« first day (1847 days earlier)      last day (3000 days later) » 

9:21 AM
@Lembik @Zgarb I'm looking for an equivalent of PCA on binary data. Basically: given a set of binary vectors, find the smallest set of binary vectors, which can yield the input set by XORing subsets. That must be a well studied problem, but I'm not entirely sure what to search for and I figured one of you might know.
 
@everyone how many people use python interpreters? should Shtriped be ported to Ruby(/.NET(/Java/Swift/Perl)) as well?
also, does anyone an ad with a phrase ('Like coding? Join Code Golf.SE') in unicode is representative of what we do here/tempting to click?
 
9:38 AM
@MartinBüttner Thanks for asking .. sadly I don't know but it sounds np-hard to me (unlike PCA). I would just ask on cs.stackexchange
 
9:49 AM
Hm, thanks. I might try that.
 
please paste the url if you do
I would be very interested
@MartinBüttner did you get an answer to your math.se question?
 
Migrated it to CS since. I got an answer but after giving it some thought it didn't seem to answer the question. There's currently a bounty on it.
 
thanks, that's interesting
 
10:21 AM
@Lembik found a paper on the Discrete Basis Problem (which is what I'm looking for, but uses OR instead of XOR). That one is indeed NP-Hard.
Although that has the number of vectors k as an input as well and tries to minimise the error.
 
10:36 AM
interesting
 
11:12 AM
0
Q: Minimal basis for binary vectors using XOR

Martin BüttnerI would be surprised if this isn't a well-studied problem, but I'm not sure what else to search for at this point: you're given a set of binary $n$-vectors $S \subset \{0,1\}^n$. The problem is to find another set of binary $n$-vectors $B \subset \{0,1\}^n$, with minimal size $|B|$, such that eve...

 
thanks!
 
Is the question clear?
 
I think so
 
11:40 AM
Pyth is giving me this weird error: AttributeError: 'module' object has no attribute 'lru_cache'
Is that normal?
 
0
Q: Create an RGB color model where luminocity can be applied without losing saturation over white/black boundaries

series0neUsually, if you want to be able to apply lightness or darkness to a color, you would use a color model such as HSL, where Luminosity is one of the component parts of the color. Applying lightness (increasing luminosity) or darkness (decreasing luminosity) would, with respect to a HSL color model ...

 
12:28 PM
TIL @Adnan is a doge
and a 'very advanced graphic designer'
 
Hahaha
 
@MartinBüttner Isn't that just finding a basis for the vector space that they generate over the field F2? Should be easily solvable by linear algebra.
 
12:46 PM
2
Q: Swap, delete and repeat

AdnanIntroduction Let's observe the following string: ABCDEFGHIJKLMNOP If we swap the ends of the string, which are these: ABCDEFGHIJKLMNOP ^^ ^^ We get the following result: BACDEFGHIJKLMNPO After that, we delete the ends of the string, which in this case are B and O. The result ...

 
@Zgarb yeah it is... Shows how much linear algebra I know
 
@MartinBüttner interesting reply!
sorry I didn't spot that myself
 
1:02 PM
Was just working on brute-forcing the APL challenge
 
@Zgarb yeah thanks. Figured it out in the meantime. I vaguely remembered Gaussian elimination from some lectures years ago and it also occurred to me treat the XOR as addition mod 2, but somehow I didn't connect them.
 
Was wondering why it stopped
 
what?
 
It turns out that the Mathematica kernel was trying to evaluate 20¹⁶!
 
@El'endiaStarman you might want to correct Nathaniel Johnston in nathanieljohnston.com/index.php/2009/05/…
20^16?
 
1:03 PM
(That is, twenty to the sixteenth power factorial)
 
oh
 
It handles 20¹⁶ just fine
Odd
My computer said that I had 3 ecm processes running
I have 4
 
1:24 PM
0
Q: Artificial intelegence

Asad MalikPlz Help to Solve This Use A* algorithm to solve the 15 puzzle problem: Your algorithm should take as input a possible configuration of the Puzzle. The output should be the sequence of steps required to solve the problem. Your program should also be able to generate a possible puzzle. The pseud...

 
Yay, now I have 63% CPU usage!
 
@Adnan @Sp3000 Posted and posted
 
@VoteToClose I'm having trouble seeing how your submission is a "function"
 
@Sp3000 m is the method caller command.
When on a separate line, it's considered a function. You call it by pushing the index of the line onto the stack and then calling m.
 
As far as byte-counting goes, I think it makes more sense to count the preceeding newline as part of the function. Otherwise it's basically the same as a snippet (might need a meta post though)
 
1:38 PM
@Sp3000 That makes sense. :P I changed my byte count.
 
But yeah, sorry, still thinking about it. It's... an interesting situation o_O
 
Yeah, Vitsy is weird when it comes to methods.
I'm posting to meta about it.
 
1:51 PM
Just for extra context: Gol><> and Jelly are pretty similar, with the function-on-new-line thing
(Minko might be too, but I forget)
 
I did not know that Gol><> did that. o.o
@Sp3000 Minko is 2D, it doesn't have different line ability.
 
Yeah, but it also has ^ and v specifying direction. :P
 
So does Gol><> :P
 
I suppose gosub could be used as a function. o-o
 
1:54 PM
1
Q: Do I need to count the newline for Vitsy methods?

VoteToCloseBackground: Methods in Vitsy are defined by specific indices of lines of code. For example, if I wanted to call the second line of a program, I would call 1m to specify the first index of the methods. In a recent challenge, I was also asked to count the preceding newline, which sort of made sens...

 
I mean, you could execute the function by both getting there with <^>v and gosub, which is a bit weird
But in gosub's case there's actually a "return" function builtin
 
o-o
 
The problem with brute-forcing programs is that you literally have to deal with every possible scenario
 
wat
 
then optimise it?
meaning skip impossible ones
 
2:00 PM
I am
 
or port to c++
or pypy
 
I need to link to the Mathematica kernel to run it
It's just that you have some like 20^16! that will never complete
 
add a timeout?
 
I have
 
I was going to ask "20^(16!) or (20^16)!?" and then realized that's a stupid question because either way the number is still enormous.
 
2:04 PM
it's 20^(16!)
 
Its FullForm is Power[20, Factorial[16]]
 
So slightly smaller than "holy cow ridiculously enormous" and just in the "really enormous" range. Pretty sure my brain can't handle numbers that big anyway.
 
@Dennis Pull Vitsy?
> Oh the difference 2 bytes made...
^ Name of the last commit. xD
 
Basically, it's asking for 20^20922789888000
About 10^10^13.4349
 
I'm an idiot
 
2:22 PM
@VoteToClose Pulled.
 
0
Q: Longest path on a 2d plane

BluePillYou are provided a set of arbitary, unique 2d Cartesian coordinates: e.g. [(0,0), (0,1), (1,0)] Find the longest path possible from this set of coordinates, with the restriction that a coordinate can be "visited" only once. (And you don't "come back" to the coordinate you started at). Example...

 
Speaking of large numbers, I watched an interesting "standup maths" video yesterday evening discussing card shuffling.
There are 52! ways to shuffle a deck of (US) cards ... which is mind-boggling huge. Let's compare that to 52! seconds. Suppose we're standing on the equator, and every billion years we take one step forward. Another billion years, another step, etc. (let's assume we can float across the water) Every time we fully circumnavigate and return to where we started, remove a 0.5mL drop of water from the Pacific Ocean and set it aside ...
Continue this process until we deplete the Pacific Ocean. Once we've done that, place a single piece of standard paper on the ground and put all the water back into the ocean. Now continue that process until we've accumulated enough paper to reach from the Earth to the Sun ...
And that's about 1/3rd of the way through 52! seconds.
 
2:37 PM
"until we deplete the Pacific Ocean" bothers me. Water will flow in from other oceans :P
 
@TimmyD luckily it's not (20^16)!!!, or (FOOT^10)(10^100)
 
Well, sure. But the Sun will be a cold husk looooong before that happens anyway, so just roll with it.
 
@TimmyD And the Earth will be long gone before that first step :P
 
pfft... only 6.134168e+61 years
when we take 5 steps the sun will be dead
 
2:42 PM
The main problem with the analogy is that the number's just too damn big, even still. The steps, water, paper, just reinforces that it's a big number, but it still can't give a clear idea of how big, because meat-sacks don't think that high.
 
@Geobits Maybe if someone were high, they could think on it better ... ;-)
 
They could at least think they were thinking on it better ;)
 
http:// perhaps?
 
damnit
 
2:45 PM
Kingdom of Loathing... very nice.
Haven't seen that in years.
 
haven't played that in years.
really fun, just i don't like the way it resets daily, because i never have the time
 
I always preferred "Ugly bags of mostly water."
 
I wonder if it works smoothly on mobile. I bet more time would open up for it if it did.
 
@somebody You got Kings of Leon's "Use Somebody" stuck in my head
 
Are you roaming around, always looking down?
 
2:49 PM
why is this called the nineteenth byte?
 
31
A: Let's think of a creative name for our chatroom

dmckeeWell, the traditional generic name for the country club bar is "the nineteenth hole", which suggests The Nineteenth Byte or something like that.

 
Bonus: it's 19 characters long
 
@Geobits but there's Jelly now, shouldn't it be the sixth byte?
 
Jelly is not my master.
 
Cjap/Pyth - the eleventh/thirteenth byte
Java - the five hundred and first byte
 
2:52 PM
Measure the output instead :P
 
The 52! byte. We'll be here a while.
 
Okay, don't try to measure the output if you're using shtriped.
 
the infinitieth byte for stuff like while True: print 1
the zeroeth byte
 
Fun fact: there are exactly nineteen bytes in while True: print 1
you win this morning's coincidence award
 
Is that why we seem to revisit the same conversations over and over again?
 
2:55 PM
@Geobits I thought we were on a Möbius strip.
 
Nah, there are sometimes more than one side to a conversation.
 
But it's the same side ...
 
Not if we're talking about listy tips :D
 
Tipsy lists?
 
hic
 
2:59 PM
Ugh, PowerShell ... so dang frustrating sometimes
The range operator .. isn't context-sensitive
 
I kinda want to object to your use of "sometimes", but I'll try empathy instead:
Yea, that sucks.
 
So, if you do something like 4:-3 in Python when indexing an array, it understands you want from the 4th element to the 3rd-from-the-end ... Not in PowerShell, it creates the range 4..-3 as 4,3,2,1,0,-1,-2,-3
ಠ_ಠ
 
Well, to be honest, that's exactly what I'd expect. I'm not used to brainwashed into python range syntax, though.
 
Le sigh
 
OOC, what does python do when the range doesn't make sense? Like if you did [4:-3] on a two-element list. Just give back an empty list or go all exceptiony?
 
3:06 PM
Python 2 gives an empty list
Python 3, too
Hovever, [4:-3:-1] gives the list reversed
 
I guess that kinda almost makes sense >_>
 
After all, a program's output only makes sense if its input makes sense
 
Sure, I was just curious about how the language handled it.
I've used it a little bit, but I tend to be careful about indices because other languages aren't so.... tolerant?
 
Fun fact: In Mathematica, [[4;;-3]] outputs:
Part::take: Cannot take positions 4 through -3 in {x, y}.
 
3:17 PM
What does it do?
 
@LegionMammal978 That's a pretty confusing way to put it - L[4:-3:-1] isn't the same as L[4:-3][::-1]
 
@LegionMammal978 (8^3)!
 
It's actually just "start from index 4, go down to index -3 (exclusive), by steps of -1". For a two element list, 4 is off the right and -3 is off the left, i.e. you get the whole list
(I think?)
 
@VoteToClose, I should learn Vitsy sometime
 
\o/
My child would be honored.
Hmm. I need to change eval and add try/catches.
 
3:23 PM
There
"!dlroW ,olleH"Z
 
\o/
 
What does \o/ mean?
 
Celebrating person.
Long haired version: \p/
Raising their arms up in the air: \ / Their head is: o or p
 
I see
\O/
Is [1] actually an infinite loop?
Or is it just TIO being slow
 
It is an infinite loop - it'll run out of memory eventually, better way to do it is 1[]
 
3:28 PM
Okay, so loops peek the stack? You should add that
 
Or you can just do <
Peek the stack? o.o
 
As in, loops don't pop the top stack element to check it
 
Marco the stack, and it Polo's a response.
 
They do pop if it's zero, then they exit.
But otherwise, they don't.
I'll add that.
 
Given a set of strings, determine if a list of those strings needs to have a delimiter
^ How would you do that?
 
3:32 PM
sniff Smells broad
 
is it simply testing to see if any one of the strings is a strict substring (at the starting position)?
 
Where did you find it?
 
my mind
actually, that's still not true
 
Dang, the kernel has become sentient
sh: 1: 6: not found
sh: 1: 16: not found
sh: 1: 016: not found
sh: 1: 2016: not found
sh: 1: 6: not found
sh: 1: 16: not found
sh: 1: 016: not found
 
given three strings "Pop", "Popcorn", and "Kernel", its always possible to tell the order of the strings, even if there is no delimiter
 
3:35 PM
@LegionMammal978 Fixed the README for loop details.
 
Does this also apply to if statements?
 
If statements always pop.
 
Although I wish they didn't. >.> That might change. I'm still mulling it over.
 
You could just use D(
 
3:38 PM
I do use that. Quite frequently.
 
It's just annoying sometimes. :P Especially for if/elif/else structures.
 
You could possibly add a rational type
 
What do you mean?
 
Internally represent it as a ratio of two BigIntegers
 
3:44 PM
Hmm. Only problem is that I've mainly been keeping the stack stacks as only Doubles.
 
I could try to help implement it for you
 
:D I mean, it would just be a lot of casting and ClassCastException catching.
 
Also, integer literals are only one-digit, correct?
 
And probably a private subclass.
 
Let me look
 
3:46 PM
could you implement it as a complex number with special methods?
 
@LegionMammal978 If you mean when putting them into the code, yes, though you can do tricks with eval to make it multi character (although using characters is just as effective).
So, for example, on an empty stack: '123'n would leave 321 (the number) on the stack.
 
I see
Currently looking at the code
 
I need comments bad.
 
Why does instruct have three levels?
Unless you have 3D programs
 
@LegionMammal978 Multi-class systems, so that higher in the call stack is still loaded.
It's more like I have multiple 2D programs. :P
 
3:49 PM
Can't figure out classes
Not very well documented
(As in ;e and ;u)
 
Yeah, I know. I wasn't sure how to put it. D:
;e declares a file as a superclass (though it's not really a superclass in Java terms, more like it's the one file that can use the shorthand K command).
;u declares a file as used by the current class and is called with k commands.
 
Wait, when did multiple files come in?
Just looking at it from the perspective of someone trying to program in it
 
The docs don't talk about it at all
 
I never finished the docs. D: School got in the way.
I'll do a bunch of them tonight.
 
3:54 PM
I see
Good thing I have some experience half-heartedly attempting to read the source code of esolang interpreters because the "documentation" page only has some examples
And then ranting about it
And 90% of them aren't even implemented
 
Basically, to use separate files, you declare that you wish to use them with ;u. If the FileHandler can find them, it'll say, "Okay, then I'll add this to the list of files we might use in the future." You call methods inside the separate file with <index in use><index of line>k
 
Which file do you put ;u in? The program file or the class file?
OH I SEE
It's like Python modules, right?
 
Look at vitsy1,2, and 3 in this.
@LegionMammal978 I've never dealt with Python extensively. D:
 
Where say you have foo.py which defines some methods
 
Yes.
 
3:59 PM
Say it defines bar
 
I'll be back later, gotta do some stuff.
 
k ping me when you're back
 
@LegionMammal978 I'll be in and out for the next hour or so - I'll be back after then.
But I'll still be around.
And, yes, I think that the modules pretty much describes this.
@El'endiaStarman What's the purpose of $j in Minko?
I get what it does, but I don't know why it's there.
 
4:27 PM
Challenge: I'm thinking of a function f. If you give me a real number n, then I will give you f ( n ). Try to find the definition of f.
 
2
 
@aditsu f (2) = 1/2
 
1/2
 
@VoteToClose f (0) = 0
 
4:29 PM
ok, 3.5
 
Wait forget that
Give me a min
 
ƒ(π) ?
 
@LegionMammal978 Division problem solved: setind(2, index(2).divide(top(), 20, BigDecimal.ROUND_HALF_DOWN));
 
@Cyoce that's clever :)
 
@aditsu ty¯\_( ͡ಠ ͜ʖ ͡ಠ)_/¯
 
4:31 PM
@LegionMammal978 Is the function possibly piecewise?
 
Not telling
Let me implement it so I don't have to do it by hand
 
playing hard-to-get, I see ( ͡° ͜ʖ ͡°)
 
Although ƒ(π) is well-defined
 
-1, insufficient test cases :p
 
@Cyoce ಠ_ಠ -1 for Lenny
 
4:33 PM
@VoteToClose Compromise:¯\_( ͡ಠ ͜ʖ ͡ಠ)_/¯
 
holifuq
That's terrifying.
-2 ಠ_ಠ
 
this reminds me of another problem: there's a rule for integer triplets, (2, 4, 8) satisfy the rule, you can check any other triplets and once you are confident, you can try to guess the rule (but you have only one attempt)
basically, it's a function f:ℤ³→{true, false}
 
┻━┻ ︵¯\_( ͡ಠ ͜ʖ ͡ಠ)_/¯︵ ┻━┻
 
@Doorknob (slow-clap) ( ͡ಠ ツ ͡ಠ)
 
4:38 PM
doing apt-get upgrades over school wifi is awesome. 4.7 mb/s as opposed to my typical ~ 200 kb /s
 
Where do you go that has 200 kb /s? Home?
 
yeah
 
How do you play games?
 
@Dennis Wait no not yet never mind.
 
Is it a string/can modem?
 
4:39 PM
@Rainbolt why would you need fast internet speeds to play Nethack?
finally got that glibc buffer overflow patch :P
 
Ooooh. No wonder Nethack is so popular.
 
Oh THAT'S why he likes nethack so much.
 
200kbps is still faster than dial-up
 
I'm digging through my brain and yep.... yep... yep... everyone I knew in college computer science who mentioned Nethack looked like a poor person.
 
My hobby: making a cached random number generator, then asking people to figure out what the function is by giving me input to the function
 
4:41 PM
I remember playing online with 26.4kbps dial-up back in the late-90s
 
@TimmyD L.O.R.D.
 
Every now and then the stars would align and I could connect at 28.8 ... and whoa, momma, hold on to your britches, that was fast
 
@Geobits Llamas (Optimizer) Really Dance
 
Oh, I forgot there were kids in here. You probably don't know what a BBS game is :P
 
The absolute cheapest plan that is even offered in my neighborhood is 50 Mbps download / 5 Mbps upload
 
4:43 PM
@Geobits No, full-blown FPS. Usually MegaTF, or Action Quake2.
 
my first remote connection (not exactly internet) was via a 14.4k modem ^_^
 
I think there's a 30 offered here, but it's like $5 cheaper than the 50.
 
And the only reason I was able to play Action Quake2 was because it was included on a demo CD inside an issue of PC Gamer, so I didn't need to download the couple-hundred-meg install.
 
@Dennis Pull Vitsy? :P
 
@aditsu I remember finally bugging my dad enough to get a 28.8 :)
 
4:45 PM
@VoteToClose Pulled.
 
But I think my first was 9600.
 
I somewhat remember downloading my first jpg… it was a long and exciting experience
3
 
Wait... I found a company that offers dial up
They don't list speeds
 
Well played. ;P
 
4:45 PM
@aditsu It was roughly as fast as a dot matrix printer :D
 
@aditsu ( ͡° ͜ʖ ͡°)
2
 
well, not quite that exciting :p
 
@Geobits You're lucky. I started with 2400.
 
my first internet speed was 2Mbps
 
4:47 PM
@Dennis Yea, I know we had a 2400 at some point, but I'm pretty sure I wasn't using it then. I didn't really get started until I was around 10.
 
> We’re sorry, chat isn’t available for dial-up Internet support.
 
(personal at home)
 
Dang
I can't get actual numbers without calling them
 
@Optimizer Two millibits per second? Now that is slow!
 
> chat isn’t available for dial-up Internet support
 
4:48 PM
> Internet isn't available for dial-up
 
> Available isn't Internet for dial-up
 
2μbps
 
> dial-up isn't Internet
 
> that dotted bar is not fooling anyone
 
4:51 PM
> dial-up being Internet isn't fooling anyone
 
> fooling anyone isn't being dial-up
 
> isn't anyone being fool dial-up'd
 
@TimmyD That's a strange place for an apostrophe imo.
 
> order in words wrong
 
4:53 PM
@Geobits Yeah, I do'n't' know what's going on there.
 
@aditsu it's correct. that was a question
 
Oh I remember then. Should we do it all again?
 
oh, there I started it again.
 
@Geobits that is not a great idea, it might give me diagrams
 

« first day (1847 days earlier)      last day (3000 days later) »