the Traveller/Troll puzzle), but at least bad enough that it's hard to say objectively that it's the correct solution.

That's the heuristic I've been using.

But it would be nice to make that official.

user61230

@Randal'Thor That seems honestly like a better wording of what "subjectively correct" was intended to do.

3:53 PM

@Emrakul Is there a close reason I've missed? Both this and also your comment here make it sound like you've already tried some of what I've been suggesting.

@Deusovi "Too broad" already works as a close reason for those puzzles, but the advantage of having a custom reason is that you can link it up to a meta post or something which gives people useful guidance on how to write a good riddle (/cipher/whatever).

True. We do only have a limited number of custom close reasons, though.

user61230

@Randal'Thor A Long Time Ago on a Site Very Different than This But Still Under the Same Name...

Not saying it's not a good idea. I think we should implement it, actually.

user61230

I was the one driving most question closures on Puzzling, in the early days.

user61230

I stopped - but mostly out of frustration, and an acceptance that the users of the site get the site they want, and I can't reasonably act as a shield for all things closeable.

user61230

3:57 PM

The number of questions closed dropped to one a day, site-wide.

Oh btw, what happens if a riddle gets to +7 in the sandbox and then gets downvoted to +6? I told this user they could post their puzzle on the main site, but now it's down to +6 again.

0

Q: Can YOU discover the word?

Over time I am worn and worn away like anything else A close cousin of mine treks through desert though I prefer the forest Double the third and remove the first and I become a little bit explosive Remove the last and you may see yourself in me someday What am I?

And there it is! ^

Eh, fine by me to be posted.

@Deusovi Well, you can abolish the "take it to the sandbox first" close reason in favour of this new "no unique solution" one, so that won't be a problem at this stage ;-)

4:00 PM

@Randal'Thor True! That's what I was thinking. Just saying it doesn't leave much room for later. :P

You can always request extra close reasons from SE high-ups if the need arises.

Stack Overflow has something like 6 custom close reasons.

user61230

Sorry, got distracted.

user61230

We were hit by the Ridiculous Spike in 2014, and after that, I sort of tried to take my hands of moderation again. It didn't go down quite as far as zero to one question a week - instead, it went down to around five or six, and a lot of crap that should have been closed wasn't being closed.

user61230

It's quite possible the site is very different now. I can't deny that it's grown a lot since then.

4:22 PM

And that growth has shown itself not just in volume of posts, but also volume of users. Which is why I think a "close vote review" chatroom would work much better now than it would have done a year or two ago: there are enough people to keep it occupied and keep those close votes churning out.

I'm off for a bit - later all!

@Randal'Thor See ya!

3 hours later…

7:04 PM

2

Q: Confusing grammatical numbers

What am I? At StackExchange science I become: Plural at a pyramid bottom Our four letters may lead to wisdom Singular with five to locate from In New Zealand a peak is my home. This has been through the sandbox.

riddle word lateral-thinking rhyme

4 hours later…

10:50 PM

@Emrakul Aha, I worked out what you mean about the "subjectively correct" thing: this custom close reason (which, believe it or not, I don't recall - I think I always just used "too broad" as the close-reason for such puzzles).

11:10 PM

"the intended solution fits the puzzle no better than one or more other possible solutions" => isn't that fairly synonymous with "too broad" anyway?

If we replace a mandatory sandbox with a close reason directing to an optional sandbox, wouldn't we want the close reason to be more "low quality"

@Alconja I guess the advantage of a special close reason for it is that you can include a link which points to a PSE meta post with puzzle-specific advice, rather than just SE's generic "too broad" message.

Yeah, i agree. but the ones being proposed are just synonyms

eg. they couldn't be used on low quality, but "well defined" puzzles like:

-5

Q: I have three legs. Who am I?

I have three legs. I am representing zeros and ones. I am the 20th century, most important invention. I am the most evolved piece of technology in history. I am a building block for the processor. Who am I?

The answer there is clearly "correct", but the riddle doesn't meet (imho) quality standards, and should be directed to a sandbox where it can be improved

@Alconja Hmm. So you reckon the criterion I proposed yesterday isn't enough to catch all the 'bad' riddles?

8 hours ago, by Rand al'Thor

My proposed criterion for what makes a puzzle 'bad': the intended solution fits the puzzle no better than one or more other possible solutions.

Exactly

Because I can't think of many "solutions" to my above example, that fit no better than the intended one

Similarly with lots of sequence/cipher questions

Another question: do riddles of the type you just raised often tend to fall in the "bad riddle, but still getting upvotes" category? @Emrakul?

11:19 PM

the answer may be obviously "correct", but still be terrible...

Because if they don't, we don't need to worry about them so much.

That one's an extreme example, but still has 6 upvotes

user61230

@Randal'Thor Which riddle?

There's many more that were just crap (as opposed to terrible) that get upvotes still

@Emrakul three legs (linked above)

@Emrakul Things like this, where the answer is 'obvious' but still unique.

user61230

11:22 PM

@Randal'Thor Since it's been fairly firmly downvoted, I'm not sure this is a great example. The +6 still irks me a little, but I'm willing to accept it as potentially interpersonal voting differences.

@Emrakul I'm asking a more general question though: do riddles like that one (unique answer, so it would pass the "low quality" test I proposed earlier, but still 'obvious' and a poor question) often fall in the "bad but still upvoted" category?

user61230

In my experience, they tend to. Though less so recently for some of the more egregious cases.

user61230

My observations are anecdotal, though.

It's hard to find examples when you need them... :)

Honestly I think 'obvious'/'too easy' may be something we just have to deal with by downvoting rather than VTCing. It's so subjective, and depends on one's specific subject knowledge. To me, for instance, that transistor riddle isn't obvious at all. I don't even know what a transistor is.

11:30 PM

but like i said, it goes beyond riddles. ciphers like this one has +5/-2 even though there are 15 upvotes on the comment pointing to the "code puzzles (what not to do)" meta post

and i'm sure that has a single "correct" answer

I guess we could say that if a riddle requires specific knowledge of some subject (e.g. electronics, in this case) rather than skill with untangling wordplay, interpreting metaphor, and so on, then it should be off-topic.

user61230

I actually think the one thing the mandatory riddle sandbox was functional at was interdicting normal voting patterns on the site. (I'm not saying it was a good idea overall, but I think it did have that effect.)

I guess my point is that your suggested "the intended solution fits the puzzle no better than one or more other possible solutions" text won't fit many cipher/sequence puzzles very well

But then that would cut out a lot of riddles which I think are actually quite nice, like those in the literature tag, about fantasy novels and stuff, which I and some others had fun with last year.

Also, I don't think having specific knowledge for riddle answers is a bad thing, but it really bugs me that they don't get tagged with additional things like trivia (or movies for movie characters, etc, etc)

A straight up riddle should be a "common knowledge" object as a solution

@Randal'Thor yeah, i think that's fine, if they're tagged riddle literature

11:34 PM

@Alconja So if the transistor one had been tagged riddle electronics ...

still crap, but ever so slightly less so. It's still just a description of a thing, which isn't really a riddle anyway..

but as I said, probably not a great example necessarily, just one that has been used recently in discussions (which is part of the reason for some of the extra downvotes, it more equal +/- previously)

I mean, this question is pretty obvious to anyone who's read Wheel of Time. It's just a matter of subject knowledge. The reason the transistor one is more downvoted is probably just that more people on SE are likely to be knowledgeable about electronics than about fantasy literature :-)

Yeah, I'm going to go ahead and downvote that one too :)

at least someone edited in the trivia tag, but now it's a case of if you've read the books it's an obvious (almost) description, and if you haven't you'll never get it

not really a riddle or a puzzle

And as I said, I think a lot of the downvotes on the transistor one is due to it being posted previously in chat as an example of a bad riddle, so it then gained a bunch of extra downvotes

@Alconja To be fair though, it doesn't say it's about WoT. Whereas the second and last lines of the transistor riddle point to the subject area as well as the solution, making it even more trivial. The WoT one requires someone to make the connection with WoT as well as to actually solve it.

True. And I agree it's better than the transistor one, but still just trivia dressed as a riddle.

11:45 PM

There was a recentish meta post about trivia riddles, wasn't there?

15

Q: What should we do about "riddles" that are pure trivia?

I've seen several examples of "riddles" recently that are pure trivia: I am known for believing my Teddy is real I am a genius. Find out who am I Riddle - Who Am I? (There are many more, but these are the most recent.) I think that riddles should have some form of misdirection, wordplay, or ...

discussion question-quality riddles

Yep. Pure trivia riddles are not even riddles IMO.

« first day (6 days earlier) ← previous day next day → last day (29 days later) »

Transcript for

Aug '1629

Riddle Sandbox

For conversations involving the new Riddle Sandbox - meta.puzz...

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile