The Nineteenth Byte

Liam

6:00 AM

Why was this one closed?

"Outputting anything you want makes answers incomparable." [and therefore too broad] doesn't seem right to me

Dr Green Eggs and Iron Man

Not sure why. I voted to reopen, but then that caused a little fiasco.

Dennis

It was deemed to broad (to many possible answers). The task is trivial to begin with, and requiring only two distinct outputs (no matter which ones) and allowing input in any base (including base 2) makes it even worse.

Liam

I agree that it is extremely trivial. Allowing input in any base seems like a valid concern, but it isn't listed in the comments as the close reason. According to the comments, changing output to truthy/falsy would be enough to make it "not too broad"?

Upgoat

anyone here know how to homebridge :3

Dennis

@Liam That and maybe remove the base thing.

But even if not closed, I don't think it's terribly interesting.

Sp3000

6:09 AM

(almost tempted to go start a site specifically for golfing really simple challenges in esolangs, because of the whole trivial-in-most-languages problem we seem to be having here :/)

Liam

I don't think it is a good challenge either, but that doesn't mean it should be closed. Let's say that input was only allowed in base 10 or unary. Still too broad?

Dennis

That sounds better, yeah.

Liam

Okay, what really got me was that allowing any output, as long as it is consistent, doesn't make it too broad, at least the way I'm thinking about it

Dennis

@Sp3000 As it is, it should be trivial even in some esolangs. 7 bytes in brainfuck...

Sp3000

Yeah, but "trivial in some esolangs" doesn't mean that it wouldn't be an interesting challenge for others :P

Liam

6:14 AM

See brainfuck was what made me think of this. If brainfuck was forced to output 0/1, you're adding a lot of unnecessary bytes.

feersum

Brainfuck. Uses little-endian binary input.

Dennis

@Liam My 7 byte brainfuck program does output 0/1. Input is in binary. brainfuck.tryitonline.net/#code=K1s-LF08Lg&input=MTAwMQ

Of course, feersum's idea reduces it to 2 bytes.

With all these liberties, it's more of a rule-abusing than a programming contest.

IMHO anyway.

Sp3000

That's true about a lot of challenges, unfortunately :/

Liam

You've convinced me that the base freedom definitely ruins (the already admittedly bad) challenge.

I'm still of the opinion that the output format is perfectly fine though.

Dennis

Yeah, the input is definitely the bigger problem.

Liam

6:25 AM

Do you run tryitonline.net?

Dennis

Yes, I do.

Liam

Does it basically just run an interpreter on the given code/input?

Leaky Nun

Is there anything else I should clarify?

feersum

@LeakyNun You mean, besides the subject of your question?

Sp3000

I'm assuming the subject is the logic gates question

Dennis

6:29 AM

@LeakyNun So, interaction is completely forbidden now?

If so, I'm assuming the functions do not have to be able to coexist. Is that correct?

Leaky Nun

What are you intending to do?

Sp3000

(personally I'm still confused whether a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p=[lambda a,b:n&1<<2*a+b for n in range(16)] is allowed)

Leaky Nun

I just removed the "u can share code" line.

Sp3000

> Score is total bytes used for each function/program, including the shared codes.

:/

That still needs fixing then, I guess

(gonna go delete my comment too, I guess)

Dennis

@LeakyNun In Julia, anonymous functions are longer than redefined operators. However, not many 1-byte operators are available. If the functions do not have to coexist, I can name all of them \. If they do, I'll have to use 2-byte operators.

feersum

6:35 AM

Hmm, Martin got +19/-2 for "disallowing" separate multi-part challenges.

Sp3000

I'm kinda of two minds for this one - it's multi-part, so that's a negative, but you really wouldn't want to split this into 16 different challenges either o_O

feersum

If you add a lot of small uninteresting things together, it just makes a big uninteresting thing, in my opinion.

Sp3000

27 mins ago, by Sp3000

(almost tempted to go start a site specifically for golfing really simple challenges in esolangs, because of the whole trivial-in-most-languages problem we seem to be having here :/)

Maybe I should get cracking...

Leaky Nun

What do you think, should the factory be allowed?

feersum

To me it would be somewhat more interesting that way.

Dennis

6:38 AM

Allowing shared code (if defined properly) would make this more interesting imho. In Julia, e.g., I was going to define helper functions to compose and curry functions.

The problem is defining it.

Sp3000

Personally, I prefer them separate, otherwise they'll all just be n&1<<b+a*2 or similar

zʏᴀʙiɴ101

Hello

Upgoat

Can anyone familiar with Golfscript answer this:

So, why should an evaluable string be a valid submission in my Golfscript variant (where you set up the arguments on the stack and then eval the string) but not in Javascript (where you set up some local variables and then eval the string)? Golfscript does have a separate function type, but the only observable difference is during addition, where function addition separates tokens automatically while string addition does not. My golfscript variant removes strings and functions entirely (because functions are just strings with special addition and strings are just arrays with special printing). — Jan Dvorak yesterday

Is a GS program that's a string to be evald a valid GS program?

Is there a meta about GS functions?

feersum

Why would a string be considered a function? It simply isn't.

Upgoat

mod hammer pls

Leaky Nun

6:54 AM

@Dennis allowed

@Upgoat I would not hammer the question if I can see that the OP is not English...

@Sp3000 Could you two (you and feersum) come up with a consensus?

Although it is very unclear.

Dennis

@LeakyNun OK, thanks for clarifying.

Leaky Nun

@Dennis Should I allow shared code? The opinion seems to be as divided as whether UK should leave EU.

New Main Posts

-2

Q: Encrypt employing RailFence cipher with 5 Rails. "KNTEC RRIE CAET CTST UEOEL SBOS WLDE''

Encrypt employing RailFence cipher with 5 Rails. "KNTEC RRIE CAET CTST UEOEL SBOS WLDE"

encryption

Dennis

@LeakyNun I'd vote for no. The challenge becomes a lot clearer this way.

Leaky Nun

@Dennis Well, you said it would make it more interesting. However, this would make the factory be a valid program, which is the opposite of interesting...

@trichoplax Yes...

Dennis

7:03 AM

@LeakyNun But I can't think of a way that allows interesting interaction but disallows the factory.

Leaky Nun

@Dennis Ok, would you post your currying here?

Dennis

It didn't work like I expected anyway. Wrong operator precedence.

Leaky Nun

still, please post it here

Dennis

For composition, I had %(f,g)=(x,y)->f(g(x,y)).

Leaky Nun

By the way, what's the consensus for answering your own challenge (except "don't")?

Dennis

7:08 AM

As it turns out, it isn't terribly useful. I wanted to write !%| for logical NOT of bitwise OR, but I actually need %(!,|).

Leaky Nun

Also, is it good if I add "for lambdas, you can omit the arguments in your byte-count"?

Because someone seriously hates lambdas right now.

@Dennis Can't you redefine the operator?

Sp3000

I know the intention is to remove boilerplate, but I feel like adding that as a rule would just open a new can of worms

Leaky Nun

@Sp3000 For example?

Dennis

@LeakyNun I don't think that's a good idea.

Leaky Nun

@Dennis You know,

a=>b=>0
a=>b=>a&b
a=>b=>a>b
a=>b=>a
a=>b=>a<b
a=>b=>b
a=>b=>a^b
a=>b=>a|b
a=>b=>~a&~b
a=>b=>a==b
a=>b=>~b
a=>b=>a>=b
a=>b=>~a
a=>b=>a<=b
a=>b=>~a|~b
a=>b=>1

0
a&b
a>b
a
a<b
b
a^b
a|b
~a&~b
a==b
~b
a>=b
~a
a<=b
~a|~b
1

@Dennis What's the consensus for answering own challenges?

Dennis

7:17 AM

It's perfectly acceptable.

Sp3000

@LeakyNun e.g. mixing boiler plates (int.__xor__ and lambda x,y:x&1-y in the same solution) or boiler plates that differ slightly between functions

Leaky Nun

@Dennis But is it encouraged?

@Sp3000 Alright then.

Dennis

It's neither encouraged nor discouraged. If you want to, do it. If you don't, don't.

Leaky Nun

ok, thanks

Jan Dvorak

7:40 AM

I would say that if {"a":"b"} has the correct behavior for some challenge, it's a bad challenge. — Jan Dvorak 16 secs ago

mınxomaτ

8:02 AM

Pixar's Piper preview looks really neat:

Piper – First Look – Official Disney Pixar | HD

►

Leaky Nun

8:13 AM

How do you guys cross out codes in codeblock?

Quill

8:27 AM

@Upgoat new blog post! 418stat.us/2016/06/computed-property-keys

Leaky Nun

8:37 AM

@MartinEnder What is truthy and falsey in J?

zʏᴀʙiɴ101

I'm making a SUSE Studio appliance.

I want a better name for that appliance than Andrew's openSUSE Leap 42.1, JeOS :/

trichoplax

9:07 AM

@Upgoat I did, and I did. Thank you :)

trichoplax

9:20 AM

@LeakyNun I was asleep when you replied. Did you get your lambda working?

Leaky Nun

@trichoplax Nope.

[lambda a,b:(i>>a*2+b)&1for i in range(16)] does not work, but lambda a,b:(8>>a*2+b)&1 works.

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Golf all the 16 logic gates with 2 inputs and 1 output! This question asked for 16 independent functions. I would like the opposite: a single function that takes an additional parameter that specifies which of the 16 logic gates is required using an integer from 0 to 15. If you don't want to use...

feersum

g++ tells me error on line 52: internal compiler error: Segmentation fault

That's helpful

Fatalize

Did I understand this challenge wrong?

Leaky Nun

@Fatalize Your solution is not recursive, I think.

Fatalize

9:24 AM

yes

Just figured that

Leaky Nun

Then you did understand the challenge wrong.

2

A: The Ever Amplifying Zigzag

Pyth, 60 53 52 46 42 39 bytes It is now on par with the bug-fixed version of Jelly! [email protected]++*]*dl`hkabhSK`hk*]*dl`hkabeSKKd [email protected]*]*dl`hkhaeSKhSKabhSKhkKd J1K.u+N=J_WsI@Y2JtQZ=-RhSKKjsM.t.eX*]*dl`hkheSKbhkKd J1K.u+N=J_WsI@Y2JtQQj-#dsMC.eX*]*dl`hkheSK...

I almost out-golfed Dennis!

I out-golfed @Dennis !

trichoplax

9:39 AM

@LeakyNun They seem to be working for me in python 3. Are you getting an error or incorrect output?

Leaky Nun

@trichoplax incorrect output. In particular, all values of a and b give me 1, no matter the value of i.

trichoplax

I get 1 for f(1,1), 0 in most other cases

(for i=8)

Leaky Nun

never mind

it should be my fault

trichoplax

I copied and pasted your code from chat

Fatalize

@LeakyNun Fixed

Leaky Nun

9:45 AM

@Fatalize Does { {} {{} {{}}} } and { {{{}} {}} {} } return true?

Fatalize

?- run_from_atom('{p:1a.}.', [[]:[[]:[[]]]], [[[[]]:[]]:[]]).
true .

Leaky Nun

Nice.

Fatalize

it will try recursively every permutations until one works

New Sandboxed Posts

10:43 AM

0

A: Sandbox for Proposed Challenges

Primitive Pythagorean triples Introduction A Pythagorean triple is a tuple of three positive integers a, b and c so that a² + b² = c². One example of that is (3, 4, 5). One subset of those are primitive Pythagorean triples which require a, b and c to also be coprimes, so their only common divis...

Quill

11:16 AM

I really like Doorknob's chatception userscript

Jan Dvorak

?

@Quill What's that?

Quill

github.com/KeyboardFire/stackexchange-chatception

Jan Dvorak

Nice! Let's see how it interacts with my colorization script

Quill

11:31 AM

colorization script?

Jan Dvorak

github.com/honnza/drops > so-chat-user-colors

Quill

I can't seem to get it to work, how do you work it?

Jan Dvorak

I'm using ES6, so if you're using Chrome you probably need to enable "experimental javascript" in flags

Quill

I have es6 on, I just don't know what this box of numbers is supposed to do

Jan Dvorak

That's just debug output. You can minimize it by clicking it.

I'll remove it from the public version

Quill

11:43 AM

okay, maybe just like a dev branch

Jan Dvorak

It's easier to test by copy/pasting into the console anyways

flawr

Can anyone take a glance at this challenge?

Jan Dvorak

11:59 AM

@Quill uploaded. I'll probably add a dev branch later.

flawr

@Quill We need to make a challenge out of this.

Quill

go for it, you can have my vote already

flawr

I just thought about a challenge where you are presented with a raster image of some text, and your program should tell whether it is written in the Latin alphabet, Koren, Chinese, Japanese, Arabic

Did you have something similar in mind?

Sp3000

@LeakyNun Python has more operators than __mul__, __xor__ and __or__ :)

Zgarb

@flawr Don't forget Greek and Cyrillic.

flawr

12:03 PM

@Zgarb oh right, thanks=)

But now looking at it, it is probably gonna be damn difficult

Jan Dvorak

Quick, is this latin, greek, or cyrillic? A

Zgarb

@JanDvorak Yes.

2

Jan Dvorak

Note the commas

Zgarb

@JanDvorak My "Yes" distributes over your commas.

Jan Dvorak

Then no. It's Latin.

But I get why you would accept any homoglyph as a valid answer

flawr

12:06 PM

So what do you think of this challenge?

Jan Dvorak

often image recognition challenges use success rate as the score

flawr

but is it feasible at all?

Jan Dvorak

Yes. If there can be an @Upgoat/Downgoat challenge...

flawr

but I suspect Latin, Korean, Chinese, Japanese, Arabic,Greek, Cyrillic might be too much

Leaky Nun

Chinese and Japanese have common glyphs.

zʏᴀʙiɴ101

12:08 PM

@JanDvorak There is an Upgoat/Downgoat challenge.

Jan Dvorak

That's what I say

flawr

@LeakyNun latin, greek and cyrillic too

Jan Dvorak

@LeakyNun OK... Chinese is valid Japanese, but cannot be declared as such by an answer

Leaky Nun

@JanDvorak Some Chinese characters do not exist in Japanese

and vice versa

zʏᴀʙiɴ101

188

Q: Upgoat or Downgoat?

Given an image of a goat, your program should best try to identify whether the goat is upside down, or not. Examples These are examples of what the input may be. Not actual inputs Input: Output: Downgoat Spec Your program should be at most 30,000 bytes The input will contain the full goa...

code-challenge image-processing classification

Jan Dvorak

12:10 PM

Thanks for finding it

Leaky Nun

@Sp3000 Well, int does not have __ge__, etc

flawr

What if we just reduce it to Latin, Greek and Cyrillic (or is there another similar alphabet?)

3

Jan Dvorak

> Mathematica has a builtin for determining goats. I don't know how to feel about that.

I guess you should ban builtins

TimmyD

@flawr That's really cool.

flawr

@TimmyD I'm trying to make a challenge out of that=)

TimmyD

12:22 PM

Oh ... that is really cool, too. I was referring to the German translation of "embiggen"

flawr

Oh sorry =)

33 mins ago, by flawr

Can anyone take a glance at this challenge?

Jan Dvorak

@flawr glanced already

flawr

@JanDvorak So I assume you're speechless because it is so bad?

=)

Jan Dvorak

I have / used to have a Haskell func that would flatten an arbitrary sorted array of sorted arrays.

flawr

~~how about unique?~~ oh flattening

Jan Dvorak

12:28 PM

Built in

You can't just concat, sort and nub because the arrays are infinite

flawr

well here they do not have to be infinite

Jan Dvorak

There are infinitely many, too, so transposing won't help

flawr

i thought of something like sort unique [x^y+y^x|x<-[2..2^32],y<-[2..2^32]]

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Latin vs Greek vs Cyrillic / Battle of the Alphabets code-challengeimage-processingclassificationpattern-recognition Meta: This is just a rough draft, the challenge is still under development, feel free to add suggestions as comments or via chat pinging @flawr fixed typeface for all three? ...

flawr

@NewSandboxedPosts Did you just ping me?

Sp3000

12:30 PM

@LeakyNun Switch to Python 3 then :P

Jan Dvorak

((nub.f)[x^y+y^x|x<-[2..],y<-[y..]]!!)where f=...

Sp3000

Also, a few of the lambdas can still be golfed, e.g. lambda*x:0

flawr

@JanDvorak So f should extract a finite list?

Jan Dvorak

Nope, it flattens an infinite list of infinite lists as long as they are sorted by head and each is sorted.

flawr

But here you have just a list of elements, not a list of lists?

Jan Dvorak

12:37 PM

oh, right.

((nub.f)[[x^y+y^x|x<-[y..]]|y<-[2..]]!!)where f=...

flawr

Your dog @AlexA. ?

Optimizer

no, its a plane crash survivor's dog

flawr

@Optimizer You wanna say @AlexA. died on all his plane crashes?

New Main Posts

2

Q: Leyland Numbers

Given a natural number n, return the n-th Leyland number. Leyland Number Leyland numbers are positive integers k of the form k = x^y + y^x Where x,y are integers strictly greater than 1. They are enumerated in ascending order. Details You may use 0 or 1 based indexing, whatever suits yo...

code-golf math number sequence arithmetic

Optimizer

@flawr birds die pretty easily on plane crashes

Nathan Merrill

1:29 PM

trying to come up with syntax for a type-safe tuple slice function

what do you think:

slice(start: static Int, stop: static Int, tuple: Tuple<T..U[start]..V[stop]..W>): Tuple<U..V>

orlp

1:46 PM

@NathanMerrill what is static in this context?

do you mean compile-time expression?

and regarding Tuple<T..U[start]..V[stop]..W>, Tuple<U..V>, that's way too handwavy and I do not see how it could generate a consistent language

How to play the Imperial March on a toad in youtube

►

3

Nathan Merrill

2:02 PM

yes, static means compile-time

I came up with a better one on the drive:

slice(tuple: Tuple<T..U>, start: static Int = 0, stop: static Int = tuple.Length): Tuple<T[start]..T[stop]>

Eᴀsᴛᴇʀʟʏ Iʀᴋ

@orlp 10/10

Nathan Merrill

but as far as handwavy, I'm not sure what you mean. The typing seems pretty unambiguous

orlp

@NathanMerrill what is T..U?

what is T[start]?

etc...

Nathan Merrill

@orlp T defines the first type in the tuple (as declared in the original tuple syntax)

and U defines the last type in the tuple

T[start] simply increments the pointer

unzip(tuples: List<Tuple<T..U>>): Tuple<List<T>..List<U>>{}

TimmyD

2:33 PM

I had thought we had a "section / paragraph / subparagraph numbering" challenge around, which would be very similar to this, but now I can't seem to find it ... — TimmyD 2 hours ago

Does anyone recall such a challenge?

trichoplax

Do you definitely recall it being on main rather than another sandbox post?

TimmyD

That I don't remember.

orlp

do you guys know what a pupper is?

a tiny doggo

and a doggo is a big ol' pupper

TimmyD

2:57 PM

@trichoplax Well, I manually searched the first 25 pages of the Sandbox (sorted by Active), which goes back to Dec 22, 2015, for the words paragraph, heading, or section and don't see that challenge. I also can't find it on main with those keywords, either. Maybe I'm just dreaming something.

trichoplax

I do vaguely remember something about nested somethings, and there have been indentation challenges. I can't recall anything specifically about paragraphs but I don't see every challenge.

Bálint

@Upgoat "cheddar is an interpreted, duck-typed language." Oh, I thought it's goat-typed

Upgoat

... Well I guess it is kinda but still ._.

trichoplax

Cheddar goes well with duck or goat

5

Upgoat

3:27 PM

@trichoplax and turtles >:|

Nathan Merrill

hey, to prevent/support CRTP, I'm considering the following:

class Node {
	getConnections(): List<This>
}

where This refers to the type of the class

Eᴀsᴛᴇʀʟʏ Iʀᴋ

hallo

zʏᴀʙiɴ101

hello

Poke

hullo

zʏᴀʙiɴ101

@Poke mon

1 message/1 min quote broken

And c-c-combobreaker

I broke the sequence. =(

Poke

3:36 PM

@zʏᴀʙiɴ101 mon

Luis Mendo

@Dennis User Suever has started working on a sort of MATL online server, including some MATL-specific stuff like searchable help and answer statistics. Since you so kindly host MATL in TIO, I'd like to know your opinion about having a second MATL server in parallel. Would that be ok for you?

quartata

I suspect that would be fine; Pyth is on both TIO and Herokuapp (which has extra docs and stuff) so

Luis Mendo

Yeah, CJam too I think

But I'd like to know Dennis opinion

quartata

Oh yeah, forgot about aditsu.net

The TIO one is so much better (sorry aditsu) :P

Luis Mendo

Personally I always use TIO

I prefer the dark theme :-D

quartata

3:46 PM

Same here but mostly because it's a nightly build which means things like ew actually work

Although I don't seem to be getting that error anymore on aditsu.net, weird....

El'endia Starman

Minkolang is also hosted in two places, one of them TIO and the other is my own.

Bálint

@quartata Nightly doesn't mean that, in fact it's the exact opposite

quartata

@Bálint That would be the case normally

except that aditsu seems not to want to make a new release even though the mercurial build is completely stable

Bálint

Wait, aditsu is Romanian with a Hungarian name?

Dennis

@LuisMendo Sure, there's no reason why TIO should be the only online interpreter. A bunch of languages have other interpreters besides TIO. I haven't implemented a lot of customization so far, so if you can provide a better experience with a tailored interpreter, go for it.

quartata

3:51 PM

@Bálint ? As far as I know aditsu lives in Hong Kong

Bálint

@quartata aditsu.net/About_Adrian

Sándor is a hungarian name, it's the same as alexander

quartata

Well I mean Romania borders Hungary

I'm not terribly surprised

Bálint

@quartata I'm hungarian, I know it :)

Leaky Nun

I used binary search to find the upper limit of OEIS sequences...

The result is oeis.org/A274176

Upgoat

Who you gona call? https://t.co/Fq5MmMCQi5

Tweeted by EverythingGoats on May 29, 2016 at 7:23 PM

2

Bálint

3:56 PM

Romanians usually have Romanic names

El'endia Starman

@Upgoat hahaha

Luis Mendo

@Dennis Thanks, Dennis. And again, I can't say how grateful I am for your hosting it in TIO. You mentioned you might have to do a trip to Madrid, how's that going? :-)

trichoplax

@Upgoat :P

Optimizer

@Bálint so Romantians have a romantic name?

Bálint

Romanians have a strange origin story,

Transcript for