The Nineteenth Byte

Qwerp-Derp

12:02 AM

D'you guys think the challenge is ready for use?

El'endia Starman

Figures that my last shift will be a 6 am shift. >.<

jrich

@ais523 the problem is there really aren't any one character preexisting names... try running something like

for(var i in window)console.log(i)

in the console of an about:blank page

aside from the i that was just created, unless im missing something there isn't one

PhiNotPi

12:21 AM

really neat reddit comment about the fabled EM drive: np.reddit.com/r/Physics/comments/5ewj86/…

2

TLDR: the "generated thrust" is most likely due to thermal expansion

noɥʇʎPʎzɐɹC

@Downgoat the results aren't good: deepart-io.s3.amazonaws.com/cache/ad/48/…

@Downgoat Certain parts of this are really good:

And this is good:

Also,

Abstract Trump: deepart-io.s3.amazonaws.com/cache/4a/ca/…

deepart-io.s3.amazonaws.com/cache/c5/7c/… .

MS Paint Green Trump: deepart-io.s3.amazonaws.com/cache/d6/5f/…

El'endia Starman

1:03 AM

@PhiNotPi That was indeed pretty interesting.

Downgoat

.__.

why is wolf borked i do not understand

Qwerp-Derp

1:28 AM

Argh how to parse whitespace-based language

El'endia Starman

1:42 AM

@Qwerp-Derp You mean like Python or like Whitespace?

→ 2 messages moved to Sandbox

noɥʇʎPʎzɐɹC

I wrote a thing on the reason why machine learning exists (for laymen): crazypython.github.io/blog/abstract-machlearning.html "Machine learning finds patterns" How's the markdown rendering?

It's extracted from a school report I did

Editor > Export > HTML > Ctrl-C > Ctrl-V

:P

(refined men do not use "lol", they use ":P")

noɥʇʎPʎzɐɹC

2:12 AM

TIL there's a bot named "Helpdesk Queue" chat.meta.stackexchange.com/users/-55

i.redditmedia.com/…

ΛεγίωνΜάμμαλϠΟΗʹ

> /realname God
~God is 420BlazeWalker

(from a Minecraft server)

El'endia Starman

3:13 AM

@noɥʇʎPʎzɐɹC IIRC that was mostly a myth.

Downgoat

Question: what do you think is the best string representation of a given dictionary?

Should it be the same representation as the dictionary literal's representation?

Or perhaps with { key => val } style syntax

ΛεγίωνΜάμμαλϠΟΗʹ

Hans Island

Hans Island (Greenlandic: Tartupaluk; Inuktitut: ᑕᕐᑐᐸᓗᒃ; French: Île Hans; Danish: Hans Ø) is a small, uninhabited barren knoll measuring 1.3 km2 (0.5 sq mi), 1,290 metres (0.80 mi) long and 1,199 metres (0.745 mi) wide, located in the centre of the Kennedy Channel of Nares Strait—the strait that separates Ellesmere Island from northern Greenland and connects Baffin Bay with the Lincoln Sea. Hans Island is the smallest of three islands located in Kennedy Channel; the others are Franklin Island and Crozier Island. The strait is 35 kilometres (22 mi) wide here, placing the island within the territorial...

Downgoat

3:34 AM

cheddar> ['a': 'b']
{
undefined
    [object Object] => <undefined>
}

O_o I really messed something up

quartata

@Downgoat Did you get the chat relay exception from the console?

Downgoat

wait wat

quartata

Also the output should be the same as the literal syntax

Downgoat

opening console now...

I don't see anything

quartata

@Downgoat a little late now probably. I'll do it again

Ths is what I was trying to test like an hour ago

Downgoat

3:40 AM

Exception in thread message_sender:
Traceback (most recent call last):
  File "/home/tf2server/serverfiles/tf/addons/source-python/Python3/threading.py", line 914, in _bootstrap_inner
    self.run()
  File "/home/tf2server/serverfiles/tf/addons/source-python/Python3/threading.py", line 862, in run
    self._target(*self._args, **self._kwargs)
  File "/home/tf2server/serverfiles/tf/addons/source-python/Python3/chatexchange/client.py", line 185, in _worker
    self._do_action_despite_throttling(next_action)

quartata

ok thanks

Downgoat

cheddar> ['a': ['b': 'c']]
{
  a => {
    b => "c"
  }
}

yay

quartata

Restart the relay in the mean time

Also the syntax really should be the same as the literal

wat

shouldn't these things be moved to the tf2 server room?

wait @Downgoat

did you seriously write cheddar in javascript?

Downgoat

no i wrote it in Malbolge

El'endia Starman

3:44 AM

JavaScript's not so bad in comparison! :P

wat

no like

why

why not something else, like a language that actually makes sense

like C++. That's a good language.

Or if you don't like static typing

Ruby or something

Downgoat

why not an avatar that actually makes sense :P

@wat of course, ill write next programming language in awk

@wat you're looking for VSL

VSL is basically 90% Cheddar, but in C++

wat

VSL?

I don't think a probiotic compound was written in C++

Downgoat

github.com/vsl-lang/VSL

we haven't pushed to GH in ages however

ages = never

Maltysen

@flawr how does one take the complex conjugate of a matrix?

quartata

3:48 AM

@Downgoat reload the relay?

Downgoat

[
  "a": [
    "b": function CheddarString() {
          var _ref;

          var _temp, _this, _ret;

          _classCallCheck(this, CheddarString);

          for (var _len = arguments.length, args = Array(_len), _key = 0; _key < _len; _key++) {
              args[_key] = arguments[_key];
          }

          return _ret = (_temp = (_this = _possibleConstructorReturn(this, (_ref = CheddarString.__proto__ || Object.getPrototypeOf(CheddarString)).call.apply(_ref, [this].concat(args))), _this), _this.Operator = new Map([].concat(_toConsumableArray(_class2.default.Operator), _toConsumableA…

haha

why does this feature even exist

 cheddar> ['a': hai]
[
  "a": "Attempted to access undefined variable `hai`"
]

i dont even

wat

4:05 AM

Umm.

Probably because Javascript

Downgoat

definetly

Destructible Watermelon

5:56 AM

anybody at all?

betseg

No

Destructible Watermelon

oh

@betseg betseg is bot!

betseg

Negative.

Destructible Watermelon

@betseg Betseg is a liar!

betseg

This sentence is false.

Destructible Watermelon

6:03 AM

no it isn't

@Downgoat ooooh sick burn

Anonymous

@ais523 I'm very pleasantly surprised by the quality of your contributions to PPCG. You've only been a user for 16 days, but you're already contributing high-quality content (this post in particular prompted this comment). It's a pleasant surprise when a new user doesn't need much (or any) nudging to become a quality contributor.

New Main Posts

0

Q: Hexasweep: A two-part challenge

This challenge consists of two parts, as described in the title. Part 1: The solver (code-golf) Your task is to solve a Hexasweep puzzle. A Hexasweep puzzle is set out on a grid of diamonds arranged in hexagonal shapes, of which the board looks like a hexagon, like so: _____ ...

Downgoat

6:24 AM

@DestructibleWatermelon tanks :D

New Main Posts

6:52 AM

1

Q: Peel the potato

This is a potato: @@ @@@@ @@@@@@ @@@@@@ @@@@ @@ More generally, a size N potato is defined as the following shape: If N is even, it is 2 centered @ symbols, followed by 4 centered @ symbols, followed by 6 centered @ symbols, all the way up to N centered @ symbols; then, N centered @ s...

code-golf sequence animation recursion

Anonymous

What do people think of this challenge?

betseg

You wanna hack into something and you're making other people work for you? What are those challanges :p

Sp3000

I saw that and think it's perfectly fine

Oh wait

It's fastest code?

Lembik

hi all

Anonymous

@Sp3000 Yep

Sp3000

7:02 AM

Does reversing MT take that long? I don't think it did...

(Hey Lembik)

Lembik

the sandbox seems to have disappeared from codegolf.stackexchange.com

Anonymous

@Sp3000 Reversing MT19337 with 19968 bits (624 32-bit values) isn't too hard. Reversing MT19337 with 32 to 640 bits (1 to 20 32-bit values) is considerably harder.

Anonymous

@Lembik I've raised a mod flag to get it re-featured. It should be taken care of soon. Every now and then, Community decides that the Sandbox shouldn't be featured anymore, and the mods have to ~~beat up Community in the back alley~~ add the tag back.

Lembik

:)

I am writing a new permanent challenge in the sandbox

Sp3000

It's just any seed though right, can't you just assume the next (624-n) values are of a certain form? Or does that not work?

Lembik

7:07 AM

which hopefully with be interesting

I am hope this one is quite open ended

Anonymous

@Sp3000 I'm not sure what you mean

Sp3000

I want to say pad the sequence to length 624 with zeroes, but I'm not sure whether that'll work. But I'm wondering if you can do something similar

Lembik

how can I add wikimedia.org/api/rest_v1/media/math/render/svg/… to a question?

I feel I solved this before but it is driving me a little crazy

Qwerp-Derp

Guys what do you think of the new challenge?

Lembik

it is from en.wikipedia.org/wiki/…

Ashwin Gupta

7:15 AM

@Downgoat I finally did the thing you told me to a long time ago, I wrote a PID control! Well actually its a simulation of one, no real hardware involved, but I do plan on writing a real one now. Seems to be working, the error gets smaller and smaller on each iteration. Didn't do any calculus though, I don't know any. I used averaging for the I and linear rate of change for the D part.

Anonymous

@Sp3000 I doubt it. There are 19968 bits of state in MT19337. You need 19968 bits of output before you can determine the entire state. Though, since the 32-bit seed is spread all over those 19968 bits, it's theoretically possible to determine the seed (as opposed to the entire state) with less bits, but the spread and twist operations make that difficult. 640 consecutive bits will almost certainly not be enough to extract the original 32-bit seed in an efficient manner.

Lembik

challenge posted to the sandbox

as ever, comments gratefully received

Anonymous

I'll take a look at it soon

Lembik

7:30 AM

thanks!

It is missing examples currently

0

A: Sandbox for Proposed Challenges

Permanent of a rectangular matrix The challenge is to write the fastest code possible for computing the permanent of a rectangular matrix . It turns out there are a number of different ways of doing this and no one method is established to be the best...yet. The permanent of an m-by-n matrix A ...

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Permanent of a rectangular matrix The challenge is to write the fastest code possible for computing the permanent of a rectangular matrix . It turns out there are a number of different ways of doing this and no one method is established to be the best...yet. The permanent of an m-by-n matrix A ...

Lembik

ah :)

betseg

8:04 AM

15 hours ago, by New Meta Posts

REALLY, I'M RIGHT HERE...

ErikGolferエリックゴルファー

8:20 AM

Please. I know how to chat now.

codegolf.stackexchange.com/help/badges/65/…

Kritixi Lithos

I got it over a year ago

betseg

i got it 16 days after i joined the chat room

ErikGolferエリックゴルファー

Yeah but I was just introducing myself into chat when I joined this chatroom, it's different.

Qwerp-Derp

9:05 AM

Hello everyone!

betseg

Hi

Qwerp-Derp

Ooh @betseg check out my new challenge

2

Q: Hexasweep (part 1): The Solver

(This is part 1 in a two-part challenge.) Your task is to solve a Hexasweep puzzle. A Hexasweep puzzle is set out on a grid of diamonds arranged in hexagonal shapes, of which the board looks like a hexagon, like so: _____ /\ \ _____/ X\____\_____ /\ \ / XX /\ \ /X...

code-golf game decision-problem

betseg

Bot already posted

Qwerp-Derp

True but I need promotion

And because I'm impatient

Oh yeah and how do you make a Node package?

Halp

Qwerp-Derp

9:28 AM

I have 1,999 reputation :(

Why, cruel PPCG? Why must you torture me so with 1,999 reputation?

flawr

9:41 AM

@Maltysen One way is just conjugating every entry, but from a linear algebra view that (usually) doesnt make much sense: Then there is the "transpose and conjugate" analogon (for the complex case) to transposing (in the real case). But I think I never came across just conjugating.

@Maltysen yes!

New Main Posts

10:07 AM

0

Q: Dobble/SpotIt card generator

Introduction Dobble/SpotIt is a card game, where people have to spot same symbol on pair of cards in shortest time, indicate it and move to next pair. Each card has multiple symbols (8 in normal version), but exactly one is common between each pair of cards. Example from physical copy of game:...

code-golf game combinatorics set-partitions

New Main Posts

10:22 AM

0

Q: Cantor's unspeakable numbers

An unspeakable number is a number which is divisible by seven or has seven as one of its digits. A children game is to count skipping unspeakable numbers 1 2 3 4 5 6 ( ) 8 9 10 11 12 13 ( ) 15 16 ( ) 18 ... Cantor's version of the game is the sequence defined by recursively filling in the sequ...

code-golf number counting recursion

Lembik

10:44 AM

@betseg ?

flawr

11:08 AM

Fatalize

So I guess your challenge is about retroprojecting things?

Flp.Tkc

anyone know an online, raw text file, which contains most english words?

New Main Posts

0

Q: Evaluate the aspect ratio of a triangle

Given three sidelengths of a triangle, evaluate its aspect ratio AR given the following formula: where The closer to equilaterality a triangle is, the closer to 1 its aspect ratio is. The aspect ratio is bigger or equal to 1 for valid triangles. Inputs The input is three real non-negative...

code-golf math arithmetic

flawr

11:27 AM

@Fatalize yep=)

@Flp.Tkc google "word list"

github.com/dwyl/english-words

gist: 1,000 most common US English words, 2012-11-26 15:19:08Z

1-1000.txt

etc

Fatalize

There's also Jelly's dictionary, which might actually be based ton the one above

Sp3000

My go-to is usually SOWPODS since it's easy to find

Flp.Tkc

11:45 AM

hmmm, I'm gonna have a look for something a little shorter, maybe ~2000 words or so

this one seems to fit my needs

Sp3000

I was going to alternatively suggest this :P

Lynn

7

Yay Jelly!

betseg

12:02 PM

Is this a font or just a beautiful had writing

TuxCopter

Hey

New Main Posts

0

Q: Add the Fibonnaci Sequence to the Lucas Sequence

Background of Lucas Numbers The French mathematician, Edouard Lucas (1842-1891), who gave the name to the Fibonacci Numbers, found a similar series occurs often when he was investigating Fibonacci number patterns. The Fibonacci rule of adding the latest two to get the next is kept, but here we ...

code-golf math sequence fibonacci subsequence

El'endia Starman

@flawr Oooh, what is this?

flawr

@El'endiaStarman A smiley face=)

Imagine the thing on the very bottom is some data you get from a CT scanner, each line is the image of the (parallel ) x-rays passing through the face at some angle. For each angle we stretch this line in the corresponding direction (middle) and sum it up (top image).

called "filtered backprojection" of the "radon transform". It isn't actually used (as far as I know) as it is very prone to noise e.t.c but it is a good starting point for other algorithms.

Lembik

12:17 PM

the sandbox is back!

betseg

@NewMainPosts just Fibonacci sequence that starts with 2,2

Lembik

any comments on meta.codegolf.stackexchange.com/a/10711/9206 gratefully received

except for.. where are the examples! :)

Martin Ender

12:30 PM

@Lynn that's a really lovely explanation, nice work :)

2

flawr

12:59 PM

@Lynn Did you get a new touchy-swipey-computer?

@LuisMendo Why are you always so fast!

El'endia Starman

@flawr That was pretty cool! *starts part 2 eagerly*

flawr

1:17 PM

@El'endiaStarman The concept of looking at the digit changes is also reflected in the "cannonical" gray code (which is super cool!)

which also pops up when solving these sequential move puzzles

Luis Mendo

@flawr Oh, were you working on a MATL answer? Sorry :-)

4

flawr

@LuisMendo For some reason I have the impression that you aren't really sorry XD

Luis Mendo

Heheh. Just a little

Pity that in your answer -v-v doesn't save with respect to current -2*v

flawr

Right, I didn't even think of that.

That is really somehow frustrating when traditional / non-golfing languages barely get any attention...

8

El'endia Starman

@flawr I've known about gray code for a while, and occasionally, I flip [light] switches in a gray code pattern...

flawr

1:29 PM

@El'endiaStarman Are you bored at work?

XD

Luis Mendo

@flawr I think that's changing. I find it more frustrating how traditional / non-golfing languages other than Python get a lot less attention than Python

Perhaps adding an Octave online link will help get attention

betseg

People from UK, US or Canada, is there a difference between "I haven't done" and " I have never done"

Luis Mendo

(Non-native here) Perhaps emphasis? The second sounds more emphatic to me

BlueEyedBeast

1:45 PM

@betseg I would take I have never done... to mean that you don't intend to in the future

flawr

@LuisMendo I don't even think that this makes a big difference. I also wrote an answer in Haskell and in JS both of which are way more frequently used, but I suspect also those upvotes are from you, while the MATL and the Jelly answers already got a lot more votes shortly after they were posted.

betseg

I have never posted an answer with a golfing language ;_;

flawr

Of course, the nice colourful images of @Lynn's answer might have helped with it's popularity too, but that pushes all the other answers even deeper into the black hole.

(Don't get me wrong, I love both Jelly answer with the visual explanation as well as the MATL answer.)

@betseg You should ~~'nt~~ try Javagony.

TuxCopter

no

trichoplax

@betseg Context affects the meaning, and how much difference there is:

TuxCopter

1:51 PM

@flawr How do you make a condition in Javagony?

trichoplax

> I haven't done my homework
I have never done my homework

Lynn

@flawr Wacom tablet. I’ve been doing some drawing recently

betseg

@trichoplax ah thx

@flawr does it have the ternary operator?

Lynn

Maybe my images should be links… They’re not crucial to the explanation

flawr

@betseg I forgot about that, nope:)

@TuxCopter make it cause errors

and use recursion

TuxCopter

1:55 PM

Oh, so to test if a numer is 0 one can do 1/n

And check for a ZeroDivisionError

flawr

for example:)

ais523

@Mego thanks; I've been lurking PPCG for a while, and golfing since before PPCG existed, so I like to think I had a bit of a head start

TuxCopter

This seems horribly hard

flawr

@TuxCopter Hence the name:) I've actually written one or two answers.

@TuxCopter see here

TuxCopter

Are infinite loops possible without segfaults?

Because IIRC

flawr

2:00 PM

I don't think there are segfault errors, but there is always a max recursion depth.

TuxCopter

void f() {
try {
f();
} catch {
f();
}
}

segfault

trichoplax

No response to the bounty on this so the rep will go to the single old answer unless there's a new post in the next 22 hours...

betseg

@TuxCopter goto?

TuxCopter

oh nvm it still hit the recursion depth

@betseg Don't exists in Java Q_Q

trichoplax

Can you catch the recursion depth error?

betseg

2:02 PM

@TuxCopter :O

TuxCopter

@trichoplax Seems to work, visibly StackOverflowError is not a subclass of Exception

oh wait it errored

betseg

Does void f(){f();} have tail call optimization in Java?

TuxCopter

@betseg Nope

betseg

:O :O

Luis Mendo

@flawr They weren't :-)

flawr

2:06 PM

@LuisMendo lol, thanks anyway:)

Luis Mendo

Heh. Now they are

flawr

@Lynn Oooh cool. I'm waiting for you to draw your own profile pic:)

Lynn

3

I drew this one to use on Facebook! But I like the one I’m using here better, somehow. °◡°;

flawr

@Lynn feature-request use it also here on PPCG!

@Lynn I love this style:)

trichoplax

Does this sound reasonable?

Would you consider using a text hosting site that doesn't require download, and also one that doesn't use deceptive stealth adverts ("This content requires flash player Update", positioned to fool unsuspecting users into thinking it is a genuine flash message relevant to accessing the text file). — trichoplax 1 min ago

Lynn

2:15 PM

@flawr Haha, aww, thanks ♥

flawr

@Lynn Which one do you have? I always wanted to experiment with one, but I'm not really good at drawing anything, so I'm probably never gonna get one^^

@trichoplax Why don't you suggest a site? (i.e. github / github gists / pastebin)

trichoplax

@flawr I was trying to be impartial...

Lynn

A Wacom Bamboo CTL-470. I’m happy with it

flawr

@Lynn I've never actually seen one IRL, but I always have the impression that they are so small, does it still work out well?

Lynn

Yeah, it works nicely, because you pan and zoom around your drawing a lot when making digital art

flawr

2:20 PM

Oh ok, that makes sense.

And what software do you use if I may ask?

Sherlock9

@Lynn I have several questions: What answer is this for? And how do you have a dyad without arguments? What is a hook? and Are you going to do more of these nice-looking explanations?

flawr

Can you use any drawing program?

Lynn

I use myPaint when I'm on Ubuntu (free, feels intuitive, has an awesome “infinite canvas” feature, but lacks some very important other features) and Paint Tool SAI on Windows (not free, UI is a bit clunkier, no infinite canvas, but really useful features and brushes)

That little avatar was the first thing I drew in SAI. I'm still getting used to it!

Mostly myPaint doesn't have select-and-move or select-and-resize, and it’s an enormous pain in the butt.

mınxomaτ

Anyone here ever used cling (CERNs C++ interpreter) for anything more than fooling around?

Lynn

If you draw something on the wrong layer, or slightly off-center, you have to either live with it, or erase it and re-draw it in the right place. (Of course, you could argue that that’s how it is on paper, too. :D)

TuxCopter

2:25 PM

@mınxomaτ C++ interpreter?

mınxomaτ

Yeah.

github.com/vgvassilev/cling

Lynn

SAI is a lot more forgiving in that aspect, so I’ll probably use that for “big” drawings in the future. (I haven’t really done any of those at all, haha.)

I’m still in the phase where I’m ashamed of nearly everything I draw. >o<;

mınxomaτ

I stick with stick figures.

Lynn

@Sherlock9 The answer explains this (slightly) more thoroughly: codegolf.stackexchange.com/a/101238/3852

This answer so elegantly demonstrated the powers of Jelly’s chain-parsing that I had to make a nice-looking explanation for it

Sherlock9

@Lynn Thanks for the link. I'll check it after I finish golfing a Hilbert curve answer :D

And that is a very nice answer. Well done!

Lynn

2:39 PM

Thanks n_n

New Main Posts

0

Q: How to Automatically Upload Image to website daily for charity donation

There's a charity promotion on a UK website where the website will make a donation every day that an image is uploaded from my image library. And the promotion is running for a month. So I can upload the same image 30 times and the charity will get 30 x donations. Is there a way I can upload i...

code-challenge tips

Fatalize

Do we already have a challenge on Niven/Harshad numbers?

Έρικ Κωνσταντόπουλος

It's really disheartening to see answers that contain nothing more than a built-in (golfing language, Mathematica, whatever) getting much more upvotes than your own answer you spent hours on. I realized long ago that sorting answers by popularity doesn't really make sense on PPCG and started sorting them by activity instead. — Dennis ♦ Sep 22 at 3:36

^ Well, especially if you're Dennis :)

Lynn

3:23 PM

@Dennis How exactly do the words hook and fork work in Jelly terminology? They’re useful, and I want to define them in the tutorial, I think

I keep mixing them up, and they’re not quite like their J counterparts

Dennis

@Lynn The only real difference is how they behave at the beginning of the chain and how they are presented. The fork consists of 2 (3 if you count them like the J docs do) links: the far link is applied first, then the near one. A hook consists of a single (2 in J docs) link that uses the previous return value and the rightmost argument verbatim.

Lynn

3:40 PM

s/and the/as the/?

Dennis

No, the previous return value is the left argument of the hook, the rightmost argument of the chain (left for monadic, right for dyadic) the right argument of the hook.

Lynn

Oh, like so. Right

Sherlock9

I am reading this conversation same as you, Lynn, and I'm worried it's going to take me months to learn this :D It fills me with determination

Dennis

The only difference between J and Jelly is that J doesn't have a concept of return value; multiple hooks/fork build a nested train. That's only a different at the very beginning: a chain that consists only of two dyadic atoms forks In Jelly, while the counterpart in J hooks.

Lynn

Oh heck. Is this a 6-byter in Jelly? codegolf.stackexchange.com/a/101258/3852

DJMcMayhem

3:48 PM

> Determination

Hehe

Lynn

This image is so good now. I feel like hooks and forks are the two main “tricks” you have, in a J-style tacit language, and it turns out they’re both represented.

mınxomaτ

It still has weird aliasing artifacts.

Lynn

I like the pixely look, myself

Although some of the edges are weirdly jagged…

mınxomaτ

@Lynn But the image tag doesn't care about that and interpolates while resizing, which makes the image look fuzzy.

Dennis

@Lynn Yep.

mınxomaτ

3:53 PM

Either use an SVG or enable anti-aliasing for the strokes.

flawr

@Lynn Have you tried using your tablet with Inkscape?

Fatalize

I just don't get why [a,b,c ] would ever get fed to _ or the second H

makes no sense to me

Lynn

1. The power of hooks! :D 2. The power of forks! :D

TuxCopter

@Fatalize Jelly makes no sense

flawr

The power of chocolate

TuxCopter

3:56 PM

But it works

Probaly magic

Lynn

(Jelly remembers the original argument to a function, and injects it as a right argument in certain syntactical situations.)

Sherlock9

The thing that confuses me is how in the world you parse Jelly. I could probably work out the rest from there

Fatalize

yeah

Dennis

@Fatalize That'a tacit languages for you. _ needs two arguments, so it has to get the right one from somewhere. Here, somewhere is the original left argument of the chain.

flawr

@Sherlock9 you don't. Jelly parses you.

9

Fatalize

4:00 PM

like why is the second H after %@ if it gets executed first

TuxCopter

... but because there is a chain at the right argument of _ (÷@), why the input is not passed in the chain at the right of _ first?

Sherlock9

Every other language I know, you just parse as you read. With Jelly, the amount of backtracking you seem to need for parsing would give most debuggers a massive headache

Dennis

@TuxCopter Because ÷@ is a dyad, but your chain has only one argument.

TuxCopter

@Sherlock9 Why 'maybe Haskell'? Haskell seems simple to parse without backtracking

@Dennis This is geaving me headaches

Sherlock9

I may be misremembering Haskell. I'll edit it out

flawr

4:02 PM

@TuxCopter I don't think so, you always have to be very aware what type you're using.

TuxCopter

The type checker is different from the parser, right?

Lynn

It’s surprisingly “easy to read” (cough) when you get used to it

flawr

@TuxCopter Some functions do something quite different depending on the type you put into it.

@Lynn I agree:)

Sherlock9

@Dennis The question is whenever there is a dyad with only a left argument (which I think only happens when there was only one input), is the default always to grab the input again?

Ok, so say I wanted to write, for the aspect ratio question: sum, subtract, double, swapped vectorized divide original by the left, product of the division, would I write S_Ḥ÷@P?

Because that seems to be wrong

I think the fork is grabbing product of the original rather than keep the product until after the division, so I have messed up somewhere

Dennis

@Sherlock9 More or less. Take the chain + which consists of a single atom. At the beginning, the return value is set to the left argument of the chain. + hooks and applies the addition atom to the return value and the rightmost argument (left for a monadic chain, right for a dyadic one).

However, if the dyad is followed by another atom that has the same arity as the chain, that atom is called first with the argument(s) of the chain, then the dyad is called on the return value and the result of the atom we called before. This is called a fork.

Fatalize

4:11 PM

tl;dr give up

Sherlock9

DETERMINATION!

Ok, so I'd have to write S_Ḥ÷@³P for my version of the aspect ratio code to work

Dennis

@Sherlock9 Here, P is monadic, so ÷@P is a fork. One way of fixing this is to put a ¹ (identity atom) after ÷@. A better way of fixing it is to use ⁸÷ instead. With the nilad ⁸, ⁸÷ is an atop.

Sherlock9

Ohhh, I get it :D S_Ḥ⁸÷P so that you have double minus sum with S_Ḥ, divide the original left by the result with ⁸÷ and take the product P

@Lynn, is it okay if I post this Jelly answer? When you mentioned a possible 6-byte answer earlier, it seemed like you might have thought of S_Ḥ⁸÷P before I worked it all out here

Martin Ender

I've basically rewritten the spec to this from scratch. Could someone let me know if they find it clear now?

1

A: Sandbox for Proposed Challenges

Figure and Ground code-golfwhitespacesource-layout For this challenge, you need to write a program which takes no input and prints the string figure. However, there are a few restrictions on the program: The only allowed whitespace characters are spaces (code point 0x20) and newlines (carriag...