The Nineteenth Byte

hyper-neutrino

15:04

@cairdcoinheringaahing is there a shorter way to get the range from -x to x than {-x}r{x}?

caird coinheringaahing

ŒR

hyper-neutrino

ah there it is :P thanks

didn't realize there was the abs part, so my search for -z didn't work

caird coinheringaahing

Kind of. If x is negative, ŒR gives [x, x+1, ..., 0, ..., -x-1, -x], but Nr$ gives [-x, -x+1, ..., 0, x-1, x]

hyper-neutrino

yeah i don't really like that OER does auto-abs

caird coinheringaahing

Comparison

hyper-neutrino

15:09

ah cool, thanks

Monica Cellio

15:42

@cairdcoinheringaahing (Lost power for the last 18 hours; pardon me while I catch up, and sorry for the abrupt departure.) No, Codidact doesn't have a rep cap.

caird coinheringaahing

@MonicaCellio Do you mind if I ask about the motivation behind that? I find the cap is a really good way at preventing power users from heavily outpacing regular users and helps level the reputation-based playing field

Monica Cellio

@cairdcoinheringaahing we haven't actually discussed it. I think the general feeling has been that since rep doesn't lead to privileges, we didn't need to restrict it. We haven't really thought about poser users creating large disparities in the number.

Adám

@MonicaCellio So what do Cod (is that how you abbreviate?) points do?

Monica Cellio

@hyper-neutrino thank you! I appreciate your participation. And yes, one of the frustrations from SE -- not the big obnoxious one that broke the camel's back, but a frustration nonetheless -- has been that SE is "one size fits all" + "smaller sites can stay if they don't get in our way". Codidact set out to be community-driven from the start. Most things are customizable, and we'll do our best to support needs we didn't know about. (The answer TOC is because of CG, for instance.)

The dev team is small so we can't always do everything right away, but we aspire to be as supportive as possible. We also welcome contributors, either for one PR to fix that one annoying thing or for longer term.

@bobble each category has an RSS feed.

pxeger

anyone know who this was?

caird coinheringaahing

15:52

Mego

142

Q: It's time for me to leave

On October 27th, I decided that the following requirements must be met by November 1st for my continued involvement with the Stack Exchange network: The forced relicensing from CC-BY-SA 3.0 to CC-BY-SA 4.0 must either be validated as legal by a lawyer or legal team experienced in copyright law ...

Monica Cellio

@Adám we usually abbreviate "CD". Reputation is basically "reputation" -- bragging rights, signal of expertise, etc. Eventually we want to also recognize (and label) tag-based expertise, but that'll be based on the actual post scores, not the rep number. I think "has lots of well-received posts about X" is more meaningful most of the time than "Y rep".

Adám

Right.

Redwolf Programs

@StackMeter What about it? It wouldn't make sense for it to be anything other than true

hyper-neutrino

@MonicaCellio Oh, is Codidact open-source? Didn't even realize (and am not surprised) - that's very nice. Yeah, I think the system being built from the community up rather than communities being created from the network down is definitely something I like and can really appreciate.

Anyway, the proposal to add a feed for Codidact seems to be going very well, just got the Nice Answer badge from it with no opposition so far

Wezl

16:07

another gripe about infix: !x == !x, -x - -x, and +x + +x all look really imbalanced. At least that's something APL's got right

hyper-neutrino

Final opinion on adding a chat feed for the Codidact Challenge category to TNB? - at +10/0, seems to be good to go

caird coinheringaahing

Personally, I'd like to get an answer about the Sandbox and Meta before adding a feed

I suspect that answer will be "no", but I'd like it nonetheless - might write it up in a bit

hyper-neutrino

fair enough. we could decide separately and add separately but doing them together is a good idea probably

Redwolf Programs

My opinion would be sandbox - yes, meta - no

hyper-neutrino

I don't have a solid opinion ond Sandbox yet, but I would vote against meta.

pxeger

16:12

@Wezl wrong

looks (and is) perfectly correct to me

Redwolf Programs

The first one definitely looks fine, I see what you mean about the rest but not why it matters.

caird coinheringaahing

Damn, you just ninja'd me hyper :/

hyper-neutrino

oops

Wezl

it's funny how it's "fgitw" on main and always "ninja'd" in chat

Redwolf Programs

Two different things though

You can ninja on main

Wezl

16:15

oh right, ninja is saying the same thing ig

caird coinheringaahing

@hyper-neutrino You wrote basically the same things as I would, I'm going to switch and write up my opinion on Sandbox posts

Redwolf Programs

CMQ: How many of you think golfing languages have had a negative impact on the site? (This is for an upcoming CGCC blog post)

user

I think they may have had a slightly negative impact on the site, but overall, I'd say they've helped people get and stay interested in CGCC

Wezl

overall, positive. But I do think they have had a negative impact

ninja'd :\ (except if this is counting people it doesn't matter)

Redwolf Programs

Can y'all list a few ways you think they've had a negative impact?

user

16:23

Discouraging users who look at Jelly and 05AB1E and go, "Look at those meaningless characters, that's cheating, I'll never be able to beat that"

hyper-neutrino

it's a lot harder to write interesting challenges now because most golfing languages have many builtins that just allow you to put like 4 builtins together and solve a task that is still rather non-trivial for a practical lang

discouraging new users is another thing i was gonna mention ^^

user

But that's kinda a good thing

Wezl

@RedwolfPrograms long term mental effects :P

user

Because if a golfing language can't do it in two bytes (without Dennis magic), it means it's an interesting challenge

hyper-neutrino

you mean not an interesting challenge?

or am I misreading

user

16:24

Oops

hyper-neutrino

@hyper-neutrino i wouldn't necessarily say this is that bad though

a lot of the recent challenges i've done in jelly have taken like 20-40 bytes due to being quite complex and interesting

and a higher metric for challenge quality isn't bad; verbosity doesn't always imply compelxity or uniqueness anyway

Redwolf Programs

I guess they sort of move the boundary between trivial and non-trivial

hyper-neutrino

hence why prime challenges are discouraged, because it takes like 50 bytes in python but 40 of them are the same as 30 other python answers

i have trouble finding too many reasons to oppose golfing languages TBH. i'll leave this one to others to answer :p

user

Because they're good for CGCC :P

Redwolf Programs

Oh no, I just went to the homepage and it's janky

user

16:26

(unrelated: is there a list of good questions for newcomers?)

hyper-neutrino

yep, already reported on meta but stuff's broken again ¯\_(ツ)_/¯

@user these are quite simple

caird's list of challenges to take new languages for a spin with; also doubles as a good list of challenges to inspire new builtins

user

Ooh, nice

Redwolf Programs

Oh, that's a good list to bookmark

Wezl

oh ouch I just noticed because usuallly I only go to the main page knowing what I'm looking for

user

caird should post that on Meta

caird coinheringaahing

16:27

I find it interesting that I've been given credit for that when Martin first made it

hyper-neutrino

:p

caird coinheringaahing

> Forked from m-ender/simple-challenges.md

Wezl

@cairdcoinheringaahing advertising is half the work :P

Redwolf Programs

Should the Yes or No argument come first in the blog post?

user

No first, that way it won't seem like we're arguing against them

And we can bias people to like golfing languages mwahaha

Wezl

16:28

Yes first because ^^ :P

caird coinheringaahing

@RedwolfPrograms Flip a coin

I don't think it matters that much

AviFS

TNB is so active these days! What happened?

Redwolf Programs

I won't shut up

7

caird coinheringaahing

We all got replaced with AIs

user

Speak for yourself, I was always an AI

Wezl

16:31

not my fault

Redwolf Programs

Okay, found a coin. Heads is Yes first, tails is No first.

I uh...hit my ceiling fan

hyper-neutrino

*lands on side*

also congrats

Redwolf Programs

Heads

Yes first

Wezl

> * hyper-neutrino lands on side

user

@RedwolfPrograms That's not very nice of you.

ngn

16:33

cmc: design a protocol for two or more people communicating over the internet to produce a fair coin toss

AviFS

I haven't gotten any feedback on this sandboxed post yet, but I'm about to post it. Here's a heads-up to the FGITWs & if anyone has any last-minute feedback!

caird coinheringaahing

So we've now got answers on all three categories from Codidact (ignoring their Q&A system), I think we should let them "sit" for a day or two to get some more feedback

AviFS

It's my first CGCC question, so I'm making a bigger deal out of it than I would otherwise. Sorry!

Redwolf Programs

Posting it here for feedback is highly recommended, no need to apologize for it :p

2

user

@ngn They message me to toss a coin for them. I toss a coin on video. I'm too dumb to do deepfake stuff, so that won't work. If one of them bribes me to toss a coin a certain way, I'm still too incompetent to do that, so it's still 50-50 :P

caird coinheringaahing

16:36

@AviFS Id' suggest changing the "note" to be more "Alternative way of phrasing it" as that way helped me get it more than the first description

Wezl

@cairdcoinheringaahing personally I think the Q&A should be posted here, for the same reason as the challenges

ngn

@user good answer but needs more fairness :)

AviFS

@cairdcoinheringaahing Good point. I just wish I knew how to make that less clunky

hyper-neutrino

@ngn have a third person do the coin toss. ez.

caird coinheringaahing

@Wezl Feel free to add an answer to the meta post - I didn't include it in the question but it's just as valid a topic to discuss

user

16:37

@ngn Something something quantum entanglement something?

hyper-neutrino

diffie-hellman creates a random key right?

so you could probably use something like that

AviFS

I don't really want to put Alternate Phrasing:

Wezl

I'm too tired to write markdown :\

hyper-neutrino

and that'd even make the outcome of the coin toss private

caird coinheringaahing

@AviFS "Another way to look at this is to arrange the numbers 1-9 ..."

Redwolf Programs

16:37

There's no way I could pick which side the coin landed on, given that it hit a fan and at least two other objects at high speed first :p

hyper-neutrino

impressive

AviFS

@cairdcoinheringaahing And the challenge is so short that I'm pretty sure everyone will read it anyway, regardless of the Note or not Note heading

Wezl

@Wezl well actually too tired to write English :\

ngn

@hyper-neutrino right but requires a trusted third party

AviFS

@cairdcoinheringaahing Oh! So just take out the Note: entirely? I can do that

ngn

16:39

@user there's no need for quantum computing, it can be done very simply using just cryptographic hashes

caird coinheringaahing

@AviFS Yeah. I generally dislike the "Note:" or "Edit:" headers that people use, as it's generally easier to just write it as standard English :P

AviFS

Although then it blends into the paragraph before, which should stand out

Extra?

user

@ngn They both choose a string, the strings are concatenated, then hashed?

AviFS

I feel like it helps break it up

user

@user Odd is heads, even is tails?

Redwolf Programs

16:40

That's a good idea

hyper-neutrino

right but the second person to send their message can manipulate the result

caird coinheringaahing

@AviFS Its your challenge, I just generally find that not having bolded bits aside from delineating sections is obtrusive

ngn

@user what i had in mind is: generate a random number and send a hash of it to the other people. once all "commitments" are exchanged, everyone reveals their random number. they are xor-ed together and the last bit is used.

user

@hyper-neutrino They won't know at first

Redwolf Programs

Both could write it in a private chat room first

user

16:41

^

It would have to be trustworthy though

Redwolf Programs

And then a mod could verify they don't change it when they see the other

AviFS

@cairdcoinheringaahing I'll start with "One can frame the problem equivalently:"

user

@RedwolfPrograms Nah, third parties complicate stuff. What if someone who wasn't ethical , like HN, decided to side with one person?

hyper-neutrino

you could post the hash to your string first

AviFS

@cairdcoinheringaahing But how I can make the problem as stated clearer?

hyper-neutrino

16:42

both parties select a string and post its hash

then, both parties post their string, and it can be verified to be unchanged

then, they are concatenated and hashed

Redwolf Programs

That's a good idea

ngn

@hyper-neutrino that works too

AviFS

"Given a turn and a square, print a draw with that move." is admittedly lame

user

@user Oh wow that struck-through comma looks like ⍪

Redwolf Programs

I thought that was some weird combining diacritical mark with - lol

caird coinheringaahing

16:45

@AviFS I think its pretty clear as is, especially with the examples. Maybe "Given an integer representing the nth turn in a game of Tic Tac Toe, and an integer representing the position on the board, output a completed Tic Tac Toe board where the result is a draw and the inputted square is filled on the inputted turn", but that's a bit wordy

hyper-neutrino

that usage of strikethrough on the comma is a perfect example of how annoying grammar is

user

English grammar, at least

Redwolf Programs

@cairdcoinheringaahing It's wordy but I think it makes it clearer

AviFS

@cairdcoinheringaahing Oh hold on, I just wrote "Given a move on the board and a turn to make it on, output a game with that move which ends in a tie. A cat’s game!"

I'll look at yours now

Redwolf Programs

Oh, I like that one

caird coinheringaahing

16:46

@AviFS That;s better :P

user

br⍺⍳nf∪⊂k grammar is a lot nicer :P

AviFS

Is it? I rather like yours. But thanks! Only thing is yours restricts the input/output more than necessary

user

Hmmm

Redwolf Programs

Maybe it's some weird form of unary?

user

Ah, good old base 1 :P (edit: nvm, base 1 would only have one digit - 0)

Redwolf Programs

16:48

(I once made a language where 10 would have meant 1: 0 was 0, and 1 was the successor function :p)

AviFS

Final: "Given a move on the board and a turn to make it on, output a drawn game where that move was played."

Redwolf Programs

"Drawn game" sounds confusing to me

Since it could also mean "a drawing of a game"

AviFS

Ah, good point. Thanks!

Given a move on the tic-tac-toe board and a turn to make it on, output a tied game where that move was played. A cat's game!

Isn't there a constraint-solving tag?

Monica Cellio

@hyper-neutrino yes - platform is open source (contributors welcome), and the particular instance that runs on codidact.com is backed by a non-profit (the Codidact Foundation), so profit motives won't drive either the code or the specific communities. (Other people can, of course, take the code and set up their own instances with whatever policies they want.)

user

Feels like it should exist

AviFS

17:01

Right?!

Adám

@MonicaCellio Will you at least fire moderators without reason?

AviFS

@MonicaCellio Hi Monica!!

New Posts

0

Q: A Cat’s Game to Claim

A Cat’s Game to Claim For every move in tic-tac-toe, there should be at least one way to draw that includes that move. Or so it seems; prove me wrong! Challenge: Given a move on the tic-tac-toe board and a turn to make it on, output a tied game where that move was played. A cat's game! Any input/...

array-manipulation grid game

AviFS

Thanks for your help, @cairdcoinheringaahing & @RedwolfPrograms!

caird coinheringaahing

Happy to help :)

Redwolf Programs

17:03

No problem!

AviFS

:))

caird coinheringaahing

Presumably [code-golf] is the winning criteria?

I always overlook that in Sandboxed challenges, cause I just assume its there :/

AviFS

17:16

I totally forgot to say that, haha thanks!

Is adding [code-golf] as a tag enough, or do I have to mention that explicitly?

caird coinheringaahing

We have a consensus that you should mention it explicitly, but it generally doesn't matter if its [code-golf]

AviFS

Ah darn. Pxeger already edited that in, haha

Monica Cellio

@Adám Someplace Else already has a lock on that. :-)

@AviFS hi!

Newly Featured Posts

18:01

29

Q: Decode USB packets

Inspired by this video by Ben Eater. This challenge forms a pair with Encode USB packets. The USB 2.0 protocol uses, at a low level, a line code called non-return-to-zero encoding (specifically, a variant called NRZI), in which a stream of bits is encoded into a stream of two electrical level sta...

code-golf binary encoding

hyper-neutrino

18:13

What would y'all estimate is the average number of questions a user posts before reaching 150 rep? Asking for some mod work.

Sandbox Posts

0

A: Sandbox for Proposed Challenges

Recursive Stalin Sort code-golf array-manipulation sorting There is a "sorting" algorithm often called Stalin Sort, where instead of sorting an array, you just remove any items that are out of order. In this challenge, you'll implement part of a sorting algorithm which recursively sorts the remo...

Redwolf Programs

@hyper-neutrino One question is all it takes if it's good, but the problem is a lot of users probably post answers before questions

hyper-neutrino

It's in regards to this; we could just raise the threshold to like 150 or so.

Redwolf Programs

Probably two on average for users that start with questions

That'd be my guess

caird coinheringaahing

if you start by posting a decent question, you're likely to get +15 no sweat

hyper-neutrino

18:18

i think it's a pretty good idea then (raise to 150); i'll forward that sentiment on then. thanks :+1:

New First Time Asker Dialog text is still too long; I would like some feedback on how we can shorten it (like cutting unnecessary bits out, pointing the user to the sidebar tips to avoid repeating information, etc.)

caird coinheringaahing

By "too long", is that displayed length or actual length (think links being shorter when displayed)?

I think it should be ok to remove the "The FAQ on Asking good questions contains more details and goes more in-depth on advice for asking on this site." bit, as it links to an answer on the Welcome page, which is already linked at the start

Or even, link the "Welcome to.." bit to the Asking answer, as its more relevant

hyper-neutrino

how it looks on my screen, 24", 1080p

Adám

18:33

@hyper-neutrino Heads up, you have a message on Teams.

caird coinheringaahing

Has Catija (or another CM) said why they can't remove the other two bullet points? They take about 1/3 of the space, and aren't even relevant

hyper-neutrino

oh yeah i read that on my laptop already and it doesn't involve me anyway :p thanks

@cairdcoinheringaahing not afaik; i'll ask

Redwolf Programs

It's probably not built into the tools CMs have

The devs probably assumed nobody would need to change or remove them

hyper-neutrino

18:47

@cairdcoinheringaahing basically, they made the third able to be turned off because many sites might not fit that philosophy, but the other two fit Q+A sites so there are only a handful of sites that might not want them (CGCC, Puzzling, SW/HW recs, CR maybe, and I think that's it...)

when this modal was made, the bullet points were created with SO in mind

caird coinheringaahing

What a surprise

But honestly, if there are more than one or two sites where it makes sense to disable/change the 2 bullets, why not design them with that in mind, or even, change it?

hyper-neutrino

yeah, I don't see how it could've possibly been hard to make all of them a toggle instead of just the third ¯\_(ツ)_/¯

Adám

My impression is that SE has quite a bit of technical debt.

hyper-neutrino

seems like it... again, with the alignment of the front page breaking today, it seems every fix brings a new bug

Adám

Big ball of mud

A big ball of mud is a software system that lacks a perceivable architecture. Although undesirable from a software engineering point of view, such systems are common in practice due to business pressures, developer turnover and code entropy. They are a type of design anti-pattern. == In computer programs == The term was popularized in Brian Foote and Joseph Yoder's 1997 paper of the same name, which defines the term: A Big Ball of Mud is a haphazardly structured, sprawling, sloppy, duct-tape-and-baling-wire, spaghetti-code jungle. These systems show unmistakable signs of unregulated growth, and...

hyper-neutrino

18:54

@cairdcoinheringaahing both good suggestions; done. yeah it makes sense to link the Asking section/answer under the general welcome post when trying to ask

new ss

caird coinheringaahing

I think that's long, but ok. There are a couple of areas that could just about get away with being trimmed, but otherwise, I think that works

hyper-neutrino

so it turns out i'm stupid and the 150 rep threshold change is regarding the answer popup and not the asker popup

I think the change to 150 still makes sense and is good though. any objections?

(proposal is to allow the How to Answer box to show for people even with the association bonus, so you will see it if it is your first time answering or have less than 150 rep, i believe)

caird coinheringaahing

Yeah, I think it makes sense either way

Redwolf Programs

Might make more sense if it's just based on first time answering

caird coinheringaahing

Ah yeah, 150 rep is more difficult to get from answers only

Redwolf Programs

19:06

You'd have to answer probably ten questions on average to reach 150 from 1, so that box will get annoying

Is there any way it could be based on rep minus the association bonus?

(reputation - bonus) <= 20 or something could be a good metric

hyper-neutrino

> Building it to actually disregard the association bonus would be much more difficult.

Redwolf Programs

Aw :(

caird coinheringaahing

101 might be an appropriate threshold then

Redwolf Programs

reputation % 100 <= 21 :p

hyper-neutrino

o.O

Redwolf Programs

19:08

@cairdcoinheringaahing That's still annyoing to new users, but essentially first time answer only for people already on the network

Seems like it has the disadvantages of both options

caird coinheringaahing

If it just goes by first answer then?

Redwolf Programs

I wonder if it could be based off "first positively voted answer" or something

Risky's 4 bit codepage is so overpowered when you can use it effectively

hyper-neutrino

at that point "ignore assoc bonus" would be easier

the point is if there is a workaround by just setting a specific rep threshold then we'll go with that

user

19:30

@hyper-neutrino Maybe 1 or 2, considering how many people post answers first

Redwolf Programs

19:47

Feedback on Recursive Stalin Sort?

pxeger

Part of me isn't completely sure about the name

but it looks good to me

(unless it's a duplicate, which seems not unlikely)

Redwolf Programs

Yeah, it's not very descriptive of the actual sorting algorithm, but the subset the challenge requires implementing is basically recursively applying stalin sort.

A username

How's Risky going?

Redwolf Programs

In what way?

hyper-neutrino

SciFi election is now live

can't wait to see how this one goes this time

Redwolf Programs

20:02

I'm probably not going to vote

I got a notif for it, but I don't know who any of the candidates are :p

hyper-neutrino

fair enough :p i don't even have the rep to, anyway

Redwolf Programs

TBF I didn't really know much about any of the CGCC candidates either lol

pxeger

> 100,000+ reputations

reputations? lol

hyper-neutrino

@RedwolfPrograms also fair enough :p

caird coinheringaahing

I've interacted with 5 of the 7 SciFi candidates, so I've ranked those 5 and left the other 2 :P

user

20:05

@pxeger Yes, one reputation for each of their personalities :P

caird coinheringaahing

Also, I don't know how to tell user931 this, but Valorum has 500k+, 100+ gold badges, is regularly active and was (long story) user 30 on SciFi, so user931's claim is significantly less impressive :P

Redwolf Programs

There are a lot of ways you can spin "I've been center of controversies in the past & I have even been banned multiple times", but "good leader" probably isn't the best one to choose :p

caird coinheringaahing

Also, check out the comments underneath their nomination - that looks to be the main cause of drama in this election

scifi.stackexchange.com/election/…

hyper-neutrino

"I maybe most misunderstood user on this site" usually does not inspire confidence.

caird coinheringaahing

I have to say, it's an interesting election strategy

Redwolf Programs

20:11

Along with discussing your real-world political opinions

hyper-neutrino

I think TLC, Jack, maybe also SQB and Jack are looking like the best candidates IMO

Redwolf Programs

VTCing Jack as a dupe

hyper-neutrino

er, i meant adamant instead of the second jack

caird coinheringaahing

I can't speak to Skooba's moderation side, but they've been consistently a friendly and welcoming user in my experience on SciFi and Movies

hyper-neutrino

oh yeah skooba looks like a good candidate too

i haven't had as much of a chance to look through SQB or Adamant's nominations and there hasn't been as much time for people to comment either, so ¯\_(ツ)_/¯

Redwolf Programs

20:16

^ My votes, based off of y'all's opinions and skimming the nominations

caird coinheringaahing

I've got the same candidates ranked, but as TLC, Skooba, Jack, SQB, Adamant, ibid

hyper-neutrino

from everything i've heard and reading the nomination posts i'd probably rank the same as caird, but i might drop ibid off my rankings. not sure though and I don't intend to get the rep to vote anyway, and even if i got rep, i don't think i've participated enough for it to make sense to vote. i haven't interacted with any of the candidates to my knowledge

Redwolf Programs

With STV it's not likely anything other than the first two or three will actually matter IIRC

hyper-neutrino

true. since there is only one slot anyway

Redwolf Programs

TBH I don't really like STV

hyper-neutrino

20:22

It seems like making the new answerer box ignore association bonus would take quite a while. So, what do we think in terms of what rep threshold we should use? Or should I post this to meta?

@RedwolfPrograms what about it in particular?

Redwolf Programs

I just like some of the other methods a lot more

I kind of liked the simplicity of the RO voting method; you just +1 candidates you like, and -1 ones you don't.

I also kind of like systems similar to cumulative voting

hyper-neutrino

fair enough (i think that's called approval voting? although with three states rather than two)

Redwolf Programs

Rep based cumulative voting (ignoring association bonii) could be cool, but might not be that good of an idea

TIL "dolphinarium" is a word

Catija

Short of just having a minimum rep to vote, I'm not a huge fan of weighting votes by rep.

Redwolf Programs

Yeah, the more I think about it the more problems it would probably cause

hyper-neutrino

20:31

@RedwolfPrograms I'd personally oppose that. I personally believe every user's vote should be treated the same (assuming they've participated enough to warrant having a vote). There's a reason moderators are totally powerless on the election page, for example, because I don't think having more privileges means you should have a larger say in site moderation. At least IMO ¯\_(ツ)_/¯

caird coinheringaahing

Given the fact that rep doesn't really scale well, its really not a great idea

hyper-neutrino

I feel like an election should be a chance for the whole community to get a say in how the site is run from the top, whereas rep-scaling votes would potentially make regular users feel like their voice isn't heard or they don't matter and the clique (is that an appropriate term?) of high-rep powerful users just get to control everything.

Catija

Approval voting is generally just upvotes. Downvotes can serve a purpose - generally for very controversial candidates - but they can nullify votes (or act as double downvotes), so they become very strategic. Pure upvote-only approval voting doesn't have any sense of nullification, you just vote for the people you do like and don't for the people you don't like. So the most-liked people win. That's it.

Redwolf Programs

@RedwolfPrograms "Penguinarium" is also a word, which seems like it would be even more niche than a dolphinarium

hyper-neutrino

@Catija ah. true. I'm not sure of the term for -1/0/+1 voting like that (if there even is a term? I don't even know if one exists), it just happened to be convenient for a substantially simpler selection process for a much less impactful thing

Catija

20:34

For multi-winner elections, they're generally the best. The big downside with STV is the risk of a multi-winner election where the second-most popular candidate gets only second-place votes and is eliminated in the first elimination round. So a less-favored candidate gets the second slot.

hyper-neutrino

Oh, that's a good point. Would it be possible to just do like a point system where you get like, 1 point for 1st, 0.8 points for 2nd, less for 3rd, etc, and then just add up points and top wins?

Redwolf Programs

The main thing I don't like about STV is that once your vote goes to someone, the rest of the ranking doesn't matter

hyper-neutrino

though I guess that allows someone with fewer first place votes to win

Catija

That's generally an unlikely outcome but it's not impossible. :)

Redwolf Programs

@hyper-neutrino That's my favorite system, I think

hyper-neutrino

20:36

13 mins ago, by hyper-neutrino

It seems like making the new answerer box ignore association bonus would take quite a while. So, what do we think in terms of what rep threshold we should use? Or should I post this to meta?

Catija

@RedwolfPrograms That's not true, though. If there's more than one slot and the top candidate has a surplus, a fractional amount of your vote will go down your ranking.

hyper-neutrino

feedback on ^^? might just post to meta soon

Redwolf Programs

@Catija Huh, I didn't know that

caird coinheringaahing

@hyper-neutrino 10 rep might actually work

hyper-neutrino

10?

caird coinheringaahing

20:37

Generally, answers only get upvotes if they're good, and if you can write a good question, you can probably write an ok answer

Catija

10 doesn't fix the issue of having people with the association bonus not see it - there'd be nothing to fix.

Redwolf Programs

That doesn't fix the problem with new users with the assoc bonus not seeing it

ninja'd

Catija

:P

caird coinheringaahing

Oh right, yep, ignore me :/

Redwolf Programs

First answer only might be the best solution, honestly

Anything over 100 is going to annoy new users without the bonus quite a bit, and anything under 100 won't be visible to users with the bonus

caird coinheringaahing

20:39

How annoying is it?

Catija

I have a recommendation - Would y'all be up for us setting it to 111 to start and seeing how people respond to it? How frequently do newer users post multiple answers, particularly in a short period of time?

caird coinheringaahing

I think that works. Most people won't be annoyed by it too much, and once they hit 111 rep, they'll probably notice its no longer there are figure that its for newer users

Redwolf Programs

Except ais523 :p

hyper-neutrino

oh yeah, lol...

Catija

It becomes a site setting, so if it's super annoying, I can drop it to 10 and we can pretend it never happened... the thing is, this isn't a feature that many sites are using, and on most sites, the extra guidance is something that only relates to new users... so, as much as I hate to say it, it's really difficult for us to have a huge backlog of bugs and FRs and to focus a ton of time on something that only helps a single site.

caird coinheringaahing

20:42

I don;t think we've ever had anyone report a bug on it, and the only FR was to change it (and this rep discussion), so I think that once this is changed, the only possible change we'd need would be to change it back and pretend it never happened

Catija

I... think y'all would generally understand that there's times where you have to prioritize things based on the number of people running into the problem or the severity of the problem when it gets run into... and this is pretty low on both counts. I ... really don't like discounting the problems y'all have because I do understand that the site has a lot of fun and value and people enjoy participating...

We just have to be thoughtful about where we spend our time. :)

I think this option makes sense as a useful starting point and hopefully it won't be something people struggle with. :)

hyper-neutrino

I think setting it to like 111 would probably be a good place - it is just a not-too-huge box that you can just skip past and the post button is right under it, so I think this'd be alright. I totally understand that this is pretty narrow of an issue to our site and putting dev work into something that only we'd end up using is not going to be very viable. It might annoy ais523 but most new users will probably get 110 rep in not too long and it's not a huge deal to have to skip

over it each time, at least IMO.

user

But all it'd take people with the association bonus is one upvote

caird coinheringaahing

That's fine

user

Isn't there some way to subtract the association bonus?

caird coinheringaahing

20:46

So long as users see it on at least their first answer, that's generally enough

Redwolf Programs

> Building it to actually disregard the association bonus would be much more difficult.

user

¯\_(ツ)_/¯

@RedwolfPrograms Oh ok

hyper-neutrino

I would rather more false positives on who to show it to than false negatives

caird coinheringaahing

Obviously, we don't have any stats on how effective it is (as we can't see people who change based on the box), but we don't get that many "wrong" answers, so it seems to work

Catija

Do y'all get a lot of bad answers from users with the association bonus?

user

20:47

Well, we also don't get a lot of answers :(

caird coinheringaahing

@Catija Not really

hyper-neutrino

lemme search by newest deleted answers

Redwolf Programs

We just need to do an A/B study. Capture 100 CGCC users and put them into small boxes to reduce the chances of external influences. Then, show half of them the popup. Compare the resulting answer quality.

caird coinheringaahing

@user We have more answers than like 90% of the network sites

user

@cairdcoinheringaahing I meant by new users (or do we have more of those too?)

Catija

20:48

@user Looks like ~75 in the last day. That's... a lot?

caird coinheringaahing

@user No idea, Catija probably has the ability to get those stats

Catija

Please don't make me write SQL.

7

user

@Catija Is that from all users or just new ones?

caird coinheringaahing

@Catija Just shoot an email over to one of the devs :P

(joking of course, we don't actually need that data :P)

Redwolf Programs

May 24 at 19:42, by Redwolf Programs

If SQL were a food, it would be socks with sandals

Catija

20:50

@user That's just from the recent answers search - codegolf.stackexchange.com/… Not any specifics for who's doing the answering.

user

Oh ok

Catija

I assume that y'all have a ton of answers from the really engaged users and not as many from newer people.

That's pretty normal.

user

ig the very nature of CGCC makes it so we get more answers than challenges

hyper-neutrino

most answers seem to either make a small mistake and get self-deleted, be comment-type or trash answers from 1-rep users, spam, or not follow I/O / loophole-related rules or such

people with the assoc bonus shouldn't be making comment-type / other non-answers

user

Invalid or poorly formatted answers are the only real danger from them, I guess

Catija

20:51

Capturing new members is going to be hard for any site and that can depend on a lot of things. I think that it's maybe hard for newer people unless they're already really well-versed in golfing.

hyper-neutrino

i'd have to dig back further to see if there are instances of people with just the bonus writing answers that don't follow standard rules, don't meet the specs, etc

Catija

Do y'all have a lot of non-golfing questions?

user

No, 10k of the 12k questions here are tagged code-golf

hyper-neutrino

@Catija it can definitely be intimidating, at least. our policy is as long as you show that you've put effort into golfing it, it's acceptable, even if that just includes making your variable names one character and removing whitespace, but your code is still thrice as long as the next solution in the same language, it's acceptable

@Catija 84.6% of our questions are code-golf, lol. so we have some but not that significant of an amount

(out of non-deleted ones)

caird coinheringaahing

@Catija No, but of the last few non-challenge type questions (think tips for golfing), one of the best was by a new user

Redwolf Programs

20:54

I wonder what the smallest subset of tags is that includes every question on the site

Catija

@user Sure. Y'all do it to an extreme, but that, at least, is pretty standard on most sites ... other than SO. What can happen with new people is that if answers come in right away and they don't have anything to add, they may feel like they don't have a place to fit in. They want to add to the site and so highly-competitive things like CG can feel hard to hop in as a novice.

user

@RedwolfPrograms That's NP-hard, I think :P

caird coinheringaahing

I think that we do have a higher bar than most (already high bars) SE sites, but we do try to be as helpful and as lenient as we can, especially with new answers

user

@Catija That's fair, I don't think I would've bothered answering if I had seen people answering in the language I wanted to use

Catija

The unanswered question is often a place for new users to start... or the question with a half answer. The closest equivalent is someone finding a solution that's one character shorter than the top answer... but y'all are so darn good at this is so many different languages, that it maybe leaves people with not much to add.

caird coinheringaahing

20:56

Answers don't have to be the shortest possible, so long as its clear some effort is made to golf. First Posts and Late Answers reviews are cleared within minutes, and most of them leave some form of welcoming comment. Our highest voted questions are (mostly) challenges that are easy to dive in (think HW, or Fibonacci)

hyper-neutrino

that's definitely true. although we are totally fine with people submitting a 100-byte python answer to something with 4 50-byte python answers already, as long as it makes an attempt to be golfed, it might not seem like that to a new user trying to find something to answer

user

@RedwolfPrograms oneboxes are annoying

caird coinheringaahing

Plus like 90% of our traffic comes from HNQ, where people see challenges and go "Oh, I can write a solution to that in <lang>", and they usually do

Redwolf Programs

Aw, I reached the maximum search size so I have to use SEDE instead

user

@hyper-neutrino And even if it's everywhere on Meta, the Tour, etc., most users are just going to browse CGCC without looking there at all :/

Catija

20:58

@hyper-neutrino And that's good. The question is whether many people really realize that. In my mind, if I only know python and there's an answer already that's half the length I could have managed, I... don't even try. Now, that's just me. It's great that y'all are welcoming of that and encourage people to post but I'm not sure that, in the end, it's how people choose to participate. It's almost as if y'all need a way to hide all the answers for the first day and then reveal them later on.

If that makes sense?

user

I suppose editing all questions to include a note saying "Hey new user, ignore other answers and just do your thing!" wouldn't be feasible :/

Transcript for