The Nineteenth Byte

Pavel

00:15

@quartata Just learned this guy is running for president while simultaneously on the run from the IRS

Neil

00:32

@JoKing I've edited my Retina answer to include a pure unary regex detector (currently 7 bytes longer than my Retina code when you take the decimal to unary conversion into account)

Deadcode

00:46

Awesome. I'll avoid analyzing it because I'd like to try that problem myself

@Neil did you do it without nested/forward references?

Wow, you did

That is an ECMAScript regex.

What funny coincidence that it's the same length as the Retina program.

Very clever algorithm. I was just going to divide by powers of two until finding one where the quotient is less than the divisor. I probably still will, just to see how it compares in length to your regex.

New Main Posts

01:12

1

Q: Partition a square grid into parts of equal area

This challenge is based on the following puzzle: You are given an n by n grid with n cells marked. Your job is to the partition the grid into n parts where each part consists of exactly n cells. Example Here is a puzzle on the left and its (unique) solution on the right: Challenge You will...

TRITICIMAGVS

Hey, It's been a while since I did much work on it but here's a new screenshot from my Brain-Flak--code-golf--rogue--like game:

Veskah

The absolute mad maan

PhiNotPi

01:27

Anybody remember AvocadOS?

El'endia Starman

...yes. I never touched the thing, but yes.

PhiNotPi

Well we should bring it back.

I was joking about it with a group of CS friends today... so there's real-world demand for an OS with such comedic value

Khuldraeseth na'Barya

01:43

CM*C: Make a case-sensitive search engine.
* M = Mega

What is AvocadOS?

DJMcMayhem

@TRITICIMAGVS Wow.

I can honestly say, 3 years ago when I thought to myself "Hey, what if I made a language where every command is made up of nested brackets", I never thought "I bet one day someone will make a video game off of this idea"

PhiNotPi

@Khuldraesethna'Barya An operating system that's entirely centered on JQuery.

It's an unfinished project.

El'endia Starman

.... definitely not touching it now! :P

PhiNotPi

It would have just enough assembly to get C working, and just enough C to get a javascript interpreter working, and then the rest would be JQuery-powered.

Khuldraeseth na'Barya

@PhiNotPi Sweet, thanks! Found it.

PhiNotPi

01:48

Pretty website: jqueryos.github.io

@El'endiaStarman you expected something else from a project with "avocad" in its name?

Downgoat

@PhiNotPi did we ever get past bootloader?

:P

El'endia Starman

@PhiNotPi I don't know what I was expecting, but it wasn't jQuery.

PhiNotPi

Pretty sure it could display character codes from keyboard input, or something like that.

Downgoat

huh

@PhiNotPi i am on board. I now have enough OS and assembly knowledge to probably get somewhere :P

PhiNotPi

I feel like we're not actually that far from the actual jQuery part. After we reach the jQuery stage, then it's a matter of adding all the "usability" features which aren't critical.

(As in, they totally are critical for any operating system to be useful, but I would be satisfied with this project if we get to the point that we do everything in jQuery)

@Downgoat I've gained a lot more C experience than when we first started this project, maybe I can help more now. Although I'm still not that good.

Deadcode

02:13

Holy crap. Using my idea came up with a 48 byte regex

so close

It's much faster though.

Downgoat

@PhiNotPi well for jQuery there's two significant aspects which is implementing the event loop, implementing DOM interface, and mounting v8. Mounting v8 would require modifying it to support a custom OS or implementing posix stuff like pthreads idk if you'd get much milage with a cross compilation

Skidsdev

@PhiNotPi What's AvocadOS? can't find anything on google

Veskah

02:31

Gotta figure out how to make a toast pun out of a daemon or something to have AvocadOS on TOAST

Bonus points for a guacamole process

Quintec

@Skidsdev They have bad SEO :P

Downgoat

@Skidsdev project some of us attempted a looonng time ago at PPCG to essentially make an OS that runs almost all on jQuery. This was back when jQuery and Avocados were everyone's favorite meme

Deadcode

@Neil submitted my regex

Skidsdev

@Downgoat that sounds like the most wonderful, horrible thing ever

PhiNotPi

02:47

Hmm... I guess I'm not all that certain what it really takes to get V8 working.

AvocadOS

Discussion for the design and creation of AvocadOS. How juic o...

IDK if there's actual interest in revamping the project. But if there is, then someone could unfreeze the room and maybe something will come of it.

quartata

03:55

@PhiNotPi well you need executable space

that is V8 needs to be able to write the JIT assembly somewhere and be allowed to jump to it

while not, you know, allowing for all the other problems that come with that

Downgoat

@ASCII-only what should syntax look like for specifying which branch of if statement is expected. e.g. for optimization

ASCII-only

06:06

@Downgoat imo i'd recommend against that, see processor branch predictors (i.e. don't do it at all, or you need a smarter predictor)

@PhiNotPi imo v8 is really specific, it'd be better if it was changed a bit so it could be relatively easily changed to any other language

@PhiNotPi yes

@PhiNotPi well. if you can do everything in jQuery (well, JS) then... that's literally 99.9% of the work

Downgoat

07:43

@ASCII-only we could technically try any other engine such as spidermonkey

H.PWiz

08:31

@Deadcode It turns out Proth was much easier than I thought. I had a brainwave yesterday night for how to solve it. Turned out to be the same as Neils. (I haven't figured out your solution exactly yet)

Deadcode

@H.PWiz Yeah, that's awesome! I see what you did there

It's kind of an implicit division, I think.

Implicit that the two odds match

H.PWiz

Yeah, I was inspired by my squares solution

Glad I solved that first

Why would (\3\3)*$) be faster than (\3\3)*)$

Is is because, then * knows that it has to go to the end of the line?

Deadcode

Because the regex engine sees the $ sooner and can fail to match if it's not at $

and can more easily do optimizations... it doesn't have to be smart enough to look outside the ()

@H.PWiz Yes, that's it exactly.

I'm playing a fast-paced game right now that's why my responses aren't that good

Okay actually it came out well enough :)

I think I've never actually used that trick that you did.

Gotta reexamine all my previous work and see if it can be used

Deadcode

11:03

@H.PWiz Could you explain why you think my Carmichael regex might get false positives?

Adám

CMC: Given a 24-hour time as an HH.MM float, return the fractional day. E.g. 18.54 (six minutes before seven in the evening) gives 0.7875.

FrownyFrog

11:32

16 bytes, J

primo

18 bytes, perl

Adám

@FrownyFrog Very clever. Translates to 12 bytes in dzaima/APL: 5÷72÷.6⊥0 1⊤

New Sandboxed Posts

11:48

0

A: Sandbox for Proposed Challenges

Irregular English Verbs Given the infinitive of an irregular English verb, output its simple past and its past participle. Rules The input and output can be given in any convenient format. No need to handle verbs not in the given list. Either a full program or a function are acceptable. If a ...

FrownyFrog

@Adám Thanks

H.PWiz

@Deadcode I was being dumb and checking not square, instead of square free. My concern is that I'm not sure that yours ensures that the number is square free

(Unless you've based your answer on some other fact)

My solution would be to assert that all prime factors only occuerd and that there were more than one of them

i.e {2} -> {2,}x$ (And \5+ -> \5* inside)

New Sandboxed Posts

12:20

0

A: Sandbox for Proposed Challenges

Fibonacci-Haiku I know that the task to write a Fibonacci-program has been given before. Therefore I have a new idea: Write a Fibonacci-Haiku. Rules You are supposed to write a program that takes one number n as input and returns the n-th fibonacci number. The sequence starts with 1. You can a...

H.PWiz

@Deadcode It gives the false positive 319345

I already am annoyed by 10k rep :(

Quintec

13:02

What?

H.PWiz

I get notified about chat flags

Adám

@H.PWiz You can start giving bounties.

H.PWiz

I suppose

Bassdrop Cumberwubwubwub

13:23

hi friends

Adám

@BassdropCumberwubwubwub Sherlock?

Mr. Xcoder

Are we going to do Best of 2019? Or shall we stop considering the delay last year and the lack of activity?

Bassdrop Cumberwubwubwub

@Adám John?

New Main Posts

13:59

0

Q: Draw a path made by direction changers

This challenge takes place on a grid. +----------+ | | | | | | | | | | | | | | | | | | +----------+ This one's 10 x 10, but it can be any rectangular shape. There are four directions on this grid. Up, down, left ...

code-golf

Skidsdev

14:50

Hmm, I have a graph curve in my head, but no idea what formula would fit it

AdmBorkBork

y=x

just make the deviation large enough

Skidsdev

don't think that quite works

I'm not sure it can actually be represented by a single formula tbh

it's y = 0 if x < 0.4, y = (x-0.4)*10 if 0.4 <= x <= 0.5, y = 1 - (x-0.5)*10 if 0.5 <= x <= 0.6 and y = 0 if x > 0.6

Skidsdev

15:15

Ok I figured it out

y = max(0, 1 - abs(0.5 - x) * 10)

Pavel

15:50

Anyone know any segfaults in Rusts perhaps due to compiler bugs

Without unsafe code

@H.PWiz Retina uses .NET and not ecmascript though?

H.PWiz

16:05

I know, the e option turns on .net ecmascript mode (which appears to be inaccurate)

Pavel

@H.PWiz docs.microsoft.com/en-us/dotnet/standard/base-types/…

DJMcMayhem

16:25

What would you name a class that calls a function asynchronously every n ms?

"Timer Class" feels a bit uninspired

Pavel

@DJMcMayhem Cron or Scheduler?

I'm a fan of Cron

DJMcMayhem

I suspect that if I named it Cron, the people who will end up reading it would have no idea what that means >_>

Very Windows-Centric

Pavel

Scheduler is good then I think

Deadcode

@H.PWiz Yeah, good catch... I left it running overnight and found the same false positive.

It's weird, I remember reading something that made me sure that "at least two prime factors of single multiplicity" would be enough for robustness.

But must be remembering wrongly.

And yes, even without running this test, from what I read last night I agreed with you.

@H.PWiz Annoyed? Why? :) Annoyed at people repeatedly congratulating you on it? Congrats on the 10k anyway!

@H.PWiz Yes I was thinking the same thing after I went to bed.

about the {2,}

the other detail I didn't remember without looking at the regex

DJMcMayhem

16:42

@Mr.Xcoder Oh hey, that's a great question

Mr. Xcoder

@DJMcMayhem How about this: I / Somebody else post(s) a meta question about best of, and if within a 2-week period we have at least, say 10 volunteers for 250 or 500-rep bounties, then we do it, otherwise we'll skip it this year? I'd say that we should try to keep the existing "traditions" in place to keep this site "alive", perhaps.

DJMcMayhem

We got 20 500 bounty offers last year. I doubt finding enough bounty-ers will be the hard part lol

Mr. Xcoder

Well then I don't see why we wouldn't try it

DJMcMayhem

Agreed. I'd be happy to post the call for nominations sometime later today

Mr. Xcoder

Then if people don't show any kind of interest in proposing stuff then we'll just leave it there and wait for next year :)

@DJMcMayhem I can do it too now if you want by modifying the last year post.

DJMcMayhem

16:52

In which case we'll have best of 2018-2019? :P

Mr. Xcoder

@DJMcMayhem You mean call for categories?

DJMcMayhem

Whoops, yeah that's what I meant

Pavel

@DJMcMayhem Is that like the PPCG fiscal year

DJMcMayhem

Yes, but our budget is $0

Pavel

I'm sure we could raise a few bucks if we really needed to

DJMcMayhem

16:57

@Mr.Xcoder Sure, go for it. Ping me when it's up, and I can pin and featured it

We're only about 3 weeks behind, that's not too bad.

My company is going to be celebrating Christmas in late February.

Pavel

@DJMcMayhem Just do chinese new year

Erik the Outgolfer

actually, we could even do it in June, 2018 posts aren't lost lol

Mr. Xcoder

Best of PPCG 2018 — Call for Categories

6

@DJMcMayhem Done.

DJMcMayhem

@Mr.Xcoder Mind if I edit that so it looks better on the starboard?

Mr. Xcoder

Go ahead

New Meta Posts

17:13

2

Q: Best of PPCG 2018 — Call for Categories

Last year we voted for the best posts of 2017 and rewarded them with bounties and challenges. I think it's a great way to reward and draw attention to some of the best content the community has created throughout the year, so let's do that again. (And hopefully, this year, we'll be able to sort e...

discussion featured best-of

Adám

@Mr.Xcoder How should I edit in if I offer multiple times 500 rep?

Mr. Xcoder

Yes, just add a new title X 500 rep bounties offered by and add your name below

Adám

@Mr.Xcoder OK, done.

Mr. Xcoder

Wow, generous rewards there :)

Erik the Outgolfer

wow that's 5k rep

Adám

17:36

If all that was awarded today, and we graduated today, I'd still only be missing 400 rep to become a trusted user.

Erik the Outgolfer

@Adám nah, you'd only be missing the site analytics... also, we've graduated, we just don't have a design (or just raised rep requirements) :P

Adám

@EriktheOutgolfer Oh, you're right. If needed, I'll double it.

Erik the Outgolfer

> If needed, I'll [just offer 5k more, for a total of an astounding 10k].

Adám

@EriktheOutgolfer I don't even know what site analytics are. What's the value of rep beyond 25k?

Erik the Outgolfer

@Adám you mean 20k ;) site analytics are next to useless

Adám

17:44

@EriktheOutgolfer I wouldn't know. No idea how to access them.

Erik the Outgolfer

@Adám review > tools > site analytics

DJMcMayhem

@EriktheOutgolfer Speaking of raised privilege levels, Robert Cartaino responded the other day:

2

A: Can we raise the privilege levels?

I wanted to update everyone on what I've been working on because I've seen many discussions, proposals, and improvised solutions about how the Community Team might allocate various "graduation" features with or without a custom design (e.g. increased reputation threshold, elections, migrations, e...

Adám

@EriktheOutgolfer TIL.

What happened in early summer '17? The site peaked then, and has been dwindling ever since.

RosLuP

18:12

Cellulare cancerogeno... Nn usare o usare il filo con auricolari per telefonare infoamica.it/la-radiazione-del-cellulare-e-un-cancerogeno-certo

Adám

@RosLuP wrong room?

Erik the Outgolfer

@Adám looks like spam

Adám

@EriktheOutgolfer I don't think it is. RosLuP is a known and contributing user.

Also, the linked material seems to be a news article rather than trying to sell something.

RosLuP

Cellphone dangerous cancerous.... Not use or use with auricular infoamica.it/la-radiazione-del-cellulare-e-un-cancerogeno-certo

Adám

@RosLuP auricular is a headset

J. Sallé

18:32

The link says "Cellular radiation is (definitely) a carcinogen", and what RosLuP linked with it reads "Cellulars are carcinogens. Do not use them, or use with a headset". Still not sure whether this is spam or not.

moonheart08

ban

LSM07

19:01

When is it inappropriate to add a new answer to an old question?

Erik the Outgolfer

never

dzaima

^

Dennis

The age of the challenge doesn't matter.

LSM07

Ok.

AdmBorkBork

CMC: code-golf Given an integer input of the area of a rectangle, output an ASCII-art of that rectangle with the perimeter minimized. You can use any printable ASCII character for the rectangle, so long as it's not whitespace and is consistent. Example: 12 should output either a 4x3 or 3x4 rectangle.

Erik the Outgolfer

19:15

@AdmBorkBork or simply given an integer n output two integers x and y where xy = n and x + y is minimized

AdmBorkBork

But then it's not ASCII art and it's a duplicate of an existing challenge.

Erik the Outgolfer

so be it, then... :P jk it's a CMC

Skidsdev

So basically find the pair of factors of n with the smallest sum

AdmBorkBork

Right

Erik the Outgolfer

um... no, actually

"pair of factors of n" is a bit ambiguous...

Skidsdev

19:26

@EriktheOutgolfer how so?

Erik the Outgolfer

let's first take the factors of n, then take some random pair of elements from that list :P

Skidsdev

would "factor pairs of n" be better wording?

Erik the Outgolfer

it must be like (f[x],f[-x-1]) in Python if f is the list of factors

and x is some index

@Skidsdev yep, that's what they're called :P

Skidsdev

@EriktheOutgolfer shouldn't've doubted myself, that's what I put initially, but then I second-guessed and changed it

AdmBorkBork

Oh, I see what you mean, Erik. Yeah, I can see how that's confusing.

DJMcMayhem

19:36

8

Q: Create a grid as close to a square as possible

Create a function or program that makes a grid as close to an square as possible You will be given an integer N as input, whole numbers (1,2,3,25, etc) The output must be a perfect rectangular grid of N letters as close to a square as possible The (wannabe)square must consist of one of the the ...

code-golf ascii-art geometry grid

AdmBorkBork

Bah

Erik the Outgolfer

lol

Adám

@AdmBorkBork May we assume the integer is larger than 1?

AdmBorkBork

Sure

Adám

@AdmBorkBork Define consistent. May we consistently use the same sequence of characters to fill with?

AdmBorkBork

19:51

No. Meaning the fill is all X or all # or something like that.

quartata

-5

Q: You have reached your question limit

What? Have a question similar to Home on the Range of Lists. Why cannot this user ask the question? Does that mean that even if the question is posted at Sandbox and well-received that the question will still not be posted at the main site?

discussion support

this persons biography is extremely normal

this person is definitely having a normal one

Adám

@AdmBorkBork APL (Dyalog Unicode), 23 bytes '.'⍴⍨c⊃⍨.5×∘≢c←⍸⊢=⍳∘.×⍳

Conor O'Brien

CMP: How would you translate ≺ and ≻ into pure ASCII?

Adám

@ConorO'Brien << and >> if multi-chars are allowed.

Conor O'Brien

multi-chars are allowed, but I think those might be reserved for the actual << and >> symbols ("much less" and "much greater")

Adám

20:01

@ConorO'Brien { and } if single chars.

quartata

@ConorO'Brien i would still use > because presumably youve made clear what kind of ordering is going on

i guess i dont know the context really

Adám

@ConorO'Brien We have no idea what the context is.

Conor O'Brien

yes this is true

let me clarify

I'm making a programming language and it's got a bunch of mathematical symbols in it, and I want pure ascii equivalents so I don't have to copy-paste these unicode symbols each time. the operator a≺b tests if a+1==b and a≻b tests if a==b+1

()[]{} already have meanings, so they cannot be used

Adám

@ConorO'Brien So they don't mean precedes and follows?

Conor O'Brien

@Adám oh they do, just in a more strict sense. "directly" precedes and "directly" succeeds

Adám

20:05

@ConorO'Brien Since there will necessarily be lots of compromises when using ASCII only, we would need an entire table.

quartata

oh

so the immediate successor

Adám

@ConorO'Brien What about non-integers?

Conor O'Brien

@Adám either default to < or error

Adám

@ConorO'Brien <| and |> maybe?

Conor O'Brien

@Adám this is also true. I don't have a full table atm. <<, <<<, >>, >>> are all taken. for this particular problem, no other operator uses < and >

@Adám not bad. I think those are pipes in F# but really who cares about that lol

quartata

20:07

-> <-

Conor O'Brien

@ConorO'Brien except for <= and >= ofc

Adám

@ConorO'Brien What do those mean?

Conor O'Brien

@Adám what they usually mean, less than or equal to and greater than or equal to

Adám

@ConorO'Brien If you're going for ASCII but otherwise resembling TMN (traditional mathematical notation), I'd make those _< and >_.

Conor O'Brien

@Adám ooo those are nice

slightly better than <- e.g. since a<-b is also parsed as a < (-b), and it may be a bit confusing given the somewhat esoteric nature of the operator

Adám

20:11

@ConorO'Brien <= => are obviously for "is implied by" and "implies that".

Conor O'Brien

oh, I forgot about those... maybe ==> and <== instead

Adám

@ConorO'Brien So will |a| mean magnitude ("absolute value")?

Conor O'Brien

@Adám just |a as a unary operator

quartata

~> <~

Adám

@ConorO'Brien OK, so you'll go the APL route. I presume !a will then be factorial.

quartata

20:13

only vaguely related to the bend of the real symbol

i have no idea if ~ is a valid unary operator

Conor O'Brien

@Adám correct

@quartata it is, what it usually means. for the bend I was thinking <: and :> or :< and >:

AdmBorkBork

Oh, definitely go <: and :> ... they're happier than the other ones.

Adám

@ConorO'Brien What will you do for assignment? :=?

quartata

:=?

half of the mustache has been shaved off

little bit at the end

Adám

@quartata no, :=

quartata

20:16

i was using it as a face

Adám

@quartata He's just smoking a pipe.

quartata

ahhhh

there it is

AdmBorkBork

Ceci n'est pas une pipe

Adám

CMC popularity-contest: Funniest APL face which has real, non-trivial meaning (and thus, it has to be syntactically correct, at least as a snippet).

Skidsdev

VTC: No objective definition of "Funniest"

AdmBorkBork

20:19

That's why it's a pop-con

Erik the Outgolfer

CMCs don't need to be sooo objective :P

just please don't start using stars as votes ;)

Skidsdev

@EriktheOutgolfer I know, I was satirically mimicking what Leaky Nun used to do with CMCs

Erik the Outgolfer

from what I've seen in the past hour or something, he might be watching...

AdmBorkBork

Leaky is always watching.... Waiting...

Erik the Outgolfer

the only way to tell is to post an advanced mathematical term and wait ;P

Adám

20:22

… and then suddenly, boom, a CMC requiring solving an unsolved mathematical problem or with computational complexity exceeding the Universe's capacity.

Skidsdev

CMC: Prove that p = np

Erik the Outgolfer

this might help...

DJMcMayhem

Crap, ninja'd

Adám

E.g ¯8○| meaning –√(–1 – | x |²)

quartata

@DJMcMayhem wait i feel stupid for not getting the last step

Skidsdev

20:28

yeah same

quartata

if P is empty that just means P is in NP

Erik the Outgolfer

it's much simpler

Skidsdev

how does it go from e^-2P = 1 to P = 0?

Erik the Outgolfer

x^0=1 as long as x is nonzero ;)

quartata

because anything to the zeroth power is 1

Skidsdev

20:30

all zeroth powers are 1 but not all 1s are zeroth powers

Erik the Outgolfer

but 1^x=1, so -2P can't be some irrational number that makes e^(-2P) result in 1

DJMcMayhem

e could be 1?

Skidsdev

√1^2 = 1?

Erik the Outgolfer

but e ≠ 1

Skidsdev

√1

quartata

20:32

i really dont know what youre talking about

Skidsdev

neither do i tbh

quartata

just take log of both sides if you want to do it out explicitly

lirtosiast

# [Wolfram Language (Mathematica)], 68 bytes

StringTrim[#<>"
"&/@Array["A"&,{x=Divisors@#~Quantile~.5,#/x}]<>""]&

[Try it online!][TIO-jrb2oyeo]

[Wolfram Language (Mathematica)]: https://www.wolfram.com/wolframscript/
[TIO-jrb2oyeo]: https://tio.run/##y00syUjNTSzJTE78n2b7P7ikKDMvPaQoMzda2cZOiUtJTd/BsagosTJayVFJTae6wtYlsyyzOL@o2EG5LrA0Ma8kMye1Ts9UR1m/ojYWqEMpVu1/moOhqYK@vkIA0KyS/wA "Wolfram Language (Mathematica) – Try It Online"

DJMcMayhem

@EriktheOutgolfer Yes, but it's already inconsistent since e^-p = -1 implies that p can't be 0

quartata

well yeah

Erik the Outgolfer

20:34

well if that was the only inconsistency...

quartata

because guess what p is

-pi i

Adám

@lirtosiast Markdown doesn't work in multi-line chat messages.

Erik the Outgolfer

@DJMcMayhem the real inconsistency there is that ² is an even power so you can't use it like that

lirtosiast

@Adám I noticed.

DJMcMayhem

The "alt-text"

Erik the Outgolfer

20:37

> It is UNHACKABLE

DJMcMayhem

@Adám I still think ⍨ is hilarious, but I have no idea what it means

Erik the Outgolfer

that is in fact syntactically correct on its own...

Adám

@DJMcMayhem Selfie (or commute)

E.g. +⍨ plus selfie means add to self. So +⍨4 is 8.

And as commute, 2-⍨5 is 3.

DJMcMayhem

How does that work?

Adám

@DJMcMayhem Which one?

DJMcMayhem

20:44

2-⍨5

Adám

@DJMcMayhem It simply swaps the arguments of the function to its left.

DJMcMayhem

So it's kinda like a jelly quick

Adám

@DJMcMayhem it is exactly the same as.

Skidsdev

so would 2-⍨8 be 6?

Adám

Jelly is an APL dialect.

@Skidsdev yes.

Pavel

20:46

@Adám I think that's a bit of a stretch

Skidsdev

inb4 dennis confirms

Pavel

It's like saying Brachylog is a dialect of Prolog

Erik the Outgolfer

it says on the tin:

> Jelly is a golfing language inspired by J.

Pavel

Not that same as being a dialect of J

Erik the Outgolfer

the source of inspiration is clearly J, and J is an offspring of APL, so one can say Jelly is a grandchild of APL in a sense :P

not directly a dialect such as Dyalog or NARS2000, though

lirtosiast

20:48

I wouldn't consider Jelly an APL dialect or a J dialect

Fixed arity alone would mean it's its own language

Pavel

We don't say English is a dialect of German even though it's a germanic language

Erik the Outgolfer

yeah, the term "dialect" isn't entirely the same as "offspring"

the latter has to do with the source of inspiration, while the former implies pure evolution and no intentional creation

AdmBorkBork

@Pavel I do. :p

Adám

@lirtosiast Like Q

lirtosiast

Is it. Maybe I should learn Q.

Adám

20:53

And Q is surely a dialect of K (which btw, has the same array model as Jelly — as opposed to APL and J), which turn is a dialect of APL.

AdmBorkBork

TIL all single-letter programming languages are dialects of APL. ;-)

Adám

As a rule of thumb, if +/ (or \+) means sum, you're probably looking at an APL dialect/derivative.

Erik the Outgolfer

@Adám note: Jelly's array model is simple flatness and vectorization is an independent property of atoms and quicks :P

Adám

@AdmBorkBork No, only A, I, J, K, and Q, afaik.

Erik the Outgolfer

@AdmBorkBork C too? :P

Adám

20:55

@EriktheOutgolfer Sure, but it was just to point out that being an APL dialect does not require multi-dimensional arrays.

RAD is also obviously an APL.

Erik the Outgolfer

@Adám if "dialect" is used in a sense that means "offspring or evolutionary separation" rather than just the latter, that is

lirtosiast

It still feels a bit strange to me to call most array-oriented languages APL dialects

AdmBorkBork

@EriktheOutgolfer Since APL was invented before C, yes.

Erik the Outgolfer

so, Python was invented after C, and it is implemented in C, but is it a dialect of C? ;-)

Adám

@lirtosiast Who does that? FORTRAN, MATLAB, and Wolfram Language are not APL dialects.

lirtosiast

20:58

MATLAB is imperative to me

and Mathematica is expression-based

AdmBorkBork

@EriktheOutgolfer No, it's a dialect of APL. Keep up. ;-)

2

lirtosiast

Do you consider Haskell a Lisp dialect, or JS a C dialect?

Erik the Outgolfer

or JS a... Java dialect?

@AdmBorkBork ancestorlessness? ;P

mbomb007

10

Making sure to fill out the developer survey accurately

Deadcode

@H.PWiz By "I haven't figured out your solution exactly yet", did you mean you haven't figured out how to implement the algorithm described outside spoiler tags in a regex, or that you looked at the regex inside the spoiler tags and haven't figured out how it works?

Adám

21:11

@mbomb007 Good point. I forgot that one.

Skidsdev

21:22

@Pavel english is also a latin language

Pavel

Not a dialect of latin so¯\_(ツ)_/¯

Skidsdev

au contraire

AdmBorkBork

@Skidsdev No, that's French.

Skidsdev

@AdmBorkBork I'm aware

Veskah

@Skidsdev I mean, it eventually used the alphabet but English is descendant from proto-germanic, Latin is father of the romance languages. They're both from the indo-european family though

Skidsdev

21:32

@Veskah huh, TIL

Veskah

upload.wikimedia.org/wikipedia/commons/4/4f/… If you want to see how it exactly shakes out

Skidsdev

only thing I know about language geneology is that finnish and hungarian are alien

Adám

21:56

@Skidsdev Let me teach you one more thing: A language is a dialect with an army and a fleet.

flawr

22:14

@Mr.Xcoder feature-request Worst of PPCG 2018 with negative bounties

Quintec

I want to give a negative bounty, give me rep

DJMcMayhem

r/UnintendedConsequence/

Quintec

Unintended? Certainly not

Adám

@Quintec I'll be happy to. Just win one of the "best of" categories.

DJMcMayhem

@Quintec It wasn't intended by flawr, I assume

flawr

22:20

@DJMcMayhem yeah, clearly not inteded by me, who would believe such a thing :P

Veskah

I' ain't winnin' nuffin'

Quintec

@flawr ~~Wow, have a negative 25k bounty~~

flawr

@Veskah what are you gonna win? D:

Veskah

@flawr Best at Everything obviously :^)

DJMcMayhem

22:37

Open question: How can we reward users other than bounties? There are a lot of categories I could think of where offering a bounty doesn't make sense as a reward, since you can only offer bounties on answers. Can we come up with other kinds of rewards that don't necessarily have to be on answers?

Adám

@DJMcMayhem Nice question in general. What truly has value?

Veskah

Janky work around is slapping bounties on unrelated answers that the user has posted

Obviously not ideal

Adám

@DJMcMayhem But yes, it is generally recognised that it is much harder to write a good challenge than a good answer.

Veskah

Hmm, You could post an Awards "Challenge" where awarded users/questions respond with links to their lauded thing. Then bounty that. Fractionally better than slapping it on random answers

DJMcMayhem

I'm not a huge fan of that idea. It just doesn't seem like something that fits on main.

In the past, we'd have honorary challenges

Veskah

22:43

I agree it's not all that great

Mr. Xcoder

@DJMcMayhem The thing is that someone has to come up with a good challenge idea and probably ask the community to help (like a collaborative challenge writing) which would be cool but who would be able to do that?

DJMcMayhem

Anyone who can write good challenges and doesn't mind volunteering some time?

Mr. Xcoder

Under these circumstances, I wish he had Calvin / Helka. :)

Downgoat

23:37

@ASCII-only hmm ??.foo can't be used because it's ambiguous w/ null coalesce, any other ideas for something that has same beahvior?

Pavel

23:57

@Downgoat For what?

Conor O'Brien

@Adám := and .=

Adám

@ConorO'Brien Don't tell me that's for global and local assignment.

Pavel

Also unless ? is valid in identifiers I don't see how that can be ambiduous with null coalesce

Conor O'Brien

@Adám .................it might not be

Adám

@ConorO'Brien Some J inspiration…

Conor O'Brien

23:59

yes

Pavel

@ConorO'Brien :> is the F# upcast operator, just fyi. (:?> is downcast)

Conor O'Brien

@Pavel oh, huh

are those anything like upgoats and downgoats

AvocadOS

Transcript for