« first day (1841 days earlier)      last day (2996 days later) » 

5:18 AM
llama@llama:...code/ruby/markov$ ./markov.rb
error-checking program is necessary to be the audio file name for arbitrary subset sum: you reproduce that my weekend, now you'll see wrong since that's not enough to use both elements in narrow circumstances; as the string. output will run on Jan's code output section of the fact I wonder if you can be a Stack Exchange answers here is already did nobody uses a list of the crash. Adding another competition in the 4 chars, trigraphs) only be stated explicitly disallow the result of posting Coffeescript demos is awesome, hopefully it as if the inte
I've taken some inspiration from @Geobits and thrown together a quick Markov chain Ruby script using Comments.xml from the PPCG data dump
 
@Doorknob Oooh, fancy. GitHub it?
 
Nice :)
 
#!/usr/bin/env ruby

data = Hash.new([])

File.open('../datadump/Comments.xml', ?r) do |f|
    f.each_line do |line|
        next if rand(10) < 9
        idx = line.index('Text="')
        next unless idx
        text = line[idx+6...line.index(?", idx+6)]
        text.split.each_cons(2) {|a, b| data[a] += [b] }
    end
end

words = [data.keys.sample]
loop {
    continuation = data[words[-1]].sample
    break unless continuation
    words += [continuation]
}
puts words * ' '
Ruby is awesome ;)
 
in Beep Boop Maggot, 37 secs ago, by Marky Markov
@Zgarb I'm not sure what the _UNK is.
Nobody is
 
> Lastly, you Mr. T U V W X may be disinclined, due to be clearly not bytes. Omit ! before posting this to have added to be reopened now? I could probably obviate the previous comment, REM+TOO... I win....
 
5:21 AM
On a scale of one to even, I can't. — VoteToClose 5 secs ago
 
> before posting this to have added to be reopened now?
Remarkably similar to some comments I've seen.
 
> Ported your program contains the OP should be avoided using Perl 6 99 in this all of unavoidable baggage, and specify that users fail the longest code as ~-n.
Haha I got an output that was literally just "So... Winning conditions?"
 
@Geobits Have you given Marky access to comment on main??
 
:P Well technically he has enough rep to comment anywhere. Should I write a script that comments on random posts at irregular intervals?
 
ಠ_ಠ
 
5:25 AM
> In Haskell If they must be a minifier to answer to combine to think it remains of course you've shifted the beauty of the question to PPCG!
> Didn't make me code, at byte count if unary -split will be outputted.
 
in Beep Boop Maggot, 24 secs ago, by Marky Markov
@Hosch250 I think I'll go to bed
 
> but @poseidon wants to satisfy the same There's no Using mathematica but you clarify that could do you can just fabulous. lmao
 
> very cool, I want to convert the spoon interpreter doesn't change Sorry, popularity-contest type into Prolog functional code cleanliness advice...
> in different but Python is encountered? an answer. was invalid. anyway. but also leave you load of insanity.
 
YES
WE NOW GRADUATE KTHNXBYE
 
5:34 AM
@Doorknob where is the image from
 
straight from okay to excellent
 
@somebody A51
 
1842
Code Golf & Programming Puzzlescodegolf.stackexchange.com

Beta Q&A site for code golfers and for those who interested in code golfing (from beginners to experts), and programming puzzles.

Currently in public beta.

 
Should we meta this?
 
_UNK
 
5:37 AM
Um, no?
What's there to say?
 
Like, CG meta in celebration. \o/
 
@AlexA. yey 10.1 is an big nummbur
 
Was that you or your markov script?
2
 
>.>
 
I didnot know we can cont that highh
 
5:38 AM
9.9 answers per question?!
 
Yeah
 
I'm surprised it's only 9.9
 
^
 
I think that's a bug
it's been at 9.9 for forever
 
I'll bet that's the highest answers/question ratio across the network.
 
5:39 AM
you think it's higher?
 
^ Yes.
 
@Doorknob I made a little Julia program that connects to the SE API to get the average q/d over the last two weeks with up-to-the-minute results! :O
 
Do they have stats listed for full sites?
 
VIVA LE HYPE
@Liam No, because they don't matter once the site has graduated.
 
It's 9.6 total all-time, so I'm sure the recent one should be higher.
 
5:40 AM
@AlexA. you can do that with search you know
 
This is easier
 
Trivia: Ethereum is the beta site with the highest q/day stat (14.3).
Followed by Arduino (14.1).
 
because the subject matter is super confusing
 
> Ethereum, the crypto value and blockchain-based consensus network
wtf
 
It just got into beta
 
5:41 AM
Then CiviCRM (13.1).
 
Sounds like a bunch of made up words
 
@Doorknob Looks like neither of them vote much.
 
I think you should restrict it to sites older than X days or something.
 
@AlexA. cryptocurrencies (e.g. bitcoin/litecoin)
 
5:42 AM
Then Law (11.6).
@Geobits I'm just going off of this :P
Oh, and then finally us (10.1).
That makes us 6th.
 
@Doorknob Some guy who's active on that site chewed me out on A51 for suggesting that a proposal for a Tea.SE be given a chance rather than being closed as a duplicate of Seasoned Advice immediately.
 
Pretty decent, for a site where we originally thought 10/day was downright impossible...
@AlexA. wat
 
¯\_(ツ)_/¯
 
Yes, because that would be a lot different than coffee.se or beer.se >_>
 
public beta start

5 years ago
 
5:44 AM
Is downgoat in here?
 
@Geobits I'm user #19 on Coffee.SE!
 
@Downgoat you here?
 
And 887 on Beer
 
@SleepingPerson Hey, are you awake?
 
@Downgoat well anyway you will see this eventually: I assume you have played the video game goat simulator? (store.steampowered.com/app/265930)
seems like it would be up your ally.
@VoteToClose have you seen:
?
 
5:48 AM
@AshwinGupta wat
 
@Doorknob Finally a math problem he gets right. :P
 
As long as an expression evaluates to 0, he'll get it right
Or 0. 0. 0. 0
 
in Beep Boop Maggot, 20 secs ago, by Marky Markov
@VoteToClose I'm not sure what the hell is.
 
Nailed it.
 
5:53 AM
I leave him alone with you guys for two minutes and you're having the hell talk?
 
XD
Btw, this is the just the craziest thing:
in Beep Boop Maggot, 2 hours ago, by Marky Markov
@CᴏɴᴏʀO'Bʀɪᴇɴ I've been trying to figure out how to make a challenge that works in the sandbox.
how
 
in Beep Boop Maggot, 14 secs ago, by Marky Markov
@VoteToClose ninja'd
 
He said that a few times earlier
It's triggered by using 'd in a message
 
@AlexA. It's not actually.
I checked.
It seems that "rickroll'd" in particular does that.
 
5:57 AM
It's his response ever time you've said "rickroll'd"
Oh, that in particular
Interesting
 
in Beep Boop Maggot, 37 secs ago, by Hosch250
Good night.
in Beep Boop Maggot, 41 secs ago, by Marky Markov
@Hosch250 Good night !
 
@VoteToClose I noticed you posted a donald trump video earlier.
I figured you'd find this one funny.
 
Oh, no, I'm not specifically after Trump.
I'm after all politicians.
 
in Beep Boop Maggot, 30 secs ago, by Marky Markov
@Doorknob I think I'm not going to post it
 
in Beep Boop Maggot, 19 secs ago, by Marky Markov
@Geobits I have no idea what I'm doing.
 
6:23 AM
0
Q: Residue Number System

LiamThis is code-golf In the vein of large number challenges I thought this one might be interesting. In this challenge, we will be using the Residue Number System (RNS) to perform addition, subtraction, and multiplication on large integers. What is the RNS The RNS is one of many ways that peop...

 
 
1 hour later…
7:26 AM
0
A: Sandbox for Proposed Challenges

alphadeltaI'm a Puzzle! Your task is to print the string "I'm a OS", where OS is the operating system that the program is run on. If run in a browser (JavaScript and/or any other browser scripting languages), the browser should be output instead. The code should be portable. Standard loopholes apply. Yo...

 
7:43 AM
Would it be frowned up to intentionally keep a challenge you write off of HNQ by including a phantom unicode character or something?
 
1
A: Sandbox for Proposed Challenges

xnorKleene star code-golf Given a set of symbols, their Kleene star consists of all strings of any length composed of those symbols. For example, {'a','b'}* = {'', 'a', 'b', 'aa', 'ab', 'ba', 'bb', 'aaa', 'aab', ...} Input: A non-empty collection of distinct characters a..z. Output: The Kleene ...

 
@xnor Why would you do that?
 
to avoid some of the voting patterns that come when easy challenges go onto HNQ
it's a matter of personal taste i suppose
i complained about HNQ here
 
so.. . we didn't graduate yet?
in Beep Boop Maggot, 3 hours ago, by VoteToClose
Can we refresh the dictionary now? :D He's had some prime material.
in Beep Boop Maggot, 3 hours ago, by Marky Markov
@VoteToClose I'm not sure if I'm not sure how to do that.
 
8:01 AM
ayyyyyyyyyy 10.1 questions/day ¯\_(ツ)_/¯ಠ_ಠ( ͡° ͜ʖ ͡°)( ͡ಠ ツ ͡ಠ)¯\_( ͡ಠ ͜ʖ ͡ಠ)_/¯
 
and we still aren't graduating
it was all a myth
graduation is a myth
 
> fraduating
from Area 51 site:
> Beta Q&A site for code golfers and for those who interested in code golfing (from beginners to experts), and programming puzzles.
(emphasis mine)
for that is how you English. ಠ_ಠ I have found why we aren't graduating. Grammar error in description
 
I blame YOU
 
We've already graduated. After design-independent graduation, they introduced an even better model: graduation-independent graduation. Nothing happens, so no one has to do any work. Great, isn't it?
 
8:28 AM
Guys, why do video games make for such strange question titles?
(thinking of gaming.se)
I just asked a question there with the title, "How do I know if I should flush?" And some people might take that the wrong way....
 
8:48 AM
@Doorknob got more numbers and an animation for you
 
9:05 AM
@El'endiaStarman I stopped computation for n = 66
it's still going, it's probably way too much
 
9:24 AM
then someone needs to find a pattern?
also, I still don't know how to make the pattern of 0s programmatically
 
9:44 AM
I found what Queens were looking for
that @somebody to love
 
9:55 AM
@MartinBüttner I haven't thought about edge case "characters" but transforming a matching part to e.g. spaces is something that comes up. Now it is doable with $1$* but with complex regexes that can be long. Anyway, T`p` `regex is enough most of the time, so not a big deal.
 
@orlp I think I got the formula for the even n:
In which b(n) is the smallest possible m greater than zero, with:
So, for n = 66, the smallest m is 33. Which gives 2^(33+1) - 2 = 17179869182
But this works for all even n, except n = 36...
 
So... you haven't found the formula :P
n=36 is so mysterious
In the meantime, I've got this:
def c(n):
 f,t,p,l=lambda w:[0]+[w[i]^w[i+2]for i in range(n)]+[0],[0,1]+[0]*n,1,1;h=f(t)
 while t!=h:
  if p==l:t,l,p=h,0,p*2
  h=f(h);l+=1
 return l
 
@orlp Yeah, I can't figure out why that is happening.
 
10:13 AM
What are you working on?
 
@randomra it seems to have quite strong patterns
but then suddenly n = 36 goes poof
 
Hahaha
 
and the real question is
why 36?!
what's magical about it
 
a(2) = 10
a(4) = 110
a(6) = 1110
a(8) = 1110
a(10) = 111110
a(12) = 1111110
a(14) = 11110
a(16) = 11110
a(18) = 1111111110
a(20) = 1111110
a(22) = 111111111110
a(24) = 11111111110
a(26) = 1111111110
a(28) = 111111111111110
a(30) = 111110
a(32) = 111110
a(34) = 1111111111110
a(36) = 101010101010101010
a(38) = 1111111111110
a(40) = 11111111110
a(42) = 11111110
a(44) = 1111111111110
 
spot the odd one out
 
10:21 AM
@orlp have you rerun 36?
 
multiple times
@Adnan have you looked at my animation?
 
@orlp It looks fascinating!
 
make sure to press + / -
 
It looks a bit like a Sierpinski triangle
 
@Adnan it IS a sierpinksi triangle
 
10:32 AM
open-cobol needs and oddly specific number of bytes of archives.
 
@orlp But cut in half it seems
 
@Adnan that's because we put the initial 1 on the left border
 
hmm
 
@orlp Oh, I see
 
10:40 AM
@VoteToClose Also exactly 5353kB of space, where two of the packages are version 5.3
 
^
1337 and 5353
 
@orlp you were stuck on 66 right?
nvm, found it
 
Hey, how would I correctly use chroot?
As in, I'm running a specific process and I don't want it to have access to higher directories.
 
10:57 AM
Hmm.
Maybe not entirely what I wanted, then.
I'm trying to restrict a single process to a single directory, with no other installations.
 
@Adnan have you checked how your wire differs from oeis.org/A160657? that has the formula you have given.
 
haven't read them properly, so idk if that's what you want
 
Hmm.
I'm looking at jailkit.
I mean, I want minimal access to anything.
 
anything?
 
No shelling, no file io whatsoever.
I'm trying to find a solution to allow us to use Databot with non-safe interpreters.
 
11:05 AM
oh
non-safe interpreters?
eval() it? apparently pypy has build in sandboxing for some stuff
 
Interpreters that can access the shell and files.
@somebody It's not all in one language, though.
 
how do you run other languages?
 
Some of them are linux binaries, some are JS, some are python, some are C(++|#)?
 
but how?
 
I'm talking programming language interpreters.
The one I'm currently working on is COBOL.
 
11:08 AM
yeah, how does databot run them
just execute them from system terminal?
 
It executes them from python Popen.
 
maybe a good idea to get PyV8 etc. instead?
 
What would that do for us?
 
PyV8 is pretty much JS for python
 
And what would that do for us?
 
11:15 AM
idk
how do you run js using popen?
 
With spidermonkey/node.
What I'm attempting to do is generalize the access restrictions so we can do it for any programming language with low disk usage.
 
@VoteToClose go to lounge<c++> chat room on stackoverflow and ask @StackedCrooked
 
Link?
 

 Lounge<C++>

Today we're daydreaming about C++26 reflection
 
Wait, never mind.
Thanks. :D
 
11:19 AM
btw guys, this answer deserves upvotes:
1
A: Compress an image to a 4 KiB preview

MoogieJava (vanilla, should work with java 1.5+), score 0.672 It does not generate particularly good dssim scores but, to my eye it produces more human friendly images... Album: http://imgur.com/a/yL31U Scoring script output: Comparing corpus/1 - starry.png to test/1 - starry.png... 2.3521...

 
@orlp Why StackedCrooked?
 
@VoteToClose he made coliru.stacked-crooked.com
 
D: why does data not have se-chatbot as a submodule?
 
@somebody Because we modified it significantly.
We majorly reworked chat access.
 
oh
killall python3 <- not always a good idea
 
11:24 AM
@somebody Yeah, we know. ;P We were having problems with threads at the beginning that weren't allowing us to kill it normally, but that's fixed now.
 
:( you need a !todo command, or open an issue with an interactive todo list
 
Eh.
 
12:09 PM
0
Q: {Junk decryption} a puzzle I found on a wargame site

adibI have found this puzzle and I wanted to share it with you to see if anyone can solve it because I can't figure it out Agent, we've just received a bunch of files from one of our field agents taken from one of senior gang members hard drives, and our team is having trouble putting it all togethe...

 
 
1 hour later…
1:19 PM
@orlp here are some of the next ones, i'm calculating the next 9 (incl. 6, leaving if overnight to see if anything happens
67 120
68 8388606
69 16380
71 112
72 1022
73 349524
74 2097150
75 4088
77 16380
79 96
81 4092
83 504
84 510
85 508
87 496
88 4094
89 16380
90 8190
91 16376
92 2046
95 64
101 1020
103 1008
107 4088
111 224
113 2044
119 240
123 248
125 252
126 254
127 1
 
@randomra if you actually need more than printable ASCII, you can still use a custom range.
 
@MartinBüttner Is there some concise way to transpose input in Retina? I mean like matrix transposition
 
unfortunately not
I thought about that a few times.
 
@MartinBüttner Sure
 
I don't want to add just a "transpose" built-in.
if I can come up with a useful generalisation that makes it possible in ~20 bytes, say, I'd be happy with that.
 
1:27 PM
yeah, that would be out of place, but it would often help a lot
in the current state it would be the shortest with balancing groups?
 
yes
although with non-rectangular input even that is pretty hard
 
if you could substitute with the whole match stack that would might help
 
hm yeah. that's planned, but I haven't decided yet how best to implement that
@Geobits I moved back to Germany in summer (and I'm actually living in southern Germany now).
 
my first idea was $SX1 where S marks the stack printing option and X is the delimiter
 
1:43 PM
hm, maybe
I think first I want to add something like # that takes the match length instead of the capture count
so you don't have to do (1)+ -> $#1 but can do 1+ -> $"0 (or using some nicer character probably)
 
2:05 PM
I didn't understood why did you choose the current version instead of match length.
 
2:21 PM
it seemed more flexible
 
2:42 PM
Yay, you did the connect the dots in Retina, now I don't have this urge to do it. :)
 
3:03 PM
@MartinBüttner Hey, can you delete stuff in other rooms?
I really, really need you to delete something really quick. c:
@Dennis @Doorknob Please?
 
what?
 
Can you delete the last message in here?
 
'swhat flags are for
 
Someone here has Windows 8.1 Pro (64bit)?
 
0
Q: Fill the holes in letters

Hannes KarppilaSometimes when I'm bored, I take some text and fill the "holes" in the letters. But isn't filling the holes the most boring thing you can do? I think we should automate it, so we could use our time better. Standard code-golf rules apply. Input A string containing sequence of alphanumeric chara...

 
3:10 PM
@mınxomaτ Yeah, I have one somewhere, whatcha need?
 
@VoteToClose Is is native? I need access to the GPU.
 
@mınxomaτ Uh, is what native?
 
@VoteToClose Not in a virtual machine.
 
@mınxomaτ Oh, yeah, it is.
 
OK, what's your screen resolution?
 
3:12 PM
Idk, it's my dad's computer. I'd have to ask him for perms to get on it.
 
@SuperJedi224 _UNK
 
Nice going @Downgoat.... :P
 
Wat hepnd
in Beep Boop Maggot, 14 secs ago, by Marky Markov
@VoteToClose done
Mark now knows the difference between happy and sad expressions.
in Beep Boop Maggot, 14 secs ago, by Marky Markov
@VoteToClose I'm not a mod.
 
3:32 PM
1
Q: ADFGX Cipher Encryption

George GibsonADFGX cipher encryption The ADFGX cipher was used by the German army in WWI. Write a program to encipher with it! It works as follows: Create a 5x5 Polybius square with a mixed alphabet. The polybias square is labled A D F G X along the top and left. Generate the pairs, so A in its non-mixed ...

 
3:46 PM
@Downgoat BTW, I'm still porpoising challenges.
Hey, can I do this challenge again, but as a ?
20
Q: Self-Interpreting Interpreter

Hoa Long TamBased on a comment by George Edison to this question, write the smallest self-interpreting interpreter. You may use the language of your choosing. Empty languages do not count. Your program must be at least two characters long. The program does not need to interpret the entire language, just a...

 
@VoteToClose what were you thinking of doing with it?
 
I feel like a lot of langs didn't compete because it was code golf and therefore biased.
@GamrCorps As a challenge.
 
@VoteToClose how would you score it then? But I like the idea of reviving it with some "modern" golfing langs
 
Plus, it should be the entire language, not just a snippet.
@GamrCorps It'd be nigh upon impossible to actually score - it'd be a language-per-language victor, I wouldn't accept an answer. More like a catalog question.
 
@VoteToClose Ah, I see. Then couldn't you just make it a catalog code-golf? (If necessary, write it very well and close the old on as a dupe?)
 
3:55 PM
I don't like the idea of it being code golf.
Just a code challenge.
 
@VoteToClose I would say go for it.
 
Time to Porpoise this Challenge!
 
LOL. Still can't believe Downgoat merged that.
 
@VoteToClose But what would be the scoring algorithm?
 
@VoteToClose You need objective scoring
ninja'ed -.-
 
3:58 PM
worst case scenario:
 
But how on earth would you score that? It'd be completely unfair on a language-per-language basis?
I would make it a popularity contest if I had to.
 
Code challenges still need a scoring algorithm / objective winning criterion.
 
@GamrCorps we use popularity contest to use human judgement for accuracy for certain criteria
 
we still need criteria the humans are able to judge
 
3:59 PM
@NathanMerrill Yes that is true.
 
Aight. It'll be a pop-con then.
 
^
 
xD
What should I do then?
 
what criteria are humans supposed to judge?
 
4:01 PM
I mean, I don't know how to score it.
 
you can have a list of features, and give points for each feature?
but the problem is that not all languages support all features
 
Hmm. But I'm talking about an entire language interpreter anyways.
I'd allow eval in this one, but suggest that people make it not eval.
I kinda wanna make it a catalog.
 
ok. Here's a question. If I make an interpreter, but wanted a better score, what could I do?
 
I don't know.
That's up to the writer.
 
then PPCG might not be the best platform for this
 
4:05 PM
^
@MartinBüttner can I use your series leaderboard from the ASCII series for some of my upcoming challenges?
 
Should I make it a code golf catalog, then?
 
@GamrCorps the ASCII series is Optimizer's (mine is Random Golf), but of course, go ahead
 
@MartinBüttner Oops, thats what I meant, but thanks!
 
@VoteToClose is that really different from the existing challenge other than allowing eval then?
 
@MartinBüttner Hmm. Damn.
That challenge is 5 years old, and we've done similar things with "99 bottles".
 
4:08 PM
there was no code golf for 99 bottles
we only had popcons
 
Oh.
 
for HW, we did have code golfs, but all of them were in one way or another
only for cat did we already have a challenge, but that had a really bad spec, so it made sense to redo it.
(I'm not saying it's impossible to improve that challenge, but the questions you should ask is are a) how can it be significantly improved? b) can those improvements be edited into the old challenge without invalidating things?)
 
@MartinBüttner Also, the old challenge did not require the full language to be interpreted.
 
If you think we can do much better than the old challenge and that it has some serious flaws that need correcting, then you can propose a new spec in the sandbox, and if peopel agree the old can be closed as a duplicate. But I'm pretty sure they will be dupes one way or another.
@VoteToClose maybe that's a good thing? :P
 
@MartinBüttner It's a different challenge.
Plus, with my allowance of eval, it's quite different.
 
4:12 PM
For anything except esolangs (and using eval, which seems boring), it becomes impossible once you require the entire language to be self-interpreted.
(by impossible, I mean practically impossible, not theoretically)
 
It is a different challenge, no?
 
is it a better one? :)
 
I'd say yes.
But I'm biased. :P
 
what kind of amazing answers do you expect that couldn't be posted to the existing one?
 
Ones that don't just do tiny bits of languages.
 
4:20 PM
I just think the programs will be come so ridiculously massive that no one will want to golf them (if they even fit in an answers)
 
Hmm.
For compiled languages, should I allow them to be compiled by the interpreter, but run by the normal runner? (i.e. Java)
 
@ThomasKwa That's one of the defaults for multiple pieces of input, so yes.
 
@VoteToClose are you going to ask for an interpreter or a compiler?
 
@MartinBüttner An interpreter. :\ Nevermind then.
 
@El'endiaStarman stop gaining rep you're only delaying the inevitable
Wow, this is cool:
in Beep Boop Maggot, 12 hours ago, by trichoplax
@AlexA. The one thing where the correct answer is zero and I get nothing...
in Beep Boop Maggot, 12 hours ago, by Marky Markov
@trichoplax ( that's the same thing )
 

« first day (1841 days earlier)      last day (2996 days later) »