Any PRNG good enough for a Monte Carlo simulation would do. You could even just specify the seed instead of giving the whole ≈8MB (or ≈170MB if in decimal) list

CMC create a million random 64 bit ints in any language you like and time sorting it

fastest-code, presumably?

If anyone can try both C and Julia on the same computer I would be very interested

@pxeger yes!

Maybe in place radix sort? Or quadsort?

@Anush generate: shuf -n1000000 -i0-18446744073709551614>f (only works upto 2^64-2 for some reason)

@Anush sort: time sort -n f

(Zsh, this is)

sort -n f 0.96s user 0.03s system 55% cpu 1.763 total

not bad

@Anush I would, but idk how to write to files in python

@pxeger that is an interesting benchmark. So basically the question is how much faster than a second can you do it

@pxeger correction: apparently only 20MB in decimal? (I don't really know how I got my calculations so wrong)

GNU coreutils sort is probably not terrible, but it still has to do decimal stuff which adds some overhead

@pxeger yes. I wonder how much better is possible

Radix sort in hand-written assembly?

@pxeger or maybe just in C/rust/Julia?

@Anush I imagine k is pretty fast

@rak1507 I look forward to the k solution!

can anyone tell me how to write files in Python?

@Anush {x@<x}

@hyper-neutrino I'm late, but postfix

write TO files, as that is

@StackMeter have you googled it?

@rak1507 timing?

what machine are you timing on?

doesn't look like ngn/k supports 64 bit ints

neither does shakti :/

@rak1507 ok I see

@att just compare your timing to command line sort or C's qsort if you don't have a command line

@rak1507 how big does it go?

@Anush the question was because the fastest sort for you will be something hardware specific

@Anush 32 I assume

@rak1507 I don't know. I mean I have a boring PC

@rak1507 you could compare with that to start with

CMC: Given a list of unique integers, output a random permutation of that list. Each permutation should have a non-zero probability, and the probability of outputting the sorted permutation (in ascending order) should be larger than the probability of outputting any other permutation

what is cmc?

Chat Mini Challenge

ah

Basically, a [code-golf] but in chat and less "formal" :P

right

heres one for vyxal: ∆Ṙ[Þ℅|s

this is assuming you mean that the probability for sorted > probability x for any permuation x

Œ!ṢJ1;Ɗị$X in Jelly

5 bytes in Jelly

;-;

:P

what was your method?

Ṣ,ẊX in Jelly

1/2 + 1/n! chance to be sorted

so if random.random > 0.5, sort, else return the perm of the input?

@hyper-neutrino Nice, I had this

random.choice([sorted, random permutation])

@cairdcoinheringaahing ah, nice

mine was just, if random.random(): random permuation, else sorted

so it gives 1/n! + 1/chance to get 0 from a random number

@NewPosts My biggest problem with these kinds of trivial challenges is that its easy to write an answer, but not always to write an interesting answer, and people, naturally, just post the easy answers

Pretty much most answers in normal languages will just be dividing by 0/1 based on the parity, which isn't interesting but is short

@cairdcoinheringaahing got the reversed list twice but not tyhe original in 26 tries

that do be rng

i forgot tack, unlike append, works that way so i didn't see that approach

@hyper-neutrino 3 times in 123 runs but still no sort - is this what bogo sort is

@hyper-neutrino Tack is one of those builtins that's amazing exactly when it shines and completely unnecessary otherwise :P

@hyper-neutrino what does tack do?

It appends x to y

@cairdcoinheringaahing that's just append

I know how argument order works :P

@StackMeter Lemme grab an example

ok

Concatenate (;) vs tack (ṭ)

so top is concatenate

Concatenate (;) vs tack (ṭ) vs pair (,)

^ is better

ok I am slowly grasping the difference

no idea why u would use tack tho

It saves bytes over ;@ (concatenate in reverse order) when working with integers, and it means that you can append a list to a list of lists without having to do stuff like W (wrap)

Compare the three in answers to my CMC, only the tack works, and the concatenate would have to be ṢW;Œ!

k

@StackMeter the probability to get the sorted list by caird's algorithm is 2/n! which is 1/60 for 5 elements, so the probability to get no sorted occurrences in 123 runs is 12.65% which is quite believable

now if you got no sorted outcomes by my algorithm, that would be... concerning

since mine has a 1/2+1/n! chance to give the sorted list

what is your algortithm

take ths sorted list, pair it with a random permutation, and then randomly return one of the two

so it's basically choice([sorted(x), shuffled(x)])

that's how I was thinking of doing it too

@user sure, I'd happily do so

This looks to have a pretty solid consensus (+8/-0 and +9/-0), looks like you should be good to go ahead and create combinatory-logic @Bubbler

I made the incorrect quotes even more incorrect (prompts + AI)

Man, I got existential quick in that first one :P

oh no, they've figured out the cookie maze

abort plan! the users have figured us out!

> Caird: That's what the other players told me it was. I can't remember who they were though. It's too hard to stay awake with so many people talking at once. Maybe one of them will help us figure out where we are and how to get out. If anyone wants to tell me anything else about the game please feel free to do so in the comments below or via e-mail at [email protected] . Thanks for reading!

Shit, how'd they know my email?

@cairdcoinheringaahing can i have my shoelaces back :(

website doesn't exist :(

> [caird] just pick up stuff and take it back to their cave with them, right?

Alright so above the line is using the incorrect quote generator, below the line is using everyone's favourite text generation AI

This is getting out of hand, my address as well??

lyxal got really persistent about me joining y'all on your experimentations on caird at the end there o.O

Honestly I just let it go wild

I love that the last message in each ramps it up to 12 :P

why am i always the (semi-)sane one in both the incorrect quotes and the GPT3 quotes

Diamond

It's always diamond :P

It had the frequency and presence penalty at non-zero values, meaning it favoured tokens not previously mentioned and new topics

diamonds correlate to sanity? gonna have to put a status-norepro on that one

oh god how are there so many flags

....

y'all need to be more active on meta so i can go through these flags faster ;P

No-one's posting interesting challenges, I'm bored :P

honestly i could just do all of the edits at once and then manually bump the few front page meta posts to the top again, lol

Fun fact: that's my 40th flag from this tagging project :P

@hyper-neutrino Of the top 50 recently active questions, 7 are active from a reason other than the status cleanup

yeah i'm just gonna clear all of these and then bump those few back up

1 2 3 4 5 6 7

Active meta posts this month (June) for not-status reasons

@hyper-neutrino Thought: after this, create a meta post along this lines of "why is meta filled with old questions?" just to provide some explanation for anyone who hasn't seen our convos

good idea lol

> My god Hyper... I am getting excited just thinking about all the things we could discover together!!! WOAH!!!! WE JUST SAW A SHARK!!! AAAHHH!!!!! OH MY GOD!!!! WHAT THE F***

This is just Lyxal's brain 24/7 :P

sounds about right :P

I'm a little worried about lyxal at the end there

don't worry, most of his body was retrieved

I'm the top reputation earner for all categories except "all time" (duh), but only by 1 reputation for this week :P

What is this story?

@lyxal Very cool, although I don't see what's so incorrect about the AI-generated ones

:P

Wow. Hyper is the third biggest meta editor with 159 edits. However, looking at his profile, he only has 45 edits. Profile doesn't include tag edits, Editor page does O.o

Mod abuse?

No, status tags

Aww

@cairdcoinheringaahing Do those also count edits on your own posts? Because I'm somehow up there before JoKing, and I've edited barely any posts on Meta

@user No, but they do count edits on CW answers, such as LotM threads

Oh, I see

> a lot of 11

mod abuse is when xnor golfs -(x&1) to x%-2

Good, some mod abuse I can scream at :P !11!111!!

@Neil That's a bit extreme, don't you think?

@NewPosts Maybe one thing to mention (@hyper-neutrino): "if you think a recent, important post has been buried, feel free to bump it to renew some activity"

oh yeah, good idea

@user inf abuse is extreme

Morning.

@Neil (what's inf?)

infinity?

Maybe?

FYI @hyper-neutrino I still have ~40 questions left that should be [status-declined], but I'll wait for meta to resettle before flagging them

What's infinity abuse?

uh oh

alright, thanks for the warning :D :P

@cairdcoinheringaahing Why not do them right now and bump important questions later?

Because it's 1 am and I have work tomorrow

Oh lol

@user I'm sure it's considered infinite abuse when one abuses everything nonstop and can't stop.

That's my context of inf abuse.

@StackExchange I'm not sure what that has to do with Neil's original pun, though

probably things like abusing x/0=inf or related things like doing math pretending infinity is a number or smth

idk tho

Let's just wait for Neil to talk then.

This question is tagged bug but has no status tags btw

status-declined - See the MM post

tip: if you flag those, you get a free helpful flag

Noice.

SEDE should update tomorrow, so I can double check my list against the updated query

@hyper-neutrino Aight, gimme my Deputy badge now :P

0/10 bad flags

@user the joke is that xnor abuses the modulo function (referred to as "mod") to golf code

Hence mod(ulo) abuss

*abuse

caird's flags all included which tag was necessary, any other retagging to be done, and reasons for the selection of flag for most of them like a linked meta post, etc

@lyxal I got that part, but what doe sinf abuse mean?

@hyper-neutrino Sorry

> Should be marked [status-bydesign] per Taryn's answer. Should also be tagged [reputation], [protected-questions] and [privileges], instead of [answers]

Behold, a good flag :P

granted the point was i should've been going through these doing the research to retag them, but i've been slightly busy :P

@lyxal I'd like the term "abuss".

@user inf -> infant most likely

Welp, that got dark

I could be reading too much into it though and the joke could just be infinite abuse

ಠ_ಠ user's stealing my flags :P

... not sure if i should 11 that. i will leave it unless anyone asks me to