Puzzles Etc.

Jonathan Allan

8:00 PM

the positive thing about bounties on questions is that they are not limited by the daily rep cap

Khale_Kitha

Because if questions go up, so do their downvotes need to go up

question_asker

I don't see why we don't just use the same values we use for answers

Jonathan Allan

personally I think both bounty and change in vote values, but the latter is the first priority

Khale_Kitha

I can see that as good, Jon - but I think you're going to have a harder time convincing people of it, because of the fact that there's a reason that the timer was put into place (to prevent mass bountying)

question_asker

@JonathanAllan I agree

Gordon Allocman

8:01 PM

The rep cap is very relevant since most upvotes on a question come in short bursts (i.e. when the question is asked, when an answer is accepted)

Deusovi

yeah, I agree with Jon

Jonathan Allan

@Khale_Kitha yes, needs to be thought about in detail

GentlePurpleRain

I think if we increase upvotes for questions, it should be to at least 15. Maybe 20.

Deusovi

personally i don't get the point of a rep cap

question_asker

me either, now that I think about it

Khale_Kitha

8:02 PM

Yeah, I think that the 15/-5 that was proposed in the original SE meta question seems appropriate, @GentlePurpleRain

question_asker

it's never bothered me, but I don't really get it either

Deusovi

15/-5 sounds good to me too

Gordon Allocman

I think I read a post about the reasoning somewhere, lemme see if I can find it

manshu

I am with the bounty coz bounty isn't affected the rep cap.

question_asker

So - and I'm only asking this to make sure we don't get accused of not having our stories straight - are we claiming that questions are more valuable than answers?

GentlePurpleRain

8:03 PM

Ideally, I think we need both a bounty (of some sort) and a increase to the upvotes.

manshu

@question_asker err...yeah

Deusovi

hm... depends on question quality

GentlePurpleRain

@question_asker That's what I would say.

Khale_Kitha

I think we're trying to make the claim that quality questions are more valuable than answers.

user61230

The reputation cap was motivated by a desire to prevent a user from getting a plethora of site privileges off one great answer or question. Its secondary purpose is to limit reputation inflation, and its tertiary purpose is to encourage even users who love the site to step back every once in a while.

question_asker

8:03 PM

because a good answer to a good question does require a good amount of work - I'd be fine with them being equal

Khale_Kitha

It does, but they also get far more votes

Due to visibility

question_asker

but that's just me, and if people want questions to be worth more than answers, I won't protest.

Gordon Allocman

Good post by Jon Skeet (the foremost expert on getting rep) about it meta.stackexchange.com/questions/136059/…

question_asker

@Khale_Kitha True, excellent points

Jonathan Allan

We have a cap to limit system gaming and very quick moderator status I think

question_asker

8:05 PM

I trust y'all who actually write questions to know better than me :)

pacoverflow

I think the rep cap should be per question, not for the entire site.

Deusovi

quickly getting mod status wasn't an issue back in beta

here, questions are rarer since they take more time to craft

Jonathan Allan

"To encourage heavily active users to step outside once in a while" - hahaha

Deusovi

ew, outside

manshu

@JonathanAllan Not in my case. :p

Jonathan Allan

8:07 PM

replace "step outside" with "sleep"

manshu

haha

Gordon Allocman

Well clearly it didn't work since everyone knows Jon Skeet never sleeps

Khale_Kitha

@pacoverflow That's actually exactly what Jon Skeet suggested in the post Gordon linked

Gordon Allocman

The stats showed that the vast majority of users are negatively effected by rep limited to questions rather than overall, but that may not be the case here

Jonathan Allan

@GordonAllocman Jon Skeet found the bug that means we need sleep and fixed it

Khale_Kitha

8:09 PM

That's simply because...

Jonathan Allan

doh

Khale_Kitha

of limiting questions to 100 would affect users who only posted one question and got the (200) rep cap, that day

So they'd lose 100 for each of those questions - but it's strictly because of the 100 per question suggestion

Had he suggested 200, there'd be 0% loss

Oh, I hadn't realized this...

Accepted answers are immune to rep cap.

Deusovi

wait, really?

manshu

@Khale_Kitha It's not fair.

Jonathan Allan

oh?

Khale_Kitha

8:11 PM

According to an answer from that thread, yes

66

A: Revisiting the rep cap (yes, again)

a user who dives straight into the community and contributes left, right and centre will still be forced to wait for privileges (and general recognition via rep) This is intentional, and by design. Regardless of how talented one may or may not be, they cannot waltz into a brand new communit...

Deusovi

that's really weird

Gordon Allocman

He is suggesting a hard cap on questions, not a daily cap I believe, which means questions like this one puzzling.stackexchange.com/questions/31557/… would only get 100 (or whatever the set value is) total which seems a bit unfair

Deusovi

yeah...

Khale_Kitha

True, good point, Gordon

pacoverflow

the +15 on an accept doesn't count towards the cap, but any other upvotes on an accepted answer do count towards the cap

Khale_Kitha

8:13 PM

Yeah, that would definitely hurt a lot of people

GentlePurpleRain

Perhaps here we need a hard cap on answers?

Khale_Kitha

And @Alconja would probably drop to like a 4k user, here :P

Ah, okay, @pacoverflow - That makes sense, then

user61230

I'm thinking about +10 on questions, and I'm beginning to wonder if we're overlooking a possible side-effect.

Gordon Allocman

I'd be interested in seeing the stats if there was a separate cap for questions and answers, but I think discussing the rep cap is a bit futile

Deusovi

what is it, Emrakul?

user61230

8:14 PM

What about questions that accumulate upvotes, but really aren't very good?

Gordon Allocman

*discussing changing it, not the relevance

Deusovi

Yeah, I had thought about that

Khale_Kitha

That's why we'd need to change the downvotes, as well.

user61230

We've noticed this problem in the past - riddles, in particular, are hard to judge the quality of before they've been answered.

Deusovi

that's why I was more in favor of bounties too

manshu

8:15 PM

@Emrakul Should we ask a StackApp user about it?

question_asker

Yeah the only thing that's exempt from the rep cap for accepted answers is the 15

Gordon Allocman

This is relevant to the riddles problem meta.puzzling.stackexchange.com/questions/4818/…

Khale_Kitha

I can see that as a benefit to the bounties, though. Most people aren't going to sacrifice their own rep to bonus a bad question.

question_asker

And bounties of course

user61230

@manshu StackApps is weeeeird. We can definitely try, but I'm not sure how helpful it'll be...

manshu

8:16 PM

@Emrakul hmmm...they might not have the same problem. True.

Khale_Kitha

I'll double-vote bounties and questions, then - which pushes bounties to the top, lol

manshu

@Emrakul Should GPR's meta question have feature tag?

Gordon Allocman

I think if vote weight on questions change, the ability to change votes should strongly be considered

question_asker

I'm not too worried about bad questions getting too highly voted. That's not the direction we lean

Deusovi

q_a, it kinda is

question_asker

8:18 PM

Ehhhh

Gordon Allocman

I'd also suggest a consideration in a recently accepted tab since people will probably hold off voting even more on questionable questions that may be good

Deusovi

plastic bag was voted fairly highly at one point IIRC

Jonathan Allan

@Emrakul well the question quality should be decided by the community, if the community like bad riddles they are "good questions", however we "regulars" may feel, no?

Khale_Kitha

That actually brings up something I'd thought of, Deusovi

I almost think that the bounty-on-a-question feature shouldn't be turned on until a question has an answer.

user61230

@JonathanAllan That's a fair point.

Deusovi

8:19 PM

sounds like a good idea to me!

Khale_Kitha

You can't excatly retract a bounty if you change your mind.

Jonathan Allan

@Emrakul in the end that should even itself out I believe

user61230

It's a little curmudgeonly of me to say "all these questions people like are bad!" ;)

Deusovi

agreed, but riddles generally aren't our highest quality questions

there are some exceptions, but...

Khale_Kitha

That's another push towards the bounty system, though. Even if someone likes bad riddles, they likely won't sacrifice their own rep for one, unless they just REALLY like it.

user61230

8:21 PM

@Deusovi very easy to write a bad riddle; very hard to write a good one.

Deusovi

exactly

Rand al'Thor

@Emrakul Same goes for any kind of puzzle though, really.

Khale_Kitha

Yeah

Deusovi

Not really - at least, not to the same extent.

Khale_Kitha

There's certain categories that are harder, but I'd say that more than just riddles follow suit for that statement.

Jonathan Allan

8:22 PM

@Randal'Thor agreed!

Khale_Kitha

Rebus, for example

user61230

@Randal'Thor That's fair. But I seem to recall we've had this argument before... [nostalgic grin]

Deusovi

Oh, I saw that meta thread! The one discussing banning riddles

?

Gordon Allocman

"crotchular region" this question's story is pretty good puzzling.stackexchange.com/questions/32762/good-apple-hunting

user61230

2

question_asker

8:24 PM

I dunno, I'm with @Randal'Thor

Gordon Allocman

also very clear reference to good will hunting

question_asker

And in any case, I've said what @JonathanAllan said above a billion times

Rand al'Thor

@Deusovi Because most of the best riddlers took their stuff elsewhere :-/

Deusovi

SMBC is fantastic

question_asker

We cant be a community site and shoot down what the community likes

manshu

8:25 PM

@question_asker ERROR 404: There aren't billion messages in this chat room.

Khale_Kitha

@Emrakul Exactly.

Jonathan Allan

weird 2 favourites 0 votes

ok 1 up now :p

Gordon Allocman

I favorited it, but I often hold off voting until it gains some traction to make sure it wasn't awful (not that this one looks awful)

Rand al'Thor

@question_asker Um, well, very highly-voted questions here have been nuked before ...

question_asker

I know

Deusovi

8:26 PM

Sure, it's easy to make bad rebus puzzles, but: (1) they aren't a problem here, and (2) it's obvious when a rebus is bad, even for the person making it. For riddles, people can more easily be deluded into thinking they're good.

Rand al'Thor

How to get to an island with a tree in the middle if all you have is rope?

Khale_Kitha

Yeah, I see a lot of favorited questions with 0 votes. That's why I dislike the idea of using the "favorite" button as our system of answering this issue.

Deusovi

I use favorites as a reminder to come back later

Khale_Kitha

Yeah, that's exactly what I said in my comment, on the meta.

And changing that would remove that feature.

user61230

@Randal'Thor To be fair, that question gathered nine delete votes.

user61230

8:27 PM

And was deleted, subsequently, by nine different users.

manshu

@Randal'Thor What is the answer though?

Deusovi

there wasn't AN answer

Jonathan Allan

@Randal'Thor swim?

Deusovi

there were several

Rand al'Thor

@Emrakul And 100-odd upvotes, if memory serves. (Note I'm not saying it shouldn't have been deleted! If anything, I'm making the opposite point: sometimes mega-upvoted stuff is still crap.)

Jonathan Allan

8:28 PM

seems like the obvious answer to me

user61230

Oh! Yeah, that makes sense.

Rand al'Thor

@manshu You have an unlimited amount of rope, according to the problem statement. An unlimited amount of rope has infinite mass, therefore generates a black hole, so you don't have to do anything - the island will be sucked towards you and into the black hole.

That was the top-voted answer.

question_asker

@Randal'Thor all I mean is that it's silly to claim to be community-driven and then go "upvotes? that's Wrong." Either we're community driven and a few people occasionally have to deal with being in the minority, or were not and shouldn't have votes

user61230

Though to be fair, questions like those prompted the introduction of this (admittedly a little unclear) close reason.

manshu

@Randal'Thor Wow, this is really clever.

Khale_Kitha

8:30 PM

Rofl Rand

Jonathan Allan

Wouldn't the island will take an infinite amount of time to reach you?

question_asker

Lmao

Khale_Kitha

that depends - are you infinite mass, too?

Gordon Allocman

@JonathanAllan wouldn't matter since you are dead anyways due to the black hole

Rand al'Thor

Happy to be of entertainment service :-D

Khale_Kitha

8:31 PM

:P

Jonathan Allan

Oh right yes sorry forgot "stab the oracle"

Khale_Kitha

I'm just glad we don't have "up/down/funny" buttons, like Steam reviews do. I've rarely read a "funny" steam review, yet most have a plethora of "funny" votes.

question_asker

Worrying about outliers is a good way to never make meaningful change

manshu

good night guys. Today's fight/argument/discussion was really great

Gordon Allocman

Every Bad Rats review: "Like skyrim with rats"

manshu

8:33 PM

Enjoyed a lot

user61230

@JonathanAllan cries quietly, the oracle lies, the conductor is dead, the train is stopped, and the poor little rock we live on has been sucked into a black hole, never to be seen again

question_asker

But skyrim has rats

Khale_Kitha

Even rats have skyrim.

Gordon Allocman

Neve star evah miryks

Jonathan Allan

gn @manshu

Khale_Kitha

8:34 PM

lol, nodroG

Jonathan Allan

@Emrakul exactly

Khale_Kitha

Need to make a game and call it Miryks

Maybe it'll be the name of a made-up place in a puzzle...

Gordon Allocman

That just gave me an idea to make a puzzle about a certain trip to nilbog

Khale_Kitha

lol

just learned the word Ananym

lol

Gordon Allocman

Apparently my computer science degree was a waste of time because I can't seem to get anywhere on that Good will hunting question

well I guess I don't technically have the degree yet

question_asker

8:38 PM

Ah yes, 'ananym', from the Greek for 'anus name'

manshu

This is, IMHO, the most generally-useful idea here. Critically, it matches the sorts of "topic challenges" that've been done on quite a few sites over the years, meaning we have quite a bit of experience to draw on when designing them. If a platform change needs to be made, this offers a lot more value for the investment than most of the other options. — Shog9 ♦ 11 mins ago

On Weasley's answer

question_asker

Lol

Gordon Allocman

That seems like one of the least useful solutions imo (no offense to wesley)

All that does is essentially give some incentive to do the fortnight challenge, which we don't really see as an issue

Jonathan Allan

We can also achieve that with question bounties if we want (albeit without enforcement)

manshu

That feature request encourages only certain tag(s) and doesn't take every tag to be equal.

question_asker

8:43 PM

Yeah it still seems like bounties+upvote-increase is the way to go

(I still prefer the latter if we were forced to choose one)

Gordon Allocman

Would that bounty only be able to be awarded to new questions posted? if not it essentially does what we want, but I doubt it will allow the bounty to be given to old questions

actually no it wouldn't, if a bunch of good riddles for example get posted, only 1 can get rewarded, thats practically useless

question_asker

And I don't really understand, honestly, the idea of "setting bounties requesting puzzles of a certain type" - how would that encourage good questions?

Gordon Allocman

We don't want a competition for the best question, we just want to reward good questions we like

2

manshu

Phew...I still have some stars left.

question_asker

I'd love to see good lateral thinking problems. I'd love to see good rebuses. I'd love to see good riddles

But a bounty simply requesting those would get us trash

manshu

8:48 PM

@question_asker Same problem with more rep on upvote

question_asker

Not really

manshu

how?

question_asker

We already have the problem of "maybe someone will post a bad puzzle and maybe people will upvote it"

We'll never not have that potential problem

manshu

@question_asker Yeah. And the questioner will get some rep to get the priviledges that he should not be having.

question_asker

But people can reverse their votes

manshu

8:51 PM

Only if there is some edit in the question.

If not. Then no reversing.

question_asker

And again, this isn't a runaway problem, and it's a problem we already potentially have with answers

And yet it isn't an actual problem

Gordon Allocman

I posted a comment on that post about the competition point

GentlePurpleRain

That's another thing that was proposed in meta. An ability to reverse votes after solution is posted.

question_asker

I like that, gpr

Gordon Allocman

I agree with both that post and also the highly voted comments in it

*the vote reversing that is

question_asker

8:53 PM

We worry an awful lot about hypothetical problems

Which, again, is a good way to make sure we never change anything

Rand al'Thor

@GentlePurpleRain I suspect that might be very hard to implement, from SE's point of view.

manshu

ooopss... I am still awake.

gn

Jonathan Allan

gn :p

question_asker

Every stranger on the street could be a murderer, but I don't go around swinging knives just in case one is.

Wesley Situ

@GordonAllocman Good points

question_asker

8:58 PM

Plus, I mean, we're meant to have mods. If the mods are doing their job, we don't have to worry too much about maybe one or two users getting rep we don't think they deserve

Gordon Allocman

9:12 PM

aww yeah, a review stayed around long enough for me to finally get my first review in!

Jonathan Allan

9:46 PM

Who will win the glory of giving me the 10 rep I need to reach 3k?

question_asker

me

(this is that shadow cabal of users who upvote each other's stuff to hit privilege rep requirements we're so worried about)

Jonathan Allan

hahaha

now I own them all or [mutter mutter]