The Nineteenth Byte

Bubbler

12:03 AM

github.com/Kindelia/HVM

Holy sh*t, it beats GHC exponentially (at least in some cases)

Ginger

wait what is that? a really good haskell compiler??

user

@UnrelatedString If you're still around and it isn't a bother could you check out the Prolog room? We need your help

Ginger

read the readme, and all I have to say is: cool

user

@Ginger It's not just a compiler and it's not for Haskell itself

Aiden Chow

@mousetail probably not, considering that the crack isnt even valid (adapt code prints "2\n" instead of "2")

Bubbler

12:08 AM

not really haskell, but a compiler for untyped lambda calculus

Ginger

@user well the readme sure looks impressive

user

It does

Bubbler

and we have a winner for

3

A: Sandbox for Proposed Challenges

Fastest untyped lambda calculus evaluator fastest-code functional-programming lambda-calculus interpreter Challenge What it says on the tin. Mainly because googling "fastest untyped lambda calculus" gives almost zero meaningful results. Each submission is expected to take a lambda term from STDIN...

yay.

Steffan

12:27 AM

@user well, don't bounty any more of my answers. I'll probably post plenty

user

lol okay

Unrelated String

12:45 AM

@user np

OldSandboxPosts

1:00 AM

Sandbox posts last active a week ago: Simplify a Cycle

lyxal

Any last feedback on ^

might post today

emanresu A

The last one is confusing.

By removing the [2, 7, 2, 7, 2] loop you get [1, 2, 3, 7]

By removing the [7, 2, 7, 2, 3, 7] loop you get [1, 2, 7]

lyxal

you're removing the first loop you see

emanresu A

What about [1, 2, 1, 3, 2, 3]?

lyxal

[1,2,3]

emanresu A

1:11 AM

... why?

lyxal

[2, 1, 3, 2] is reduced to just 2

no wait

[1, 3]

smooth brained being able to read the first 1 there

emanresu A

What about [1, 2, 1, 3, 1, 4, 5, 6, 7, 8, 3]?

lyxal

[1, 4, 5, 6, 7, 8, 3]

[1, 2, 1, 3, 1] -> 1

emanresu A

... even though it could be simplified to [1, 3]? That's... inefficient

lyxal

> Note that the simple cycle may not be the shortest cycle, and the shortest cycle may not be the simple cycle.

emanresu A

1:17 AM

hmmmmm

New Posts

2:16 AM

0

Q: Write a program with the smallest width

Print or return the following text: Hello world! The quick brown fox jumps over a lazy dog. However, your submission will not be scored in bytes. Instead, primarily it will be scored by the width in pixels of the program when written in Arial in a 12pt font. In particular, the score is: $$ (\tex...

emanresu A

@NewPosts Unary with ZWSPs, length 0

lyxal

Don't even need unary

Just use stuck: esolangs.org/wiki/Stuck

Wait there's more text

Smooth brained it

Bubbler

I have a JS template that allows coding in one column

Aiden Chow

4:14 AM

@Steffan bro i find it way easier to program in knight lol

prolog is just way too confusing

knight is easier to understand

Unrelated String

knight is also way harder to write something coherent in

Aiden Chow

@UnrelatedString ya i suppose if u were to look at someone else knight code it would be confusing but i find it relatively easy to code in

and tbh prolog looks just as, if not even more incomprehensible than knight

could be im just a noob at prolog tho lol

like i cant understand shit in prolog

the predicate stuff and unification or whatever, and cuts like my brain hurts just trying to understand it lol

Unrelated String

don't worry about cuts

2

yet

worry about cuts once you understand everything else

Aiden Chow

ye true, still tryna understand everything i read in the tutorial linked in lotm post

thejonymyster

5:06 AM

@NewPosts relevant

Bubbler

5:55 AM

I̝̝̝̝̝̝̐̈̄̐̅̆̐̌̆̐̌̆̐̏̆̐́̂ win?

Unrelated String

nice

Bubbler

(changed the message to "Hello!")

Aiden Chow

bruh what was that huge line stretching across my screen a second ago

my screen like glitched out or smth

Bubbler

That's literally the source code for my answer :P

Aiden Chow

@Bubbler oooh wait, didnt see ur msg lol

Bubbler

6:31 AM

@Bubbler Now scoring less than 1 :P

emanresu A

6:46 AM

In a hypothetical version of Vyxal that did base decoding with large amounts of the same chars (WIP, may become a feature) it would be .28

1

Q: If an event has a statistical probability of only 50%, is it possible to use a neural network to predict it with more than 50% accuracy?

For example using a neural network to predict a coin toss. Can a trained neural network to predict it with more than 50% accuracy?

neural-networks

🤔

lyxal

7:32 AM

Purely statistically, no.

Accounting for physical and environmental conditions (like coin weight, force of flip, weather etc), maybe

Bubbler

If the outcome is 50:50 but does depend on several factors, NN will do much better than when the outcome is 99:1

Otherwise NN just becomes a sophisticated RNG

graffe

8:53 AM

morning all

mousetail

11:10 AM

@user how does your bounty work? First answer is 100, second 200 etc.?

user

11:55 AM

@mousetail yes

mousetail

Ok time to write 10 prolog answers

user

Obviously aftr 500 it gets confusing so I’ll probably stop

@mousetail go for it, there’s a prolog room if you want

I know barely any prolog myself but it’s cool to see

Sandbox Posts

12:46 PM

0

A: Sandbox for Proposed Challenges

Print the source code of the previous answer This is an answer-chaining challenge. The objective is simple: print the source code of the previous answer! When writing your program, it must print the source code of the latest answer. I have written the first answer. No answer will be accepted, so...

thejonymyster

@SandboxPosts i could swear we have this already

Radvylf Programs

Whoaaaaa

Rust is seriously OP

4

My parser worked first try after I got all the compiler errors cleaned up

user

@RadvylfPrograms Yes, it is God's chosen language.

@RadvylfPrograms How complicated was it? That's really cool

mousetail

There's a reason for it's popularity

Radvylf Programs

@user Moderately. About 300 SLOC, has to deal with different types of parenthetization, somd really unusual parsing rules for unary and binary operators, insertion of default variable names when an operator is missing a required argument, etc.

Took quite a bit more debugging on JS, and still had some weird issues

user

1:00 PM

Wow

And this was without a parsing library too?

mousetail

Pattern matching is really nice for language design

pxeger

@RadvylfPrograms I especially like Haskell because it gives me lots of compile errors

Radvylf Programs

@user Yep

Seggan

1:33 PM

@RadvylfPrograms SLOC is what lines of code?

Radvylf Programs

Significant, I think. Just non-empty ones.

lyxal

I always thought it was spaceless lines of code

Wheat Wizard

@pxeger If it compiles, it works.

mousetail

Source Lines of Code right?

pxeger

@RadvylfPrograms I think it also excludes comment lines

@WheatWizard That's what they say, at least :P

Radvylf Programs

1:35 PM

@pxeger Yeah, it usually does

SNBeast

1:59 PM

if you know me, you know that gbz80 is the best language :P

thejonymyster

@SNBeast what's the best language if i don't know you?

java

glad to know ya

|:|

thanks for the oreo

2:08 PM

what kinda oreo is that

|I| is more accurate

thejonymyster

yeah

lyxal

@pxeger the kind with eyes

thejonymyster

the better to... see oh oh kay eye ee you with?

pxeger

sideways oreo: =

nope doesn't work properly

thejonymyster

i actually had a really good oreo ascii

but i ated it

Ginger

2:11 PM

we are not amused

lyxal

> we

Ginger

me and my clone army

lyxal

oh boy it's early 2021 all over again :p

The Ginger Hivemind(tm) now exists apparently

@user get in here we have an organisation to topple!

Ginger

does he.. does he suspects us?

thejonymyster

@lyxal *2012

lyxal

2:14 PM

@thejonymyster *January 22 2021

Ginger

why was your name capitalized then

how do I know that's not evil mirror dimension lyxal

did you have a goatee too?????

thejonymyster

back then it was pronounced "licks-all"

Ginger

I still pronounce it like that

sue me

thejonymyster

(i mean same, it also makes "lyxtures" easier to say)

it also makes more sense for "vyxal", sounds like "vixen"

lyxal

@Ginger because lowercase lyxal was officially sanctioned by hyper-neutrino

Ginger

2:18 PM

what

thejonymyster

im trying to think of how to spell it so itd be unambiguously pronounced "likes-all"

(without just doing that)

Ginger

liecksa\y\l

lyxal

Jun 4, 2021 at 6:44, by lyxal

in hyper-neutrino's domain, May 5 at 23:07, by Lyxal

@hyper-neutrino I have a reasonable need of your moderator help. I would like a username change to lyxal (very important to notice the lower case letter at the start). I can't do it myself because 30 day policy reasons

Seggan

@thejonymyster i say "likes-al"

woops wrong quotes

lyxal

@Seggan good. Excellent

thejonymyster

2:19 PM

like "Like Sal"?

ya

better like saul

>:(

Aug 30 at 14:18, by lyxal

better call lyxal!

Seggan

@lyxal although i pronounce Myxal as "Miks all" :P

lyxal

2:21 PM

@Ginger if you think that's something, emanresuA used to be A username

Ginger

and I used to be Gru ensgdrtsnIiei

pxeger

@user why do you have so many sockpuppets? I've been following the chain of links in their bio's for 3 hours

lyxal

@pxeger all that prolog rep has to come from somewhere

do you think it just magically appears out of nowhere?

Ginger

rep don't grow on trees

lyxal

now admittedly, explicitly linking all those sockpuppets isn't a good idea - it makes it obvious there's rep laundering happening

what @user should do is open a car wash or a nail salon and launder the rep that way

Ginger

2:25 PM

how? rep isn't physical afaik

lyxal

bounties

you think they exist for awarding good answers

Seggan

hmm is rep laundering against site rules?

lyxal

when really, they're used by stackexchange underworld kingpins to move reputation between accounts :p

Ginger

I'll keep that in mind :b

lyxal

@Seggan probably, but you gotta turn that dirty rep into clean rep somehow

stackoverflow.blog/2022/08/31/…

Major assumption that blockchain is a solution to a problem and not an invented problem like it is

mousetail

2:35 PM

Oh no

user

2:48 PM

@pxeger Huh, I don't remember modifying my bio to say that

pxeger

It's not your bio, it's one of your sockpuppets'

user

@lyxal Yes, quantum cloud blockchain NFTs are perfect for money laundering

4

My sockpuppets' bios are my bios too, after all

mousetail

Petition to rename OTTNB to "byte 20 and up"

thejonymyster

@mousetail surely theres a joke about code bowling to be made, but i dont know anything thatd make sense lol

mousetail

the 9th pin?

Ginger

3:01 PM

The Bowling Green

pxeger

→ 32 messages moved to Off-Topic TNB

I haven't been following the conversation so I don't know what exactly you wanted moving

Please let me know if there's anything else

Ginger

that's good thanks

user

Well I kinda also meant the stuff about oreos and money laundering but that's less noisy, this was the more urgent stuff

pxeger

@WheatWizard How many mods are to be elected? Is it even a fixed number?

Ginger

1

Wheat Wizard

3:07 PM

1

pxeger

Right

@cairdcoinheringaahing your prophecy was right!

Wheat Wizard

You can see the number of positions on codegolf.stackexchange.com/election/4, but it's fine to just ask.

pxeger

Oh didn't notice that

Wheat Wizard

No problem.

Just idk if noone is online the next time you want to ask about the number of slots in a moderator election you can look there. (I guess this is helpful only in an incredibly niche scenario.)

pxeger

thanks lol

I was deciding whether or not to start writing a nomination for myself. Because I know who I want to be moderator, and if there's only one position I would rather them than me ;)

Seggan

3:41 PM

ugh im under 18

nobody would vote for me anyway lol

thejonymyster

CMQ: (genuinely) We have code-golf for minimizing, code-bowling for maximizing, and code-shuffleboard for approaching some specific value. Are there any other code-<sport>s which would make sense at all?

Ginger

code-decoding

Seggan

code-challenge? :P

thejonymyster

idk im just wondering if we could have like. Code tennis

Ginger

Did you mean: answer-chaining?

2

oh neat the label's in italics too

Seggan

3:46 PM

@thejonymyster i think thats called a king-of-the-hill

pxeger

king-of-the-hill is Code King Of The Hill

thejonymyster

code shotput

pxeger

What about Code Capture The Flag? I guess that's kind of what cops-and-robbers is

thejonymyster

i can see that

well no in cnr the flag only goes one way

Aiden Chow

@Seggan riiip, im under 18 too :|, but i feel like i wouldnt stand a chance either lol

thejonymyster

3:50 PM

do you want to be mod?

yall are crazy

Seggan

@AidenChow id never thought so

Ginger

~~imagine having responsibility, couldn't be me~~

Wezl'

@pxeger cops and robbers is cops and robbers :P

Seggan

though someone should make a capture the flag style challenge

whichever side has most points?

tho its kinda like a koth

PyGamer0

@RadvylfPrograms damn

thats too OP

Seggan

3:56 PM

#nerfrust

pxeger

Code Chess?

Ooh, code golf on a chess clock timer would be fun

5

Aiden Chow

@thejonymyster bro i got some free time to kill lol

like might as well

Wezl'

@Seggan capturing a flag could be like a combination of answer-chaining and cops-and-robbers

so it's captured when you can extend it and you try to make it hard to extend?

@pxeger the first step could be to figure out how to make it turn-based

thejonymyster

itd be funny to have a team aspect though

for ctf

imagining a rule like "you join a team by submitting an answer for that team, and youre not allowed to switch teams"

and its like, the challenge is to print either Cat or Dog or something and you have to follow specific rules on how you can modify the program

Wezl'

makes sense

thejonymyster

4:09 PM

likely set by the previous answer

team Cat and team Dog

pxeger

This isn't chess clock based but it's a similar concept:

a judge or panel of judges sets a problem, e.g. "output sequence A0123456 in under 80 bytes in Python"

then two players or teams compete to try and solve that first

basically like competitive programming but with a length restriction

Ginger

oh boy

thejonymyster

@pxeger its fur elise lol

Ginger

but we already have a timed challenge upcoming (hopefully)!

thejonymyster

@thejonymyster this ruels actually i might write it up at some point

ill hvae to think of rules rules though

and a better challenge

PyGamer0

4:16 PM

TIL you can play sequences on OEIS

Wezl'

@pxeger have you seen codingame's clash of code

pxeger

No

Wezl'

and i'm sure there are other similar games

thejonymyster

code football

Wezl'

its automated and it doesn't have subjective "best program" but there are code golf challenges and shortest coding time challenges

pxeger

4:22 PM

@thejonymyster that's basically just code bowling then

thejonymyster

ok wait thing ive realized

pxeger

Most sports are just "highest number of points"

thejonymyster

the three tags we do have are competitive sports but they mainly feature one person playing at a time, rarely directly interacting with the other players

and then koth and cnr are like... are those sports? or

pxeger

What is a sport, and what is merely a game? These are the kind of philosophical questions we answer on... Code Golf Stack Exchange.

Steffan

clash of code is a multiplayer coding game. there's three modes, one is code golf, another is just whoever can solve it first, I forget what the last one is

graffe

4:27 PM

@thejonymyster Hello!

Wezl'

@Steffan it's whoever can solve it first but you only get the test cases rather than the challenge statement

called reverse I think

Steffan

exactly

yes I just ermembered

Ginger

you rember

Wezl'

I always have problems dismembering stuff :P

Sandbox Posts

0

A: Sandbox for Proposed Challenges

write the "index by number" sequence Imagine writing out positive integer numbers, and indexing each character in a 1-indexed array (indices read vertically on the top two rows, numbers read horizontally on the third): 000000000111111111122222222223333333333444444444 12345678901234567890123456789...

Radvylf Programs

4:33 PM

One disadvantage of Rust though:

Ginger

rookie numbers

New Posts

4:56 PM

0

Q: Prove the Mathologer 20 card solution

This very interesting video of Mathologer states the following problem: Given a sequence of 20 cards face down, one move consists of taking a card face down and flipping it up, and then flipping the card to its right (if it exists). Prove that any sequence of such moves terminate. Pick a language...

code-golf puzzle-solver proof-golf

thejonymyster

@graffe aloha graffe :-)

caird coinheringaahing

5:35 PM

@thejonymyster Yes

It'll give me an excuse for why I don't have Marshal yet :P

@RadvylfPrograms You can still read the tab text, this is amateur hour over here :P

DLosc

@RadvylfPrograms Sounds like my Haskell experience

Radvylf Programs

I'm still not really sure if I prefer JS-style stumbling across errors randomly with a decreasing frequency, or Rust-style error: could not compile 'twitch' due to 19 previous errors

DLosc

@RadvylfPrograms This also sounds like my Haskell experience :P

(and Prolog, and Scala, and BQN, and...)

Seggan

@thejonymyster i like this

though id make the rules on which team you go on be based on something immutable, such as user id

otherwise it could turn out to be a 4v28

caird coinheringaahing

You're describing Red vs Blue

Seggan

5:46 PM

ik

DLosc

@Seggan That can also happen if teams are based on user ID, depending on who chooses to sign up

Seggan

true but itd be less likely

imo people would just choose the winning team

caird coinheringaahing

Have you or a loved one had a challenge idea you think is new and creative? If so, you may be entitled to "Calvin has already posted this challenge" duplicate insurance. Call 1-800-CALVIN today

5

Seggan

that does not apply in this case :P

i knew of redVblue

DLosc

@thejonymyster code-dodgeball 🤔

thejonymyster

5:47 PM

@Seggan lmao

@DLosc YES

Seggan

whatd it be

try to avoid the attacks of others?

thats koth

caird coinheringaahing

@DLosc OP produces a piece of malicious code, answers have to edit 5 characters at a time to make it safe

DLosc

@Seggan Yeah, probably

Seggan

@cairdcoinheringaahing and the winner is...?

caird coinheringaahing

(Don't actually do that, challenges around malicious code are banned)

DLosc

5:49 PM

@cairdcoinheringaahing Sounds like answer-chaining code-bomb-defusing

caird coinheringaahing

@Seggan Everyone, as we all had fun :P

Seggan

@DLosc also answer-chaining

anyway i feel like writing up a ctf challenge now

DLosc

@cairdcoinheringaahing I don't need Calvin insurance, since I've never thought of a challenge as creative as one of Calvin's :P

@Seggan Hmm yes, answer-chaining answer-chaining code-bomb-defusing ;)

thejonymyster

@cairdcoinheringaahing radiation hardening koth?

Seggan

@DLosc oops :P

caird coinheringaahing

5:52 PM

@DLosc Is answer-chaining answer-chaining where the nth answer must continue the chain from the nth posted answer-chaining challenge?

One challenge format I've played around with is "answer tree", which is like answer chaining, but multiple answers can shoot off from the same answer at any point in the chain. The biggest problem I've had with it is that the SE format really doesn't work for it

DLosc

@pxeger So like, on their turn each coder can add or change one character of their program? Could be interesting. They'd probably need to use the same language, tho.

Seggan

@cairdcoinheringaahing kinda like Evolution of OEIS?

caird coinheringaahing

@Seggan Yeah, that challenge is why I think the format doesn't work on SE

DLosc

@DLosc Hmm, and it wouldn't have to be strictly "shortest program wins"; it could be "first person to complete a working program (without losing on time) wins." Because of the turn-based nature, it's likely the first working program would also be shorter than whatever the opponent was trying to write.

Is there a default I/O rule about taking string input as a list of codepoints? I can't seem to find anything.

Seggan

iirc there was

doesnt keg do that

Radvylf Programs

6:12 PM

Damn, I was hoping there wouldn't be another mod election until late 2024

~~I guess I'll just need to make sure one of the current mods conveniently vanishes in the next two years~~

thejonymyster

answer chaining idea: modify the previous program to be a solution to the problem in another language, score is bytes added, deleting bytes is free, minimize your score

Radvylf Programs

But early congrats to caird on becoming a mod next month :p

Wezl'

(seconded :)

Seggan

6:24 PM

~~eh, radvylf would be better~~

Radvylf Programs

If I could run I would :p

(would run, not would be better. caird would probably be better :p)

thejonymyster

6:54 PM

@thejonymyster oh my god, and this can be both self validating and self scoring if the challenge itself is to take two strings and count how many character additions it takes to get from one to the other LMAO

idk if thats necessary but itd make some fun tags

or wait, maybe not both of those

itd be more cohesive but probably make the challenge less good :P

Wheat Wizard

Challenge idea: You have

thejonymyster

It's true, I do.

caird coinheringaahing

7:16 PM

@RadvylfPrograms This is good timing, means that 2024 is the likely date for the next election :P

user

Would it be acceptable to award a bounty immediately after 24 hours even if the answer only has 3 upvotes?

Newly Featured Posts

7:35 PM

15

Q: The second even sublime number

easy mode of my previous challenge A perfect number is a positive integer whose sum of divisors (except itself) is equal to itself. E.g. 6 (1 + 2 + 3 = 6) and 28 (1 + 2 + 4 + 7 + 14 = 28) are perfect. A sublime number (OEIS A081357) is a positive integer whose count and sum of divisors (including...

code-golf math kolmogorov-complexity number-theory

Steffan

7:47 PM

8

A: What's a string?

If it looks like a string and acts like a string, it's a string Examples of strings: Native string types: C++ strings, Java Strings, JavaScript Strings, etc. Native character types: C/C++ chars and wchars, Java chars and Characters, etc. An iterable (tuple, array, vector, list, etc.) of characte...

thejonymyster

@user What's the bounty for?

user

Being an answer to an old question

...and I just realized it wasn't eligible for my bounty but whatever, a lot of effort was spent on it

thejonymyster

@user I don't see what would be bad about awarding it immediately, why would that make a difference?

user

Well, less people would see the answer and upvote then

thejonymyster

what are the points in the bounty for then :P

user

7:55 PM

Yeah I guess so

Steffan

A lot of Ajax1234's answers might be eligible for your bounty

user

I'll check those out then

Steffan

Oh wait, he does unanswered ones, and I guess you excluded them

95% of them are to unanswered old hard questions

user

Yeah there's a good bounty for unanswered ones already

How does this person do that?

Steffan

idk

the question is 3 days old lol

user

8:03 PM

Whoops

I'm gonna put that on you Steffan

You guided me to Ajax1234 :P

Wezl'

lol want a refund bounty :P

I mean rep's not exactly meant to be a currency but

thejonymyster

lol

thats definitely how we use it

emanresu A

8:39 PM

@user "old" questions with few answers

Steffan

are u trying to drain ur rep or what

user

8:55 PM

@emanresuA Yeah I didn't realize

Steffan pointed that out above

@Steffan Yes

hyper-neutrino

oh? why's that

user

This account may become abandoned later so I'm just making the most of it while the rep is still accessible

Unrelated String

9:13 PM

wait what

user

I've been spending too much time on CGCC, I'm going to take a break for a month or so and if it turns out I'm fine with it I might just leave entirely

caird coinheringaahing

Traitor smh :P

user

Case in point: I've been switching to this tab like every 15 seconds. Every few minutes I shut it down and then two minutes later I log back in (I'm opening in Incognito as a mostly failed attempt to make it a bit harder for me)

Unrelated String

...understandable

user

In other news, Math.SE has cooked up some more drama

What is it about math that drives people to such extremes?

emanresu A

9:19 PM

They just refuse to accept that 1 plus 1 is 3.

Unrelated String

everything

Steffan

9:32 PM