« first day (3543 days earlier)      last day (1596 days later) » 
00:00 - 15:0015:00 - 00:00

15:12
Also Jelly Balls is pretty nice for understanding shorter programs
It gives explanations for any program that doesn't have an explanation
wouldn't recommend trying to develop in it much
I've never really liked Jelly Balls tbh
Not quite sure why, just never really thought it's useful/helpful. But if others do, ¯\_(ツ)_/¯
Is there an easy way to type all the non-ASCII characters it uses?
Adam has a toolbar browser extension, but beyond that not really
Sep 26 '17 at 20:26, by caird coinheringaahing
ಠ_ಠ I hate what Jelly has made me do. I just copy pasted the character W
15:17
I'll make a mac keyboard layout for husk if I'm not feeling lazy
Maybe I'll just memorize all the codepoints and Ctrl+Shift+U them...seems easier :p
gasps
So Jelly has already been language of the month, right?
It's too popular for LotM
Ah
Even if I did have an excuse for spamming Jelly answers, there's probably one from Jonathan Allan on every single question
15:19
There are at least 10 users with >100 Jelly answers
or you
or Dennis idk
Yeah, it's difficult to find questions without existing Jelly answers
So the best way to do it is answer unanswered question
easy peasy
Not the best language to be using for long-time unanswered questions :P
Or politely ask everyone from JHT to stay away from a newly asked question
15:22
What's the site policy on challenges that don't require code? e.g. "Write a mathematical function that does <insert thing here>" with an objective scoring criteria
If you do , you can trick people into doing that, but on its own you can't
There was a meta post a few months ago about that exact question IIRC
We've had code free challenges before ((-a)*(-a) = a*a)
Yeah, but I think that was atomic code golf. It's still sort of code.
28
A: Is "Proof-Golf for proofs in general" on-topic?

Nathan MerrillYes, as long as it is well-specified For proof-golf, we validate the prove by mentally following it. For compass and straightedge, this means that a reader should be able to come out with the same solution given the same set of steps. Furthermore, each of the possible steps should have a well-sp...

@Razetime Does Husk have a builtin for "convert to base 10"?
15:28
I believe this is the meta post I was thinking of: codegolf.meta.stackexchange.com/questions/10287/…
@cairdcoinheringaahing B should be the one
B16 converts from base 16 to base 10
given a list of numbers
So what's B10 doing in this program?
Wait, I meant "convert from base 10"
That's d
Ah, I thought it existed :P
Was very surprised that Husk didn't have a "convert from base 10" builtin
d converts from and to base 10
15:33
Yay, I love it when both of my recent Jelly answers get downvoted with no comment ಠ_ಠ
some people just hate golfing langs
I just got used to random downvotes at this point
I'm used to them, but they're still irritating
I stand by the principle of "If you can't stand behind your reason for voting enough to explain why, maybe you shouldn't've voted that way"
hard to tell that to anonymous voters innit
Yeah, but in my experience most drive by voters will stick to their vote no matter what you say
Also annoying is when a trivial answer in a BF derivative gets dozens or even hundreds of votes because it "looks compilcated" or "has a funny name" or something
e.g., the dozens of "upvoting because it's ArnoldC" comments
15:39
At least they're willing to say why they voted
ah thatkinda annoying
but some challenges are enuinely hard for bf derivatives
so I guess thats fairer
JS is weird. There's an argument you can pass to this function that makes it return true: f = (n) => (typeof n == "object" && n > 8 && n++ < 4 && typeof n == "number")
(If anyone can get the answer in under an hour, 50 rep bounty on any answer they choose)
15:59
In two hours I should be able to answer it in an hour.
Maybe I'll do another one later then, I've got multiple
summons Shaggy and Arnauld
Javascript has plenty of weirdness to go around
@RedwolfPrograms spoiler
Close, only works if a is defined though.
I think that counts though
16:07
@RedwolfPrograms rather that it isn't defined, and is run in non-strict mode
Interesting unintended solution. I'll award the bounty at the end of the hour.
*start
@RedwolfPrograms no need for a bounty, i don't use SE
(this works, but only once. i'm guessing it's possible to get a reusable solution with a proper object)
My intended solution was much more complicated, mostly just because I wanted to avoid global variables
16:11
this works and is repeatable
Ah, that's a good idea. Much simpler than what I did.
obvious golfs (now that i looked at the intended solution)
(i couldn't remember how valueOf was called, but toString works anyways because JS ¯\_(ツ)_/¯)
Not sure why inequality operators would ever call string related methods tbh
@RedwolfPrograms iirc it's that valueOf falls back to toString
Interesting, o = {}, !(o < "[") && o < "[" is true because of that
16:18
@RedwolfPrograms that's false
Oh, the second [ should be a ]
Sorry about that
0
A: Sandbox for Proposed Challenges

caird coinheringaahingHow many times do I have to call you? Define a function \$f:\mathbb{N}\to\mathbb{N}\$ such that \$f^\alpha(n) = n^\alpha\$, where \$f^\alpha\$ denotes composition with itself \$\alpha\$ times. For example, \$f^1(n) = f(n), f^2(n) = f(f(n)), f^3(n) = f(f(f(n)))\$ etc. Therefore, the function you d...

Jeez NSP's getting quick
is there more to golf from that? i've got this
Seems optimal.
I wish there was a way to get rid of those parentheses around the object literal, but the only way I can find just adds a byte
16:52
heh, I made a tongue-in-cheek answer on an HNQ, and that Q made the SO interesting questions blog, so I've got 28 upvotes without trying
0
Q: We're all unique in our own way

caird coinheringaahingThis is an answer-chaining challenge where every part of your answer should aim to be unique from every other answer. This question will work in the following way: I will post the first answer. The next answer will stem from that, and all other answers will originate from it. In this answer I wi...

 
1 hour later…
18:18
@cairdcoinheringaahing Do you know any reason review stalker would be redirecting to a different site? It's giving me worldbuilding review tasks.
@RedwolfPrograms If you've visited a sites review queues it adds those in
Is there a way to get rid of a site from it?
Log out, visit the queues for the site you want to remove, then go back to PPCG and log back in
I just deleted and reinstalled it, seems faster :p
 
4 hours later…
22:13
Morning everyone!
Did you see the list of bytes I chose?
I did
Is there any relevance to them or were they random?
Did you find it funny?
Oh, I just saw that :/
What;s the 15?
22:15
Number 15 burger king foot lettuce
...ok
It was a clever way of referencing that meme :P
I think the only languages that can work with those bytes are going to be ones where the characters don't mean anything
E and * are the only printable characters when ASCII encoded
The bytes are D)”Ǧ®¤Œ in Jelly
@cairdcoinheringaahing Make that E*¡Ð©µ¥Þ because Jelly uses 1-indexing and I always forget that
22:46
It's the latter. lol
I'm trying Jelly too because it's very tolerant of random meaningless things being spammed in without breaking anything
Pretty sure its impossible in Jelly
unfortunately both 0 and 1 are taken, as those are relatively easy to output since 0 is the default and we can force 1 with *
¡Ð©µ¥Þ are all quicks and E only outputs 0 and 1, meaning * can only output 0 and 1
i found a way to get it to output a list, but that's not helpful
since we have no way to turn that into a number other than E which again, only can give 0 and 1.
22:50
and the copy quick is literally useless here
wait, can we find a way to output with exit code...
© doesn't do anything, Þ, µ and ¥ cannot change the values in a list and ¡ and С can't yield anything other than changes to 0 and 1 with the two atoms
@HyperNeutrino Nope, Jelly's exit codes are either 0 or 1
oh. bruh, well there goes that
all errors get sent to code 1? that's odd but unfortunate
05AB1E might work
@HyperNeutrino It's a Python thing IIRC
@cairdcoinheringaahing oh, does seem to be. huh, never actually noticed that
The 05AB1E characters are E*ǝтεθγΣ
not sure what any of them mean, though
Me neither :/
E*ǝтεθγΣ errors by itself though
22:57
these don't seem like they'd work.
E requires input
ε is each and requires a closing }
this definitely isn't correct
Might see if any of my languages work
Lol I think I've got a language where it can work
But I'm not telling you what it is
did you have something in mind or did you just do it cuz 69 420 funny numbers lol
Yes
Funny numbers
nice :+1:
23:06
Hehe
Ugh, I keep thinking you two are the same person
Lol
small icons be like that
theoretically someone could just give the only available byte as 0 and wouldn't that just basically end the challenge?
Unary
23:08
oh. lol
(if that's what the other one is called)
So long as no one uses them before someone posts the "the only byte you can use is 0" answer
mhm
oh, the time limit is 14 days this time. cool
But there'd exist some other language
Probably
I actually don't know
23:10
quick, time to start making a bunch of arbitrary languages that just do arbitrary things given any random bytes so you can just randomly obtain some number :P
But that's been explicitly banned
> You may use any language that wasn't specifically invented to answer this challenge
I'm 14 parallel universes ahead of you HN :P
(was mostly joking) also is there an official policy on language postdating answer-chaining? because I know we removed [non-competing] and just allowed new languages on old challenges as long as it wasn't like MetaGolfScript
And I had something to do with that lol
@HyperNeutrino ... - _- you win this time
@HyperNeutrino Using a language made specifically after the challenge has posted, in order to do that specific that is a loophole isn't it?
23:12
Uh. sorry, that wasn't clear. I don't mean a language specifically designed for a challenge, but like in general, if a language is made after a challenge, should it be allowed in answer chaining
IMO yes, so long as it wasn't specifically aimed to answer it
Not sure about site policy
example: OEIS answer chaining, a lot of problems came from not having more Python versions to work with, so if I made a language (or a bunch of languages) that transpiled into python in some way, how would that work, or even if someone else made languages without seeing the challenge and I just took advantage of the newly existing languages
I think a single rule might not be enough to cover everything and really it's mostly common sense as to what is in good faith of a challenge and what's just exploiting loopholes and taking the fun away
The problem of course being that "Use some common sense, of course that's exploiting a loophole, don't be an arse" isn't how site policy usually goes :P
Perhaps "only languages that existed before this challenge are allowed"
And then, you'd have to say "before this was sandboxed" to avoid challenge sniping
yeah, definitely :P I think the rules here make sense, mostly. like it's pretty easy to tell if a language was made specifically to just have random byte sets pass without error, or if it's a legit language that just happens to be new
23:16
TBH I'd be fine with the site rules being "Here is a list of things we've decided aren't funny. When posting an answer, ask yourself, is this on the list? If there's any chance the answer isn't no, don't post it"
None of my languages work btw :/
Lol
The problem is that E isn't a common command, and * is almost always multiply, which usually works on the 0s on the stack :/
Well everyone has 14 days to find a language
* can almost always be assumed to be multiplication or exponentiation, neither of which help with 0 and 1 being the only two available things
I think custom code pages is the way to go
23:19
Definitely
If only there were a language which didn't treat letters as commands. Shoulda made that while it was still sandboxed lol.
(the language that is)
You need a permissive language with either a custom code page, or where E has some meaning
I've got a language where it's easy to solve
Husk's commands are E*¤‰←¦↓≥ BTW
But it ain't my turn so I can't use it
Tell us in 13 days then :P
23:23
Okey doke
I'll drop a hint in 7 days if it isn't found by then
> A-L are numbers
That sounds promising
Huh, that's not what I was thinking about but sure
00:00 - 15:0015:00 - 00:00

« first day (3543 days earlier)      last day (1596 days later) »