The Nineteenth Byte

exedraj

14:00

@BrowncatPrograms Yeah but prefix has the luxury of having expressions grouped together

That doesn't exactly happen in stack land

Anyhow

Browncat Programs

It doesn't use that

exedraj

I'm going

Browncat Programs

That's not how it works

exedraj

It's midnight

Browncat Programs

o/

exedraj

14:00

And I'm tited5

user

\o

exedraj

*tired

o/

Neil

@BrowncatPrograms replace should be shorter than map...join, no?

user

@exedraj No you're not, you're lyxal

Browncat Programs

@Neil Forgot about that, yeah

If anyone's interested, the way Ash's dynamics work (the name for any operators that take a function) is that they have an implicit {

So it's just a normal function, but using a dynamic (where a function would be necessary anyway) starts the function implicitly

Neil

14:04

@PyGamer0 Retina, 19 bytes

Browncat Programs

@BrowncatPrograms So there's no prefix weirdness, it'd work just as well with postfix, stack based, or other types of tacit

Neil

@AaronMiller Charcoal, 11 bytes: ∧⁼ηＩΣηＩ⌕θΣη Takes x and y as a list and outputs "", "0", "1", or "-1".

Sandbox Posts

14:25

0

A: Sandbox for Proposed Challenges

A phonetic letter input a letter of the alphabet, A-Z (upper or lower or a mix, I don't mind) output the phonetic (i.e spelled name of the )letter of the one passed in, along with the appropriate indefinite article a or an (again, in any case) full set of test cases (lowercase) - see This english...

Unrelated String

@exedraj my third ever answer was in brachylog and the first two were in clojure lmao

user

@exedraj I think 05AB1E, although I only used that once. My first "regular" esolang would probably be Husk

Wheat Wizard

I don't think I've ever used a golfing language.

user

Huh

Unrelated String

neat

user

14:36

What's the first esolang you used?

Wheat Wizard

Brainflak.

It's just the first language I used.

Actually I seeems I have some Husk answers:
https://codegolf.stackexchange.com/a/211958/56656

I don't recall making this but I have apparently used Husk before.

Dude coinheringaahing

6

Q: User card is missing padding between username and reputation in review queues

In the Close Votes and Low Quality Posts queues, when reviewing a question, the padding between the username and reputation seems to be missing:

Browncat Programs

14:51

@BrowncatPrograms Dang, I can't post a bunch of resources and studies about the effectiveness of the COVID vaccine in their google classroom, because they disabled class comments

user

lol

@WheatWizard For LotM?

Browncat Programs

15:12

Oh no...someone in my class is trying to write a program to translate Python to Java, with no knowledge on how either language is parsed/interpreted and no relevant libraries

2

PyGamer0

@BrowncatPrograms tell him that i said: "good luck :P"

user

@BrowncatPrograms ...Why?

Not only is that hopelessly doomed, it'd be about as useful as umbrellas with sprinklers

Browncat Programs

No idea. I'd ask them, but I really don't want to be the one to crush their dreams.

pxeger

Tell them to just use Jython

Browncat Programs

15:38

  public Vehicle(int n) {
    location = n < -20 ? -20 : n > 20 ? 20 : n;
  }

  public void forward() {
    location = new Vehicle(location + 1).getLocation();
  }

  public void backward() {
    location = new Vehicle(location - 1).getLocation();
  }

You can tell I didn't want to re-write the ternary mess :p

Even though it'd be way easier

But this is more abstract

Wheat Wizard

Isn't that just min(20,max(-20,n))?

Browncat Programs

Yes

But we're not supposed to use those yet :p

Oh nice, they're finally teaching us about comments 45% of the way through the class

much comment, very advance

Aaron Miller

Browncat Programs

Well frick

The instructions for our assignment are on github

And github.io is blocked here

Aaron Miller

.-.

Browncat Programs

15:57

Okay I've figured almost everything out from running it and seeing what errors or is marked incorrect

Can someone go to amooc.github.io/right-triangle/RightTriangle.html and tell me what formatting I'm supposed to use for toString?

Aaron Miller

404

Browncat Programs

Uh...well then I guess I'm really not going to get a 100% on this lol

user

Even github.com/amooc 404s

Aaron Miller

F

Browncat Programs

I think I might've found a cached one

> A String representation of this right triangle. This is in the form "right triangle with base a, height b, hypotenuse c"

Found it!

Aaron Miller

16:03

\o/

user

🎊

Aaron Miller

🎉

user

@BrowncatPrograms Why tf are you literally making a new Vehicle object just to change the location?

Browncat Programs

Abstraction, of course

user

That's not abstraction lol

Browncat Programs

16:06

We don't need to know how the location is determined, we just try to put a car there then figure out where it ends up

This is obvious and very good practice

Aaron Miller

lol that's exactly how I drive

Browncat Programs

._.

user

Aaron Miller is 1D convirmed

No right() or left()

Aaron Miller

No, I am 2D because am plate. My car is 1D.

user

How does a 2D object fit in a 1D object?

Aaron Miller

16:08

Practice

Browncat Programs

Linear transformations

user

Unless your personality is so one-dimensional you can be compressed into a single dimension

Aaron Miller

How many dimensions is a plate-like personality?

Browncat Programs

Cursed languae feature: You need a getter to get methods from an object

car.getMethod("forward")(1)

That prevents risks from being able to directly access the method, of course :p

user

*Haskell has entered the chat*

(idk how optics work in Haskell but I hate the record syntax and I hate Haskell)

(Haskell is pretty cool though)

Aaron Miller

16:11

That reminds me: does anyone have any ideas for what Dysfunctional Functions™ means? I have no idea, but its name is good enough that it gets to be included in Ｃｕｒｓｅｄ.

Dude coinheringaahing

@BrowncatPrograms Isn't that just JS's [...] constructor syntax?

user

@AaronMiller Like a dysfunctional family, they don't always listen to each other (certain functions can't be called from certain other functions)

Aaron Miller

@user How do we communicate this to the user?

user

ERROR: `foo` isn't talking to `bar` right now because `baz` cheated on `foo` with `bar`. Their parents `fred` and `waldo` are also going through a divorce right now, so please try to understand the situation.

pxeger

that's a lot to unpack

Aaron Miller

16:15

lol what determines if a function can talk to another function or not?

Dude coinheringaahing

Their names

user

Random

Dude coinheringaahing

If they share 3 or more letters in their names, they can talk to each other :P

user

@BrowncatPrograms tio.run/##fZExT8MwEIX3/…

pxeger

but then you could just prefix every name with abc_

Dude coinheringaahing

16:16

@user I've found that randomness is a boring way to make something cursed

user

@user Here's a start to how you should be doing abstractions

@Dudecoinheringaahing True

@pxeger Well you do need workarounds to the cursedness

Aaron Miller

@Dudecoinheringaahing Meaning that functions with names that are less than 4 characters will be unable to communicate with each other. That's evil and I love it.

user

The fact that all functions start with that is cursed

@AaronMiller Also, if they share more than 90% of their names but not 100%, they hate each other too

Dude coinheringaahing

Or, if they share 3 or more letters in their names, but those 3 letters can;t be in the same order :P

So abcd and adbc can talk to each other, but abcd and abce can't :P

Aaron Miller

@Dudecoinheringaahing Yeah, I've been trying to avoid unintended randomness and arbitrary restrictions in the cursedness.

I like these ideas. I'll probably combine them into some kind of Cursed Dysfunctional Functions™ stew.

Browncat Programs

16:19

^^^^^^^^^^

Also, trying to force the cursedness is the fastest way to make something uncursed

It's cursed when it seems unintentional or just borderline enough to hit the uncanny valley of cursedness and normality

Aaron Miller

lol true

now I'm trying to figure out if that was unintentional or not

Dude coinheringaahing

Yeah, the best cursed features are stuff that makes sense by the letter of the feature, but make you go "wtf, why?"

Browncat Programs

Sort of like my idea of having integer division error with a remainder; on the very surface it seems potentially something you'd do, but it's just annoying (that's not a very good example, since it still feels a bit contrived)

Aaron Miller

Switch statements that use gotos instead of breaks

pxeger

switch statements but the switch and case keywords are swapped

Browncat Programs

16:24

Switch statements in one of my languages just move the switch block above the switch statement, and turn switch into goto

Aaron Miller

Switch statements but instead of normal cases it's regex matching

Browncat Programs

That's useful

Dude coinheringaahing

Switch statements but switch switches the case of the value. So switch {"Hello, World!"} case "Upper" returns HELLO, WORLD! :P

Aaron Miller

Indenting lines with ZWSP to settle the spaces vs tabs debate

Browncat Programs

Right aligned code

Dude coinheringaahing

16:27

Indent with backspaces

Aaron Miller

@Dudecoinheringaahing I like this

pxeger

@BrowncatPrograms you mean with a Unicode RTL marker?

Browncat Programs

No

Dude coinheringaahing

Use spaces and tabs to right-align them :P

Aaron Miller

Input and Output are variables you read from and assign to, but any variable that includes in or out in the name will work, so a variable like isLooping will actually just provide input.

pxeger

16:30

‮abcdefghijklmnopqrstuvwxyz

Browncat Programs

Dude coinheringaahing

		def fib(n):
	a = b = 1
	for _ in range(n):
a, b = b, a+b
	return a

Browncat Programs

@BrowncatPrograms There's something kinda nice about this actually

Aaron Miller

@Dudecoinheringaahing So increasing the indentation effectively decreases how indented it is?

Dude coinheringaahing

Kind of, yeah :P

Aaron Miller

16:34

I'm gonna refer you back to the language motto:

in ＴｈｅＭｏｓｔＣｕｒｓｅｄＲｏｏｍ, Aug 10 at 15:33, by Aaron Miller

That's a horrible idea.
Let's do it.

One especially cursed idea we came up with is having regex syntax similar to JS's, except that instead of being delimited with /, it's delimited with r. That doesn't seem super cursed on its own, until you realize that some keywords have an r in them, so a keyword like During would have to be written as Du!ring (the escape character is ! (except when escaping ! (then the escape character is \ (like in most languages))))

Ectogen

@BrowncatPrograms the same code but golfed (if it was Javascript, that is.) (can probably be golfed more): n=>[...Array(n).keys()].filter(e=>e%2)

Browncat Programs

No, my function takes an array

It'd be n=>n.filter(x=>x%2)

Ectogen

@BrowncatPrograms Ohh I didn’t see that hehe (yep that’d be the golfed version then)

Aaron Miller

16:49

Are there any languages, esoteric or otherwise, where 0 != -0?

Browncat Programs

Not to my knowledge, I think IEEE 754 even says they should be equal

Dude coinheringaahing

I think JS, but -0 doesn't actually create a signed zero. You can, I believe, only do that with fp errors

Browncat Programs

No, 0 == -0

Dude coinheringaahing

> -0 doesn't actually create a signed zero

pxeger

console.log(-0) -> -0

Browncat Programs

16:53

@Dudecoinheringaahing Wherever you got that, it's wrong

Adám

But toString() gives "0"

Browncat Programs

Adám

@BrowncatPrograms Even 0===-0 gives true

Browncat Programs

Yes, like it should

Dude coinheringaahing

@BrowncatPrograms It's JS, all my knowledge comes from half-misremembered SO questions :P

Browncat Programs

16:54

@Adám Object.is does not though

Adám

Huh, Math.sign(-0) gives -0

Browncat Programs

Yep

Aaron Miller

why does ![] == []?

Browncat Programs

Because [] is cast to a boolean

The more important question is why new Boolean(false) is truthy :p

Aaron Miller

But if [] == false, why does ![] == false?

Adám

16:58

Because JS gotta be JS.

Browncat Programs

new Boolean(false) == false is true

But new Boolean(false) is truthy

That's so cool

@AaronMiller Because arrays are truthy

JS probably uses a separate truthiness algorithm for ==

Wait no

They're probably cast to numbers

Aaron Miller

ah

Adám

Aaron Miller

In Ｃｕｒｓｅｄ, -0 and 0 will both effectively equal 0, but will be unequal to each other.

Browncat Programs

There are for equality algorithms in JS

pxeger

17:05

four?

Browncat Programs

SameValue, SameValueZero, Abstract Equality Comparison, and Strict Equality Comparison

SameValue, used by Object.is, is the only one that considers 0 to be different from -0

TIL:

Adám

Why are there no <<< and >>> and <== and >== as alternatives to < and > and <= and >= but without coercion?

Browncat Programs

>>> is already in use

Adám

Oh, I didn't know.

Browncat Programs

And I doubt any of those are useful

Adám

17:10

(![])<[]
false
[]<![]
false

⍨

[]==![]
true
[]===![]
false

Browncat Programs

What should it return if it doesn't cast them and they're unequal types?

Aaron Miller

Object.is(-0, 0) === false

Browncat Programs

It's not like equality where you can just state that they're not equal

Adám

Either it should implement a total order, or error due to incompatible types.

Browncat Programs

I've never run into and never heard of anyone running into an error due to < and friends doing casting

@AaronMiller See above, already mentioned

Object.is uses SameValue

Which means it also considers NaN to be the same as NaN

Whereas NaN === NaN is false

Aaron Miller

17:13

oh, missed that

tfw x = NaN, and x != x

Browncat Programs

Okay, so what's happening with ![] == [] is that [] is cast to a primitive, and ![] (false) is cast to a number (0), then they're compared

So evidently [] is 0 as a primitive, which makes sense

It's probably stringified to "" then cast to a number, which would be 0

Aaron Miller

i.e. it's magic

user

@Adám Ooh, this is very helpful

Browncat Programs

Interesting, SameValueZero still considers NaN == NaN to be true

So [NaN].includes(NaN) is true

user

@AaronMiller I really like this, because if you allow assigning the groups to names, it would actually be very useful

But horribly unintuitive and inconvenient too

pxeger

17:24

and, if you tried to use it on anything other than strings, horribly error-prone

Browncat Programs

Also, IEEE-754 has multiple NaNs, which can be accessed using TypedArrays, meaning that you can have two NaNs which are not identical with Object.is

Fun

user

@pxeger Well I was thinking of making it stringify everything

pxeger

well that's what I mean

user

Oh yeah

pxeger

imagine trying to regex-match the string representation of an array

or, worse, a float

user

17:25

Not too bad honestly

Aaron Miller

@user I'm not sure which groups you're referring to

Browncat Programs

Actually wait, they are identical with Object.is

More fun

user

@AaronMiller Captured groups

$1, $2, and so on

Oh and because we believe in functional programming, everything should be an expression, including statements like goto

pxeger

but in Python and other flavours you can name them like (?P<my_group>.*)

@user so does goto return the bottom type?

user

There's no return keyword, the last expression is implicitly returned

pxeger

17:27

that's not at all cursed

particularly for a functional language

user

@pxeger I was thinking of the result of executing that label but that works too

Browncat Programs

But...goto doesn't have a last expression

user

@pxeger I know

pxeger

@BrowncatPrograms that's why it returns the bottom type!

user

I like that about Scala, but in a language like this, it would get confusing

@BrowncatPrograms Like if you have ;label; <some expr> and then goto label, it would return <some expr>

All blocks are expressions that return the last expression in that block

Browncat Programs

17:28

goto should return the machine code/bytecode generated for the expression being jumped to

Aaron Miller

@user You're in charge of implementing the regex and pattern match and whatnot, so I'll leave it up to you. I like the idea, though.

user

Great, thanks!

What do you think about statements as expressions?

pxeger

but if it's being jumped to, so it can't possibly return anything?

Browncat Programs

I was about to point that out yeah

pxeger

unless you want it to implicitly jump back, which just sounds like function calls with extra steps

user

17:29

@pxeger After you jump there, it comes back to the goto ;)

Aaron Miller

@user I like it, if we can figure out the logistics of it.

Browncat Programs

@user That's useful

pxeger

@user sounds like function calls with extra steps

Browncat Programs

@BrowncatPrograms We call them "functions" :p

user

> usefuk

Ah true, Redwolf's idea might be better

Browncat Programs

17:30

goto can't return anything though

user

Or pxeger's, although just returning ()/Unit/never whatever feels lazy

Browncat Programs

When you goto, you go somewhere else and can't use the return value

user

@BrowncatPrograms Maybe gotos start a new thread

Aaron Miller

The goto statements don't return implicitly, but there is a label called ;camefrom; which will return to the last place jumped from, so we could make it return something if we wanted.

user

So in the current thread, they return the bottom type, and it goes to the new label in a new thread

@AaronMiller That's good

pxeger

17:31

@user like Unix fork?

but with threads

user

I guess?

Or Go's go, I think

Aaron Miller

@user that sounds similar to what I was thinking for asyncgoto

user

Nice

Maybe asyncgotos are allowed inside normal (non async) functions but they just fail horribly

Somebody please write all of these ideas down, they're all deliciously evil

Aaron Miller

Also, implicit conversions are going to do some weird stuff, for example it might do something like cast to an int if possible before casting to a string, so something like null will cast to the string "0" instead of "null"

user

And you can define your own implicit conversions, like in Scala

That might also slow down programs by a ridiculous amount and make them hard to debug

Aaron Miller

17:35

@user asyncgotos will probably be the only async stuff in the language, they'll simply start a new thread or something, but continue execution in the current thread as well.

user

Can threads be joined or awaited or deleted?

Aaron Miller

Nope, they run until they hit a statement to terminate them (not sure what that will be yet), but they will run alongside the other threads, so they can change variables and stuff for the other threads.

user

Oh no

Aaron Miller

Oh yes

And all variables are global

user

I think that's a little too cursed

Browncat Programs

17:40

This sounds like assembly with more compexity :p

Aaron Miller

@user the async, or the variables?

Browncat Programs

compexity™

I am spell

user

Plus, with scopes, we can have Python-like cursedness where you need nonlocal foo to assign to foo instead of creating a new variable foo

@AaronMiller All variables being global

@BrowncatPrograms Hi spell I'm user

Aaron Miller

@user nope, they're all global

:)

user

grrr

Aaron Miller

17:42

It makes it easier for the user, since they don't have to keep track of scopes and stuff within all the functions and gotos and stuff. :)

user

lol

user

17:58

github.com/minimaxir/big-list-of-naughty-strings

Browncat Programs

( ͡° ͜ʖ ͡°)

Also I'm not sure how I feel about instantly recognizing every string there that taken from the wikipedia page on the scunthorpe problem

user

I don't understand "evaluate" and "expression". Are there languages where those are functinos/reserved keywords?

Dude coinheringaahing

18:15

@DLosc Why not go back to Snoopy for dog week? :P

Dude coinheringaahing

18:53

Feel free to brandish your pitchforks: Top bar is now sticky for everyone

pxeger

If I had the rep, I'd edit that to be a bug

Browncat Programs

For every downvote on CGCC (on undeleted posts, I think), there are 164 upvotes

Wait no

I'm bad at sql

Wait yeah

I'm not bad at sql

It's 164

Query:

select cast(count(case votetypeid when 2 then 1 else null end) as float) / count(case votetypeid when 1 then 1 else null end) from votes

The ratio on CGCC meta is slightly lower (as expected), around 124

And on stackoverflow it's 14 :p

Dude coinheringaahing

@BrowncatPrograms votetypeid = 1 is accepts, not downvotes

Browncat Programs

Aw :/

Dude coinheringaahing

The ratio is 19.3 :P

data.stackexchange.com/codegolf/query/1452725

Browncat Programs

19:07

Yeah yeah shut up :p

user

Why did they even do that?

Browncat Programs

On SO it's around 7.6

On our meta it's around 6.3

user

> Edit: from your comment reply to Sonic pointing out this SO bug report, it sounds like the issue at hand there was that your own team was not aware of the feature. That is not really a bug, so I'm guessing there are other issues.

Aaron Miller

@Dudecoinheringaahing I didn't even know that could be disabled :p

Browncat Programs

On MM it's around 3.57

Dude coinheringaahing

19:09

Here's a fun one :P

Enter 2, 6 or 10 to find the ratio of a type of vote vs its opposing vote (up/down, close/reopen, vtd/vtud)

pxeger

6 gives me a division by zero error

ngn

@Dudecoinheringaahing i don't have enough rep to downvote so i'll just wave my thumb down in the air

Dude coinheringaahing

Wait yeah, they stopped counting close votes that way in 2013

For every vote to undelete here, we have 4 votes to delete

user

@ngn Post something in the (MSE formatting) Sandbox so I can bounty it and you'll have enough rep to downvote :P

Dude coinheringaahing

@ngn Downvoting requires 101 rep on MSE

Browncat Programs

19:11

Windows phone has a ratio of 5.5:1 for up:down

> We didn’t change the default. We’ve only removed the ability to change the default.

Ah, reassuring

pxeger

??? that means literally nothing

Browncat Programs

> We didn't change something that takes you ten seconds to change back, we changed something forever and frick you

ngn

@Dudecoinheringaahing i have 99 since monica and stuff :)

Dude coinheringaahing

Got to love downvoting answers :P

pxeger

> We didn't tell your child your house burnt down. We've only burnt down your house

ngn

19:14

if downvoting is -1 rep, why isn't upvoting +1?

Browncat Programs

Hmm, I wonder why :p

ngn

rhetorical question :)

@user this sandbox? it seems i don't have enough rep to add an answer :) but thanks

Adám

32

A: New Community VP seeks your input - what should we be looking to change and what is inviolable?

One thing that you would advise me (in my role as VP of Community here) never to touch. Stack Exchange… …without asking the community first. The core community is tired of radical changes being announced as they happen. One thing that you think I should change as quickly as possible. Stop l...

Dude coinheringaahing

@Adám I'll make sure to ping you the next time a change is made so you can add it to that :P

Adám

@Dudecoinheringaahing Thanks. Is the existing list there missing any old ones you can think of?

user

19:21

@ngn Oh, that's too bad

Dude coinheringaahing

@Adám Not that I can think of

The most recent profile changes (not shipped yet) were demoed for feedback, I guess that wouldn't count

Aaron Miller

How did the topbar used to act when you disabled the sticky?

Browncat Programs

Probably was just static

Adám

@AaronMiller Like on GitHub.

Browncat Programs

I'd make a userscript to fix it, but this is really an issue for userstyles

ngn

19:30

@AaronMiller it hid when you scrolled down

Aaron Miller

isn't it pretty much as simple as a .top-bar { position: static; }?

Wezl

@ngn can you take into account castling (whether the king and knight have moved)?

ngn

@Wezl it's 4 bits: can {white,black} x {castle queen-side,castle king-side}

Wezl

wow, you answered quickly after I responded two days later

Aaron Miller

20:03

Is there a way to tell in a userscript when the chat finishes loading, like some sort of event or something? window.load and document.load are too soon, and not everything has loaded yet, and adding some arbitrary setTimeout feels really jank.

user

20:16

DOMContentLoaded doesn't work?

There's MutationObservers, I think

Aaron Miller

I forgot about DOMContentLoaded. I'll try that.

Hmm... I have
window.addEventListener('DOMContentLoaded', (event) => {
console.log('DOM fully loaded and parsed');
});
This was taken directly from the MDN docs, but it doesn't seem to execute at any point.

Wait, does the fixed font thing only work if the whole message is fixed font?

emanresu A

Hi

gyes\

Aaron Miller

20:32

hOI!!!

emanresu A

G'day mate

user

@AaronMiller Yes

Aaron Miller

oof

I'm going to try MutationObservers next. From the docs, it looks like that might just do it

Hey, MutationObservers will actually fix a problem I was having with one of my other userscripts. Nice!

DLosc

20:58

@Dudecoinheringaahing I'm gonna return to Snoopy eventually. As long as we're having special weeks, I figured I'd rep some other comic characters I like.

If we do sharkboy/lavagirl week next, it'll get harder, but I have a couple ideas :D

Unrelated String

21:18

lol

Browncat Programs

21:41

@AaronMiller Use @run-at document-start

Otherwise the script doesn't run until the DOM's ready

But the default behavior is to run when everything's loaded, so you might not need to do anything

emanresu A

21:52

When there are eight copies of the same answer

Dude coinheringaahing

What about 22 :P

(tho not really, the search isn't a great one)

emanresu A

More like 3.

Have you coinheringaahed today?

Dude coinheringaahing

Yeah, I do my daily coinheringaahings when I wake up :P

5

Transcript for