« first day (1964 days earlier)      last day (2957 days later) » 

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
Not sure why. I voted to reopen, but then that caused a little fiasco.
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.
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"?
anyone here know how to homebridge :3
@Liam That and maybe remove the base thing.
But even if not closed, I don't think it's terribly interesting.
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 :/)
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?
That sounds better, yeah.
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
@Sp3000 As it is, it should be trivial even in some esolangs. 7 bytes in brainfuck...
Yeah, but "trivial in some esolangs" doesn't mean that it wouldn't be an interesting challenge for others :P
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.
Brainfuck. Uses little-endian binary input.
@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.
That's true about a lot of challenges, unfortunately :/
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.
Yeah, the input is definitely the bigger problem.
6:25 AM
Do you run tryitonline.net?
Yes, I do.
Does it basically just run an interpreter on the given code/input?
Is there anything else I should clarify?
@LeakyNun You mean, besides the subject of your question?
I'm assuming the subject is the logic gates question
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?
What are you intending to do?
(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)
I just removed the "u can share code" line.
> 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)
@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.
6:35 AM
Hmm, Martin got +19/-2 for "disallowing" separate multi-part challenges.
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
If you add a lot of small uninteresting things together, it just makes a big uninteresting thing, in my opinion.
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...
What do you think, should the factory be allowed?
To me it would be somewhat more interesting that way.
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.
Personally, I prefer them separate, otherwise they'll all just be n&1<<b+a*2 or similar
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?
Why would a string be considered a function? It simply isn't.
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.
@LeakyNun OK, thanks for clarifying.
@Dennis Should I allow shared code? The opinion seems to be as divided as whether UK should leave EU.
Q: Encrypt employing RailFence cipher with 5 Rails. "KNTEC RRIE CAET CTST UEOEL SBOS WLDE''

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

@LeakyNun I'd vote for no. The challenge becomes a lot clearer this way.
@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...
7:03 AM
@LeakyNun But I can't think of a way that allows interesting interaction but disallows the factory.
@Dennis Ok, would you post your currying here?
It didn't work like I expected anyway. Wrong operator precedence.
still, please post it here
For composition, I had %(f,g)=(x,y)->f(g(x,y)).
By the way, what's the consensus for answering your own challenge (except "don't")?
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 %(!,|).
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?
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
@Sp3000 For example?
@LeakyNun I don't think that's a good idea.
@Dennis You know,
@Dennis What's the consensus for answering own challenges?
7:17 AM
It's perfectly acceptable.
@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
@Dennis But is it encouraged?
@Sp3000 Alright then.
It's neither encouraged nor discouraged. If you want to, do it. If you don't, don't.
ok, thanks
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
8:02 AM
Pixar's Piper preview looks really neat:
8:13 AM
How do you guys cross out codes in codeblock?
8:27 AM
8:37 AM
@MartinEnder What is truthy and falsey in J?
I'm making a SUSE Studio appliance.
I want a better name for that appliance than Andrew's openSUSE Leap 42.1, JeOS :/
9:07 AM
@Upgoat I did, and I did. Thank you :)
9:20 AM
@LeakyNun I was asleep when you replied. Did you get your lambda working?
@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.
A: Sandbox for Proposed Challenges

NeilGolf 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...

g++ tells me error on line 52: internal compiler error: Segmentation fault
That's helpful
@Fatalize Your solution is not recursive, I think.
9:24 AM
Just figured that
Then you did understand the challenge wrong.
A: The Ever Amplifying Zigzag

Leaky NunPyth, 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 !
9:39 AM
@LeakyNun They seem to be working for me in python 3. Are you getting an error or incorrect output?
@trichoplax incorrect output. In particular, all values of a and b give me 1, no matter the value of i.
I get 1 for f(1,1), 0 in most other cases
(for i=8)
never mind
it should be my fault
I copied and pasted your code from chat
@LeakyNun Fixed
9:45 AM
@Fatalize Does { {} {{} {{}}} } and { {{{}} {}} {} } return true?
?- run_from_atom('{p:1a.}.', [[]:[[]:[[]]]], [[[[]]:[]]:[]]).
true .
it will try recursively every permutations until one works
10:43 AM
A: Sandbox for Proposed Challenges

DenkerAffePrimitive 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...

11:16 AM
I really like Doorknob's chatception userscript
@Quill What's that?
Nice! Let's see how it interacts with my colorization script
11:31 AM
colorization script?
github.com/honnza/drops > so-chat-user-colors
I can't seem to get it to work, how do you work it?
I'm using ES6, so if you're using Chrome you probably need to enable "experimental javascript" in flags
I have es6 on, I just don't know what this box of numbers is supposed to do
That's just debug output. You can minimize it by clicking it.
I'll remove it from the public version
11:43 AM
okay, maybe just like a dev branch
It's easier to test by copy/pasting into the console anyways
Can anyone take a glance at this challenge?
11:59 AM
@Quill uploaded. I'll probably add a dev branch later.
@Quill We need to make a challenge out of this.
go for it, you can have my vote already
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?
@LeakyNun Python has more operators than __mul__, __xor__ and __or__ :)
@flawr Don't forget Greek and Cyrillic.
12:03 PM
@Zgarb oh right, thanks=)
But now looking at it, it is probably gonna be damn difficult
Quick, is this latin, greek, or cyrillic? A
@JanDvorak Yes.
Note the commas
@JanDvorak My "Yes" distributes over your commas.
Then no. It's Latin.
But I get why you would accept any homoglyph as a valid answer
12:06 PM
So what do you think of this challenge?
often image recognition challenges use success rate as the score
but is it feasible at all?
Yes. If there can be an @Upgoat/Downgoat challenge...
but I suspect Latin, Korean, Chinese, Japanese, Arabic,Greek, Cyrillic might be too much
Chinese and Japanese have common glyphs.
12:08 PM
@JanDvorak There is an Upgoat/Downgoat challenge.
That's what I say
@LeakyNun latin, greek and cyrillic too
@LeakyNun OK... Chinese is valid Japanese, but cannot be declared as such by an answer
@JanDvorak Some Chinese characters do not exist in Japanese
and vice versa
Q: Upgoat or Downgoat?

UpgoatGiven 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...

12:10 PM
Thanks for finding it
@Sp3000 Well, int does not have __ge__, etc
What if we just reduce it to Latin, Greek and Cyrillic (or is there another similar alphabet?)
user image
> Mathematica has a builtin for determining goats. I don't know how to feel about that.
I guess you should ban builtins
@flawr That's really cool.
@TimmyD I'm trying to make a challenge out of that=)
12:22 PM
Oh ... that is really cool, too. I was referring to the German translation of "embiggen"
Oh sorry =)
33 mins ago, by flawr
Can anyone take a glance at this challenge?
@flawr glanced already
@JanDvorak So I assume you're speechless because it is so bad?
I have / used to have a Haskell func that would flatten an arbitrary sorted array of sorted arrays.
how about unique? oh flattening
12:28 PM
Built in
You can't just concat, sort and nub because the arrays are infinite
well here they do not have to be infinite
There are infinitely many, too, so transposing won't help
i thought of something like sort unique [x^y+y^x|x<-[2..2^32],y<-[2..2^32]]
A: Sandbox for Proposed Challenges

flawrLatin 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? ...

@NewSandboxedPosts Did you just ping me?
12:30 PM
@LeakyNun Switch to Python 3 then :P
((nub.f)[x^y+y^x|x<-[2..],y<-[y..]]!!)where f=...
Also, a few of the lambdas can still be golfed, e.g. lambda*x:0
@JanDvorak So f should extract a finite list?
Nope, it flattens an infinite list of infinite lists as long as they are sorted by head and each is sorted.
But here you have just a list of elements, not a list of lists?
12:37 PM
oh, right.
((nub.f)[[x^y+y^x|x<-[y..]]|y<-[2..]]!!)where f=...
Your dog @AlexA. ?
no, its a plane crash survivor's dog
@Optimizer You wanna say @AlexA. died on all his plane crashes?
Q: Leyland Numbers

flawrGiven 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...

@flawr birds die pretty easily on plane crashes
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>
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
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]>
but as far as handwavy, I'm not sure what you mean. The typing seems pretty unambiguous
@NathanMerrill what is T..U?
what is T[start]?
@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>>{}
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?
Do you definitely recall it being on main rather than another sandbox post?
That I don't remember.
do you guys know what a pupper is?
a tiny doggo
and a doggo is a big ol' pupper
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.
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.
@Upgoat "cheddar is an interpreted, duck-typed language." Oh, I thought it's goat-typed
... Well I guess it is kinda but still ._.
Cheddar goes well with duck or goat
3:27 PM
@trichoplax and turtles >:|
hey, to prevent/support CRTP, I'm considering the following:
class Node {
	getConnections(): List<This>
where This refers to the type of the class
@Poke mon
1 message/1 min quote broken
And c-c-combobreaker
I broke the sequence. =(
3:36 PM
@zʏᴀʙiɴ101 mon
@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?
I suspect that would be fine; Pyth is on both TIO and Herokuapp (which has extra docs and stuff) so
Yeah, CJam too I think
But I'd like to know Dennis opinion
Oh yeah, forgot about aditsu.net
The TIO one is so much better (sorry aditsu) :P
Personally I always use TIO
I prefer the dark theme :-D
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....
Minkolang is also hosted in two places, one of them TIO and the other is my own.
@quartata Nightly doesn't mean that, in fact it's the exact opposite
@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
Wait, aditsu is Romanian with a Hungarian name?
@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.
3:51 PM
@Bálint ? As far as I know aditsu lives in Hong Kong
Sándor is a hungarian name, it's the same as alexander
Well I mean Romania borders Hungary
I'm not terribly surprised
@quartata I'm hungarian, I know it :)
I used binary search to find the upper limit of OEIS sequences...
The result is oeis.org/A274176
Who you gona call? https://t.co/Fq5MmMCQi5
3:56 PM
Romanians usually have Romanic names
@Upgoat hahaha
@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? :-)
@Upgoat :P
@Bálint so Romantians have a romantic name?
Romanians have a strange origin story,

« first day (1964 days earlier)      last day (2957 days later) »