The Nineteenth Byte

Redwolf Programs

12:19 AM

how to juic advogad

user

Avogadro looks like he tried juicing an avocado for well over 30 minutes, then gave up and tried chemistry instead :P

ATaco

Yeah but then he went and made my chemistry text book like, twice as thick.

user

Or maybe it's just a really unflattering painting

Adám

@user But how?

user

@Adám How would you expect me to know if even such a great chemist could not defeat The Avocado?

Adám

12:23 AM

Pleas.

user

try for ~6.022E+23 minut insted of thirtee minut

mak sur two rmove pulp

Adám

Where is the original post?

user

Best I can do

lyxal

I've had an idea for juicing avocad - heat avocad until it melts

user

but wat if advocad bun. wat if advocad sublimet

ATaco

12:26 AM

Okay someone try heating it up to 7-9

3

user

HOW TO MAKE AVOCADO JUICE

►

OMG THEY CRACKED IT! THE MYSTICAL FORMULA TO JUIC AN ADVACADO!

lyxal

The idea is that you never cut the avocad, no?

ATaco

That's technically an avocad smoothy

Adám

Yeah, not juice at all.

user

Alas, it appears mankind will never achieve avodad juic, teh nector of the gods

What would really make it juice, though? Would it just have to be thinner than a smoothie?

Adám

12:30 AM

You'd need to separate the liquid (probably mostly water, and a little oil) from the dry parts.

Wezl programs redwolf

you can tell whether it's really juic by whether the answer is accepted

@Adám great, now I'm sucked in an xkcd rabbit hole that I won't escape until it's too late to accomplish anything more today

Adám

\○/ but at least there are a finite number of xkcd cartoons. See you next week!

user

1:24 AM

Good thing it’s not an SMBC rabbit hole, there’s a lot more of them

Wezl programs redwolf

Super Mario Bros. Crossover? :P

hyper-neutrino

2:24 AM

Planned Maintenance: Sat. Oct. 30 13:00-15:00 UTC, Sun. Oct. 31 13:00-16:00 UTC

Sandbox Posts

0

A: Sandbox for Proposed Challenges

Fibonacci-like gap formula code-golf math fibonacci Background The recurrence of the Fibonacci sequence is defined as $$ f(n+2) = f(n+1) + f(n) $$ From this recurrence alone, the following gap formulae (recurrences relating three terms with certain amount of gaps) can be derived: $$ f(n+4) = 3f(n...

Redwolf Programs

3:57 AM

@Wezlprogramsredwolf Hey, you stole half my name! :p

Hey waiiit, you stole all of it!

lyxal

Redwolf Programs != Programs Redwolf

two different things

Redwolf Programs

It is a pretty good name, so I'll let it pass

But anyway, I have some sort of test tomorrow so o/

lyxal

o/

PyGamer0

^[

Redwolf Programs

I'll be in room 2214, if I forget now it's in the transcript :p

Hey that better not be my english classroom

Okay good it's not

2211 is the one I'm thinking of

lyxal

4:01 AM

hey so quick question, what happens if the room where you do your exam is suddenly destroyed before you get there?

do you still have to do the exam?

@RedwolfPrograms

Redwolf Programs

Probably, it's a big school, they'll find room :p

lyxal

dang

Okay better question: what happens if the school is destroyed?

do you still have to do the exam?

Redwolf Programs

Then...probably not

But I have a lot more to worry about in that case :p

lyxal

@RedwolfPrograms nah you'll be fine

I'll go get my teleporter and send my fridge to your school

Redwolf Programs

Don't you mean I would be fine? Right? Right...?

lyxal

4:03 AM

yeah you'll be all good so long as you stay outside the blast radius/radiation zone

because when my fridge is inevitably nuked, the school will be destroyed as well

Redwolf Programs

Yeah no worries I'm about 4 miles from the school

lyxal

oh that's perfectly fine

Redwolf Programs

It's pretty far from other structures, but you'll probably irradiate a nearby elementary school

lyxal

@hyper-neutrino next time you nuke my fridge, make it non-radioactive for redwolf's sake

there

now there's no radiation

Redwolf Programs

I want to take the PSAT though D:

lyxal

4:05 AM

oh

yo hyper ignore my request

Redwolf Programs

I gotta get a funny number again

lyxal

I'll be sending my fridge elsewhere

Redwolf Programs

You should mount a missile defense system to your fridge

Then sit in it

And have a friend push it into a lake

And then you have a submarine

But anyway I need to go to sleep o/

lyxal

why would I make the fridge immune to nukes?

that'd ruin the local fridge industry

gosh redwolf, think of the economy gosh dang it

Redwolf Programs

You're not in the pocket of Big Fridge, are you?

I thought you were someone we could trust

lyxal

4:09 AM

@RedwolfPrograms I said local

i support my local fridge producers

Redwolf Programs

Ah selling out to Little Fridge is fine

lyxal

exactly

someone has to support the fridge store down the road

Redwolf Programs

It's like my idea to have tire companies hire people to slash random cars' tires

lyxal

Oct 15 at 12:24, by lyxal

the guys at the fridge shop know me pretty well now

Redwolf Programs

@RedwolfPrograms It supports local businesses

lyxal

4:11 AM

Oct 15 at 12:25, by lyxal

heck, they even give me a 0.5% discount on Saturdays and Tuesdays because I've gotten that many new fridges from them ;p

Redwolf Programs

I've heard the mods are considering using chemical and biological weapons against your fridge since nuclear and radiological don't seem permanent enough

lyxal

Big fridge wouldn't accommodate my requests to tune the fridges specifically to the SE chat noise

@RedwolfPrograms well duh. Of course it's not permanent

I just buy new fridges all the time

Redwolf Programs

Watch out then, a Fridge Plague outbreak might start going around in your town

lyxal

oh no

oh frick

Redwolf Programs

Anyway o/ again lol

For real this time

lyxal

4:13 AM

o/

Redwolf Programs

I'm even going to click "leave"

See y'all

lyxal

o/

hyper-neutrino

> nuke

> non-radioactive

now, i'm not particle scientist, but

emanresu A

Hello

Is it just me or did chat go down for about two minutes just then?

lyxal

idk

PyGamer0

4:21 AM

@hyper-neutrino but ur hyper-neutrino

emanresu A

Also I discovered that Firefox completely bricks itself if you try to save a jpeg as png while opening a file.

Don't ask.

hyper-neutrino

4:35 AM

@PyGamer0 ok, and?

PyGamer0

@hyper-neutrino ur a particle

hyper-neutrino

are you an anthropologist

PyGamer0

@hyper-neutrino idk whats that

so NO

hyper-neutrino

anthropology is the study of humanity

so if you, a human, don't study humans

PyGamer0

ok

hyper-neutrino

4:37 AM

then it is perfectly reasonable that I, a particle, don't study particles :P

4

PyGamer0

lol

lyxal

@hyper-neutrino I more meant next time you go to nuke my fridge, switch out the nuke for a standard bomb

That way, you don't actually nuke it

You just destroy it without the radiation side effects that usually happen

hyper-neutrino

6:07 AM

→ 1 message moved to The Nineteenth Bakery

pxeger

7:32 AM

@RedwolfPrograms I believe the mafia used this exact technique, in fact

@AviFS the past few weeks I've not been working on ATO, but I'd like to get it into a fully feature-complete state soon™ enough

I've recently decided to rewrite the entire backend using a language that isn't so utterly terrible at concurrency

emanresu A

What? Weren't you using Typescript?

pxeger

7:56 AM

that's for the frontend

the backend is in Python

Python's async doesn't work well with lots of complicated IO (combining an http + websocket server with writing to files and executing shell commands = somewhat painful and very slow)

rak1507

what lang are you gonna rewrite it in

emanresu A

@pxeger Use node - It's practically made for those sorts of things.

pxeger

@rak1507 probably Go

emanresu A

The classic example is a Hello, world webserver

And then you have full-stack JS which is kinda cool

pxeger

@emanresuA yeah I was considering that but I want to use this as an excuse to learn a new language

@emanresuA ...and kinda sucky. I don't really like the js ecosystem either

emanresu A

8:02 AM

... If you're going to learn a new language, don't do it for something that requires actual security and consistent functionality.

pxeger

@RedwolfPrograms what's wrong with "unwanted commercial content or spam"?

@emanresuA you say that as if ATO has either of those things already

emanresu A

@pxeger ಠ_ಠ

Ugh userscript slow

pxeger

and hopefully Go is well-designed enough that it's hard to get security massively wrong

emanresu A

I suppose it makes interacting with Docker easier

pxeger

not Docker, a custom sandbox

Docker's startup times are too slow

emanresu A

8:05 AM

If it's fast enough for redwolf... Oh well ¯\_(ツ)_/¯

I just realised you have Whython on there

pxeger

@emanresuA yeah, seeing that he's using it is making me reconsider, and I might switch to something else soon: github.com/attempt-this-online/attempt-this-online/issues/31

emanresu A

Idk much about setting up online interpreters, your choice

rak1507

@pxeger yuck

Bubbler

From "golang vs rust" search result:

> Rust has a much steeper learning curve compared to go. However, it is also important to note that Go has a much steeper learning curve than other languages such as JavaScript and Python.

pxeger

@rak1507 why? Go seems pretty well suited for this task, and that's all that really matters

Bubbler

8:12 AM

If I picked up Rust successfully before Go, does it make me smart?

pxeger

@Bubbler I was gonna disagree with that, but I think the problem is that I think any language I have already learnt has a horizontal learning curve, and any language I have given up on has a vertical one

Bubbler

Which ones are vertical for you?

rak1507

@pxeger go generally sucks in lots of ways from what I've seen

emanresu A

Jelly has a half exponential and half logarithmic learning curve.

pxeger

@Bubbler in this case, Rust, because of the borrow checker

Although in this case it's a me problem because I never tried to understand it and just tried to fumble my way through

emanresu A

8:18 AM

Btw, would anyone like to see the eighth book of Harry Potter?

(Markov chain from the previous ones)

Bubbler

No

emanresu A

Ooh, idea: Markov chain programming!

I kinda want to generate some Vyxal programs like that and see what happens

PyGamer0

@emanresuA this?

pxeger

@emanresuA youtube.com/watch?v=6rEkKWXCcR4 ?

goddamn oneboxxing

PyGamer0

@emanresuA cursed child?

pxeger

8:22 AM

oneboxing? oneboxxing?

PyGamer0

@pxeger 1⎕ing

emanresu A

Remove the ` ?` and it shoudl onebox

pxeger

@emanresuA I want it not to because it's annoying

I was trying to work out where to put the zwsp

lyxal

@pxeger use Scala and then transpile it to JS

PyGamer0

@pxeger Use Vyxal :P

6

New Posts

8:33 AM

0

Q: Inverse n-bonacci sequence

We all know about the fibonacci sequence. We start with two 1s and keep getting the next element with the sum of previous two elements. n-bonacci sequence can be defined in similar way, we start with n 1s and take the sum of previous n elements to get the next element. Challenge Given a positive ...

Bubbler

10:18 AM

I can't resist doing some Jelly when I see a challenge where œi is the obvious last step :P

lyxal

This is a reminder for people to go upvote new answers to old posts

It's easy, free and takes only a small amount of time

emanresu A

Agered

Lol this was one of my first answers and it somehow has +8

lyxal

My first answer has +4, which mostly came about a year or so later

pxeger

10:34 AM

Mine (+4) is from when Grain Ghost was Ad Hoc Garf Hunter

emanresu A

@lyxal The compression in that one's interesting. How does it work, and why didn't you make those strings autocomplete?

pxeger

@emanresuA the actual answer itself doesn't have a close delimiter; it's only there in the explanation

so I'm guessing they do autocomplete

lyxal

@emanresuA they do

emanresu A

Oh. Forget the second half, but how did Keg's compression schematic workin general?

lyxal

Uh just let me try and remember for a bit

emanresu A

10:37 AM

@cairdcoinheringaahing If you're awake, I really should be asleep.

pxeger

oops misread you lol

caird coinheringaahing

@emanresuA I've been awake for the past 4 hours, just haven't been on my laptop :P

pxeger

@cairdcoinheringaahing caird being awake at a normal time? Is this what university does to you‽

caird coinheringaahing

its what alarms do to you :P

lyxal

That's alarming.

@emanresuA basically, Vyxal dictionary string compression but worse

There are 6 kinds of strings: Normal strings, Normal spaceful strings, SCC only spaceless strings, SCC only spaceful strings, special spaceless strings and special spaceful strings

Normal strings allow for SCCs to be used after a ;, meaning that to compress the word Hello inside a normal string, you'd use `H%;`

(Spaceful here means that a space is appended after each SCC)

emanresu A

10:50 AM

How do the special ones work?

lyxal

In a very cursed way

Okay so there's two parts: The SCCs and the joiners

«SCCs|joiners«

It's basically " ".join[decompress(x[0]) + x[1] for x in zip(SCCs, joiners)]

emanresu A

Hm

That is a bit cursed

lyxal

«H%c¡|,! -> (H%), (c¡)!

also, the dictionary is 60k words (compared to the 20ishk that we use), and it uses the whole codepage for SCCs

@emanresuA yes

yes it is

and it totally isn't accidentally over-optimised just for the Hello World program

but then again, the whole language is basically cursed

just take a look at the "preprocessor" if you don't believe me

emanresu A

True, but without it Vyxal wouldn't exist so ¯\_(ツ)_/¯

lyxal

Also, here's my early attempt at making type cohesion: github.com/Lyxal/Keg/blob/master/Coherse.py

emanresu A

10:55 AM

@lyxal That looks fairly reaosnable, liek something I'd write

pxeger

"coherse"??

lyxal

It's basically an ugly first version of the type dictionaries we use

emanresu A

I can sorta see that

pxeger

my point is that "coherse" is definitely not a word lol

emanresu A

It's like overloading but a horrible mess

lyxal

10:56 AM

@pxeger I figured that

¯\_(ツ)_/¯

emanresu A

> #TODO: Replace everything here with something more reasonable

lyxal

You think that's bad? Object strings.

and the "parser" is probably just as cursed

emanresu A

That's spookily similar to Jyxal's parser

Wezl programs redwolf

11:39 AM

@RedwolfPrograms nonsense, it was my name first and I just programmed part of it into you :P

lyxal

@Wezlprogramsredwolf what do you program redwolf to do?

and how can I program redwolf to do my bidding as well?

because that's something I'd totally be interested in ;p

pxeger

@lyxal Fool! Redwolf can be programmed only by unanimous mental concentration among all members of The Hivemind!

lyxal

@pxeger @user can we get this changed plz?

I'd do it myself, but that sounds like something the leader needs to do

pxeger

CMQ: given . is already in use in my language, how terrible would it be to use -> for struct field access? (à la C)

lyxal

not terrible at all

I kinda like switching between . and -> when dealing with normal objects and pointers

pxeger

11:52 AM

can you think of anything better?

Adám

You probably use : already, no?

pxeger

@Adám yes, I do

lyxal

@pxeger restricted to ascii?

pxeger

@lyxal this is a practical lang which needs to be easily typeable, so yes

Adám

So you probably use all the ASCII symbols already, no?

lyxal

11:53 AM

object~attribute / object~method()

pxeger

@Adám not all of them have meanings as single characters though

lyxal

or object=>whatever maybe

Adám

Give us the list of options then.

object`attribute / object`method()?

object%attribute / object%method()?

object!attribute / object!method()?

lyxal

object,attribute / object,method()

pxeger

@Adám backticks are the one character I don't want to touch, simply because of markdown

@Adám modulo

lyxal

11:56 AM

object..attribute / object..method() ?

pxeger

@lyxal tuples

@lyxal ranges

maybe I could use ... for ranges actually

lyxal

what about ~

Adám

Right, this is unfair. You're asking a question without giving us enough info to answer.

pxeger

@lyxal I like this one the most

@Adám I'm trying to work out which single character tokens are available, wait!

lyxal

@pxeger the only thing about that is that it might give connotations of it being a destructor

although that could just be the C++ from the data structures course I'm doing talking

pxeger

11:58 AM

$, ~, and ! are the only ones left

and $ would require some careful kerfudging to make it unambiguous

lyxal

~ is the only one of those three that looks like it has a logical connection/flow between object and attribute/method

pxeger

@pxeger so really I only want suggestions for digraphs

lyxal

And so why not ->?

pxeger

@lyxal I was thinking maybe $ because it kinda relates to variables in shell and derivatives, but you're right about that tbh

@lyxal I don't like it lol, (even if it's pretty reasonable objectively), because it's a bit inconsistent with arrows' actual meaning

lyxal

just whatever you do, don't make it *object.attribute or *object.method()

also

We do a little trolling

May 6 at 0:21, 2 minutes total – 21 messages, 5 users, 0 stars

Bookmarked May 6 at 0:24 by lyxal

I knew I've seen this kind of conversation somewhere before

May 6 at 0:21, by hyper-neutrino

~>

pxeger

12:03 PM

@lyxal yeah I think C++ is the only place where ~ means destructor, but it's not really caught on as a common meaning I think so that's fine

lyxal

object~>attribute / object~>method()

squiggly arrow looks cool

pxeger

how about .$? is that too ugly?

object.$attribute

lyxal

$ feels like it should be either a prefix or postfix thing

but that could be bias from $var kind of names and Jelly

pxeger

well $ is already used for symbols/atoms, where $ is a prefix

so x.$y is a bit like shorthand for get(x, $y)

lyxal

@pxeger that kinda makes sense for attributes, but not exactly methods

pxeger

12:06 PM

@lyxal luckily there are no methods in this language

lyxal

functions?

pxeger

yes of course

lyxal

so no classes

just structs

pxeger

correct

caird coinheringaahing

@pxeger ~>

it is the only correct option

pxeger

12:08 PM

@cairdcoinheringaahing that's one more shift-key press for not much improvement IMO

lyxal

@cairdcoinheringaahing vtc as dupe of chat.stackexchange.com/transcript/message/59450325#59450325

pxeger

lol

lyxal

5 mins ago, by lyxal

object~>attribute / object~>method()

get ninja'd by 5 minutes

caird coinheringaahing

@pxeger it's squiggly, that's infinitely more improvement

lyxal

@pxeger so why not just . then? If structs are the only objects, then there shouldn't be a need to have different syntax

pxeger

12:10 PM

@lyxal because x.y <=> x.y() <=> y(x)

I really like the no-parens function call syntax

lyxal

but you just said there's no methods

oh wait

pxeger

they aren't methods, that's just syntactic sugar

lyxal

yeah I just realised that lol

so what do you think you'll choose?

pxeger

my shortlist is now ->, ~, and possibly $ (depending on if it can be sufficiently disambiguated)

lyxal

@pxeger I still don't quite understand how that works

what does $y mean?

caird coinheringaahing

12:14 PM

What's the odds on Glorfindel getting his 7th diamond? :P

lyxal

3/7

caird coinheringaahing

There are 3 positions, if you're assuming every candidate has an equal chance of being chosen :P

pxeger

@lyxal have you used Ruby?

lyxal

@pxeger no

pxeger

@lyxal not how that works I'm afraid

lyxal

12:18 PM

@pxeger then 4

(chosen by fair dice roll)

2

pxeger

@lyxal it's 1-(1-1/7)**3

@lyxal $y is like a bit like an implicitly created Enum value called y

caird coinheringaahing

We're still assuming the every candidate has an equal chance f being chosen, which seems... unlikely :P

lyxal

69

A: What does the "$" character mean in Ruby?

$ identifies a global variable, as opposed to a local variable, @instance variable, or @@class variable. Among the language-supplied global variables are $:, which is also identified by $LOAD_PATH

pxeger

the universe is deterministic, so each candidate has a chance of being chosen that is exactly 0 or 1; we just don't know which in Glorfindel's case. The average chance per candidate is 1-(1-1/7)**3, so that's what we assume Glorfindel's to be

@lyxal Ruby spells it : not $

caird coinheringaahing

Looks like CesarM got promoted, and that there's some new CMs incoming

pxeger

12:22 PM

$ is something different

and they're called Symbols in Ruby

lyxal

Language design question of my own: what should indexing something with a string do that a) isn't an error and b) is actually useful?

(e.g. [1, 2, 3, 4, 5, 6]["text"])

pxeger

why do you not want it to error? that's clearly the best option

or is this not a practical language?

lyxal

@pxeger errors are not golfy

pxeger

if you really want it to not error, it should return undefined or null or nothing or void or whatever

caird coinheringaahing

@lyxal convert to ordinals, sum, modular index

lyxal

12:26 PM

@pxeger it totally isn't the one language I work on and that is named very similarly to my username

pxeger

@cairdcoinheringaahing that's way too specific to be useful in almost all cases IMO

@lyxal in that case, overload it to be something completely different?

lyxal

but what?

pxeger

move a commonly used digraph command onto it?

lyxal

@pxeger none of the string digraphs seem like they'd be a good fit

they either have the wrong arity, don't seem consistent with their types, or definitely need to be a digraph

Aaroneous Miller

~~move a commonly used constant diagraph onto it?~~

lyxal

12:30 PM

@AaroneousMiller ~~none of the string digraphs seem like they'd be a good fit~~

~~they either have the wrong arity, don't seem consistent with their types, or definitely need to be a digraph~~

Aaroneous Miller

> ~~constant diagraph~~

lyxal

@AaroneousMiller none of the constant digraphs seem like they'd be a good fit

pxeger

@lyxal the arity can be different, can't it? If the top of stack is a string, then it can be monadic

lyxal

@pxeger I'm trying to move away from having to rely on types for arity

pxeger

that's not very golfy

lyxal

12:32 PM

as in, I'm moving towards only fixed arity

pxeger

41 secs ago, by pxeger

that's not very golfy

lyxal

@pxeger Jelly only has fixed arity elements and it's golfy.

pxeger

it needs fixed arity elements because it's not stack-based

lyxal

plus, type determined arity is a pain to transpile

"R": (
    "if len(stack) > 1 and types.FunctionType "
    "in vy_type(stack[-1], stack[-2]):\n"
    "    rhs, lhs = pop(stack, 2, ctx);"
    "    stack.append(vy_reduce(lhs, rhs, ctx))\n"
    "else:\n"
    "    stack.append(vectorise(reverse, pop(stack, 1, ctx), ctx=ctx))",
    2,
),

pxeger

hmm, true

lyxal

12:34 PM

that's just for reduce

compare that to something which can be done with fixed arity:

"T": process_element(truthy_indicies, 1)

one line, one function call vs several lines of string concatenation

actually I've had an idea

(string, non-string) / (non-string, string) just orders the arguments to make it nice

and (string, string) is just øp

Aaroneous Miller

@lyxal Why not just have it pop the maximum amount it will need and then push the extras back in the part where it actually performs functions based on type? Then you only need to check for types and stuff once.

lyxal

@AaroneousMiller because then you're risking taking implicit input

source: that's what happens in 2.4.x

Aaroneous Miller

ah

lyxal

I was never keen on that happening, hence why I want to not do that

Redwolf Programs

@pxeger We could always vaporize Glorfindel and then collapse his wavefunction

@pxeger TBH if . is already used you don't deserve struct field access

:p

Wezl programs redwolf