« first day (2702 days earlier)      last day (2133 days later) » 

as a side note, you can put this song in loop mode and do your business without having to watch the video ;)
lol look at the comments
 
12:39 AM
@LuisMendo likely not for a good while
since it took quite a bit of effort to get it set up
 
I think it was just a switch :P
the fixes had been probably made independently a long time ago
 
yes, but it took us a good amount of effort to finally get around to doing it
 
oh you mean our effort
well, we're also spending our effort in raising the privilege levels to where they should've been since a long time ago :P (the topic has been inactive for some time over meta, but I don't think the efforts are dead yet)
 
12:53 AM
@LuisMendo For some reason, when I looked at that thumbnail, I got Living on a prayer in my head
 
1:13 AM
9
A: Sandbox for Proposed Challenges

Conor O'Brien10-character golfing language language-design test-battery {{{rewriting}}} Your task is to design and implement a new golfing language. The catch: it must only have 10 significant characters as commands. Design and Implementation rules Your language should have the following characteristics:...

I've revised this question a bit, some feedback on the spec part would be nice
 
1:55 AM
hey all: if you don't have it for whatever reason, tametsi is 75% off on steam right now (it's $0.74 now)
 
@Doorknob could you clear the comments on ^^? they're outdated
 
@ConorO'Brien sure
 
thanks!
 
2:14 AM
Hello peoples
 
I was just reading about some psych study where researchers calculated the number of bits in human short-term visual memory.
Their answer is slightly over 8.
 
huh! do you have a link?
 
"An Ideal Observer Model of Visual Short-Term Memory Predicts Human Capacity–Precision Tradeoffs"
I mentioned this in the Primes and Squares room. In an every shorter summary, the subject saw N differently-colored dots briefly on a screen, then a single colored dot, and was then asked which direction that dot had moved relative to its position in the first image.
It's like lossy compression: more dots to remember, and more randomness per dot, equals less precision. Figure out what the tradeoff is, and you've got memory capacity, assuming that the person learns to perform the task optimally over time.
 
3:07 AM
@PhiNotPi i wonder if they can get a more precise range
@Doorknob :O you're back. that means ppcg podacst is back on right
 
@ASCII-only It's just a matter of getting more data.
 
 
1 hour later…
4:37 AM
Anybody able to help with Retina 0.8.2?
 
kinda
 
I wrote this to try to get a factor of a unary number. Was expecting output to be 11 or maybe 11111111; got 9. What gives? It's a replacement stage, isn't it? Where's the decimal number coming from?
 
@DLosc the last empty line means "count the number of blanks"
 
@LeakyNun Wait, there's an empty line?? [headdesk] Wow, I'm blind tonight. Thanks.
 
5:00 AM
@DLosc probably too late, but working
 
5:31 AM
oops
anyone here good with scala?
 
6:16 AM
@ASCII-only what is considered 'good'
 
like knows a decent bit about the typesystem
i'm wondering why Seq[AnyVal] isn't considered an AnyVal (i.e. can't be implicitly casted) to be specific
 
Why was this deleted?
Is Bubblegum officially No Longer Funny™?
 
@EsolangingFruit i have no idea. the meta link isn't even relevant
 
@ASCII-only Actually, looks like one of the comments on it is
I think the point is
> downvoting every answer we see in a language which is deliberately designed to try to exploit PPCG-specific loopholes
 
@ASCII-only idea: func download(auto path: Path) and lets say path has constructors (file: String) and (existingPath: Path, suffix: String) you can do download(existingPath: path, suffix: "hi") and it will implicitly act as if you did download(Path(...))
 
6:25 AM
@EsolangingFruit ppcg-specific loopholes?
@Downgoat hmm ok
 
Bubblegum is the definition of "PPCG-specific loopholes."
 
@ASCII-only idk about synatx though. Idk if should be like func download(path@ Path) or something
 
@Downgoat kinda hacky though which might not be what you want for VSL
@Downgoat hmmm. does VSL have named tuples (/anonymous structs)
 
@ASCII-only yes
 
@Downgoat imo you should just be able to pass as anonymous struct then
download({existingPath: path, suffix: "hi"}) or w/e
 
6:35 AM
@ASCII-only but then you'd have to do func download(path: (existingPath: Path, suffix: String))?
 
@Downgoat i was thinking if the struct can be passed to constructor then it's automatically constructed. but i guess {foo, bar, baz: quux} isn't a thing?
 
@ASCII-only what would that do?
 
like you can do new MyClass(arg1, arg2, name: value) right
@Downgoat ^
 
6:57 AM
@ASCII-only you can assuming the call isn't ambiguous
 
7:10 AM
I was looking at disassembly of echo and apparently it prints to stderr? Am i mistaken:
lea     rbx, asc_100000FA8 ; ": "
mov     edi, 2          ; int
mov     edx, 2          ; size_t
mov     rsi, rbx        ; void *
(followed by call to write)
 
7:24 AM
@Downgoat wait wat
@Downgoat as in bash echo?
 
@ASCII-only yeah
 
@Downgoat fwrite?
@Downgoat hang on. what is this ": " from
 
@ASCII-only just normal write
@ASCII-only it's a string? in the data sectino
 
@Downgoat osx?
 
@ASCII-only bsd echo yeah
 
7:29 AM
ah. can you TIO/pastebin disassembly
@Downgoat try echo foo 2>/dev/null maybe?
@Downgoat yeah but why would echo have that at all (more specifically, why would it have ": ")
 
does anyone know if the request/requestvalue have standardized values or are they device-specific
 
@Downgoat oh. debian?
wait wat?
@Downgoat also it should be super late for you again right
 
@ASCII-only no not really only midnight
 
1
Q: Let's design a digit mosaic

Mr. XcoderChallenge Given a positive integer \$N\$, repeat each of its digits \$d_1, d_2, d_3, \cdots, d_n\$ a number of times corresponding to its position in \$N\$. In other words, each digit \$d_k\$ should be repeated \$k\$ times (for each \$1\le k\le n\$, 1-indexed), thus creating the new number: $$\...

 
8:35 AM
have we had a challenge to make a random map from 1...n to 1...m ?
 
@ConorO'Brien I usually cast a few flags...who doesn't like to rack up their helpful flag count :P
 
0
Q: Make a random hash function

AnushThis is simple challenge to make a random function that maps from the integers 1...n to 1...m. The mapping must be uniform. That is an element of 1...n must have an equal chance of being mapped to any of the integers in 1...m. The mapping of each integer in 1...n must be independent as well. Y...

 
9:05 AM
0
A: Sandbox for Proposed Challenges

DLoscPleasinortmanteaus code-golf string natural-language A portmanteau word is a combination of two words that takes a part of each word and forms a single new word. For example, lion + tiger => liger. Let's write a program to generate portmanteaus from a pair of input words. Computers aren't the ...

 
9:55 AM
@Mr.Xcoder um...huh? "the higher digit always corresponds to the higher index"? so, test cases 65, 203 and 5202 are actually invalid?
@EriktheOutgolfer So? How does that matter? Since the higher digit always corresponds to the higher index. — Mr. Xcoder 1 min ago
 
I am unable too see your point and I also don't really have time for analysing carefully again.
 
like, the indices for 5202 are {1: 5, 2: 2, 3: 0, 4: 2}
so, the digit at index 2 is lower than the digit at index 1
 
No, that is not what I meant by that.
 
for example, the output of 5202 is
5220002222
2220002222
2220002222
0000002222
0000002222
0000002222
2222222222
2222222222
2222222222
2222222222
however, if we instead took "the maximum digit that corresponds to the column or the row", I would imagine the output to look something like this
5220002222
2222222222
2222222222
0220002222
0220002222
0220002222
2222222222
2222222222
2222222222
2222222222
 
I hope I've addressed this, but I am in a hurry so I gtg
 
10:06 AM
@Mr.Xcoder yeah, looks like you have addressed the issue now, don't worry :)
 
Ok, great, I'm done, my mom needed help with something :||
 
you can delete the comments; I've already flagged them anyway :)
 
I did so a couple of minutes ago
 
@ASCII-only same output
 
10:23 AM
@DJMcMayhem I prefer Living on a prayer :-)
 
10:52 AM
@Downgoat that looks like part of printing an error message, if it has any
 
11:08 AM
CMC: given positive integers N and R, find the pair of positive integers K and C, K×(c^R + c^(R-1) + c^(R-2) + ... + c^0) = N, that have the smallest sum
 
11:32 AM
actually R is not given, you have to find it for all R's, for example, if N is 248, K and C would be 8 and 2 (and R would be 5) for a sum of 10
 
 
2 hours later…
1:31 PM
0
A: Sandbox for Proposed Challenges

LaikoniBlack Box Trigonometry code-challengetrigonometryfunction Write a program that can distinguish the following 12 trigonometric functions: sin, cos, tan, asin, acos, atan, sinh, cosh, tanh, asinh, acosh, atanh. Your program is given one of the above functions as black box and should output one o...

 
 
2 hours later…
3:10 PM
"What is a list" meta question turns out to be more complex than I expected.
@NathanMerrill: yes, your head and tail make 1 a proper list of all square numbers. But your list implementation works only for that specific list. It's like creating a new language that solves a specific challenge in 1 (or 0) bytes. Boring, but valid. — nimi 23 hours ago
(that is: it's possible to abuse the format to make the program shorter.)
Because I expect that if it's overused (and boring) it will be made a loophole, and it's appears to be very unobservable.
 
 
3 hours later…
5:46 PM
@Cowsquack So basically solve N=K(C^(R+1)-1)/(C-1) for minimal K+C, where K,C,N,R are all positive integers?
 
5:58 PM
that is a nice way of putting it, and yes
 
this feels wrong
but it's true
 
(this came up in golfing a sed expression for creating a string that is N characters long, I've since realised that there is the trivial K=1 and C=0 and used that for my answer)
 
6:40 PM
CMC: beat me here
 
You've beaten xnor already so :o is that even possible
 
Well Ton Hospel has beaten me in perl
 
I've just been knocked out in Python... I was solving Happy numbers and was quite happy with my 85 byter, but right before submitting mmy answers, I've seen that most people have 58 :c
 
6:55 PM
An article that might be relevant for the e task
 
7:12 PM
@H.PWiz I did one challenge
then I figured out it counts characters, not bytes
nevermind
 
@H.PWiz I don't know how much you are interested in sed, but mitchs recently got <100 bytes for fizzbuzz in sed at anagol
 
@H.PWiz also the problems are woefully underspecified
"Write a program that will receive various Brainfuck programs as arguments"
as arguments? you mean argv?
> Assuming an infinitely large array, the entire Brainfuck alphabet matches the following pseudocode:
what happens -- on 0?
++ on 255?
is there a leftmost cell?
etc etc
 
@orlp Yeah, I've mostly stayed away from the challenges where that matters. It also means for cool perl 6 features that you dont see here. I can't argue for the quality of the problems either
 
@orlp that's how we have all got 58 bytes over Happy Numbers...
 
@EriktheOutgolfer I figured that out when the solutions on emirps were suspiciously short
 
7:22 PM
um, when you start typing code, it explicitly says "chars"
hm, nobody has beat me here (anagol) yet, lol (note: it's not easy)
(also, anagol counts in bytes, but is more open to test case abuse, however I don't think you'll have great success with it in this one)
 
@EriktheOutgolfer Wait what ಠ_ಠ
That's.... strange... Char count? Seriously?
 
7:40 PM
Well, looks like we (Germany) suck at football this year.
 
Anonymous
@mınxomaτ At least your team qualified :P
 
We'll be out soon by the looks of it.
 
Anonymous
My country is terrible at football because we keep trying to pick up the ball and throw it to pass
5
 
hah once you get it
 
America is great at football, we've won the superbowl every year.
11
 
7:49 PM
joke spoiler :/
also...yeah, that's true
 
Holy. Goal in the last 5 seconds.
\o/
 
My IRL German friend seems to be pretty excited right now: "Jetzt erkenne ich mein Team"
 
CMC: Find the shortest python snippet to generate the number 2136894800297704.
Btw, has anyone ever written a python integer metagolfer?
 
8:07 PM
hmm are there examples of this?
 
Anonymous
@Mr.Xcoder 15 bytes
 
Anonymous
Hex is almost always the shortest in Python, barring perfect powers and products of small integers
 
It's one byte shorter than hardcoded :)
Oh the powers-of-two approach definitely doesn't work in this case
 
Anonymous
I meant perfect powers, including perfect powers of 2 :P
 
Anonymous
The fact that hex is the largest base that has a literal syntax makes it usually the best choice
 
8:16 PM
how long is your solution @Mr.Xcoder?
 
the hex seems weird enough
0x7977e977e7ee8
 
To my own CMC? I haven't been able to come up with anything shorter than Mego's
 
@Mr.Xcoder ... I thought it's a special number
 
It is special in some way.
Its binary representation is a lookup table for the integers in [0,51) whose binary representation contains a prime number of ones.
 
...
very obscure
 
8:19 PM
Yeah but it helped me get 51 for Pernicious numbers on codegolf.io (spoiler alert!)
 
I see
 
Of course xnor got 49 and a guy called edre got 47... :/
 
I was thinking that it might be able to be written as a sum of perfect powers but I didn't get very far with that approach
 
I exhaustively tried that too but no luck here either
 
@Mr.Xcoder edre is something nobody of PPCG has managed to get to yet
like, the llhuii of code-golf.io
 
8:31 PM
How can we be sure they aren't actually active on PPCG too under a different name?
 
like, orlp? not sure, but it looks like it's a GitHub profile
@Mr.Xcoder no, just like Dennis is the Jon Skeet of PPCG (although llhuii has been now beaten by xnor :P)
 
Ah I see
@EriktheOutgolfer Well you can only login via GH... I never assumed it's orlp, and I doubt so. I've got no idea who it might be though
 
ah, confused it with something else I guess...
looks like his full name is Eric Roshan-Eisner
 
hears Dennis laughing in the background as both edre and llhuii (hope I've spelled that correctly) are just his aliases :P /s
 
or they're a secret council
of golfdom
 
8:39 PM
llhuii, I think it's Ls, not Is
or llhuii is actually edre
inb4 Dennis pops in
 
Conspiracy theory aside, they're absolutely fantastic at golfing in Python
 
9:00 PM
CMC: Given N, print an array of signs s, bases b, and exponents e for which s [i] * b [i] ^ e [i] sum to N, where the sum of exponents is maximal.
scratch that it's poorly worded
 
9:20 PM
@ConorO'Brien VTC: potentially impossible (i'm still finding papers to support my conjecture)
 
it's not impossible lol
the maximal bit is a vague
 
Pillai's conjecture is still open: it states that for every number k, there is a finite number of quadruplets (a,b,x,y) such that a^x-b^y = k
if it's false for some k, then your cmc would be unsolvable for that k
 
yeah I misread my own cmc
barring "maximal" it's trivial
 
one could just do a binary expansion if you don't need maximal
or just unary
 
or just N ^ 1
 
9:23 PM
sure
 
if we assume abs(b) < N a brute force would be possible
 
so what is your CMC?
 
no clue
 
...
 
lemme rewrite it
 
9:28 PM
working on an earthshattering wsbsite to help promote my brand...expect it when yoh least expect it
 
CMC: Given an integer N, yield an array of integers A = [...s[i], b[i], e[i]...] such that sum(s[i]*b[i]^e[i] for all i) = N. Your score is the lengths of the arrays resulting from N = 2136894800297704 and N = 32032411679864578048. Note that b[i] cannot equal N
 
do we want to make our score maximal or minimal?
both are bad, really
 
oh?
right
minimize
 
well I already said both are bad
 
eh, it sounded good in my head :P
 
9:32 PM
0
Q: FIFO cache anomalies

orlpThis is the followup challenge from this one, if you're confused please check that one out first. First, let \$m(s, k)\$ be the number of cache misses a sequence \$s\$ of resources accesses would have assuming our cache has capacity \$k\$ and uses a first-in-first-out (FIFO) ejection scheme wh...

 
forgot to say that I edited
 
@ConorO'Brien Python: lambda N:[[1,1],[N-2,2],[1,1]]
maybe some special case for n=2, doesn't matter
 
mk
well there's an idea there anyhow
 
I don't see it anyhow
I can foresee some good changes, that would lead to the solvability of a^x-b^y=k
but I think that's open
 
well the idea was the python metagolf from earlier :P
 
9:36 PM
I see
let's try to minimize the sum of abs(b[i]) and abs(e[i])
 
does that stop b[i] = [1...] and e[i] = [1...] which sum to N?
 
yes, because N would not be minimal
 
@ConorO'Brien I don't understand the challenge
A = [...s[i], b[i], e[i]...]
what is this?
 
an array of signs, bases, and exponents
 
I thought it was an array of triplets
 
9:42 PM
well yes it is
 
but leaky's answer doesn't work with that
he has an array of tuplets
 
it’s inside out maybe?
he has 2 triplets?
 
I think leaky's is transposed
 
ah
 
admittedly my notation is nonstandard and prone to confusion
A = (s[i], b[i], e[i])
 
9:44 PM
I mean
what stops me from making N (1, 1, 1) tuples?
or arbitrarily large
 
well leaky was trying to minimize abs(b[i]) + abs(e[i])
I think that meant over all tuples
 
@ConorO'Brien that's not a minimizable expression
 
9 mins ago, by Leaky Nun
let's try to minimize the sum of abs(b[i]) and abs(e[i])
 
over all i?
 
I think so
 
9:47 PM
so disregarding the length of the array altogether?
 
I suppose so, it doesn't seem to be a good parameter minimize
 
so can we simplify?
remove the sign
represent a number N as sum(b**e for b, e in A)
trying to minimize sum(b + e for b, e in A)
 
that should work
 
as an example 8^8 is more optimal than 2^24
 
correct
 
9:56 PM
\$x\$
:(
 
chat + mathjax = :(
 
s/mathjax/any sophisticated markdown/
 
@EriktheOutgolfer why?
 
maybe because it's not supported?
 
@orlp why remove the sign?
2**23 - 3**2 is not bad
 
10:08 PM
@LeakyNun complicates things
 
ok
 
10:30 PM
@ConorO'Brien [2136894800297704, 0, 0], [32032411679864578048, 0, 0] :P
 
read up :P
signs are too complicated, especially those ones
 
nowhere does it say that s[i] ≠ N :P
49 mins ago, by Conor O'Brien
an array of signs, bases, and exponents
hm, so s[i] is supposed to be in {-1, 0, 1}
 
right
since it's a sign
 
then I suppose a little bit of brute force and there it is, a theoretical solution
it's not like it can go infinite until it finds the result
 
@LeakyNun very interesting probability problem
given a random shuffle of a deck of 52 cards, what are the odds that no two adjacent cards share the same suit or face?
 
10:42 PM
very low
 
e.g. if you draw a 7 of hearts the next card can not be a 7 nor a heart
 
no idea
 
why not 1/(3*12)^51
oh right
no replacement
 
10:57 PM
Jelly, 16 bytes: 13p4Œ!=ƝẎẸƊÐḟ÷52
lol good luck processing 80658175170943878571660636856403766975289505440883277824000000000000 permutations
 

« first day (2702 days earlier)      last day (2133 days later) »