« first day (1196 days earlier)      last day (3647 days later) » 

12:01 AM
well, the PDC coding phase is over
only 5 entrants :(
 
12:42 AM
@Joe wait, um, what art thou referring toeth?
 
1:11 AM
205
Puzzling

Proposed Q&A site for those who study the creation and solving of puzzles.

Currently in commitment.

yay!
 
1:59 AM
Anyone feel like reviving the old "should we enable latex on PPCG?" question?
8
Q: LaTeX/MathJax support would be nice?

TimwiSeveral StackExchange sites have support for mathematical formulae using MathJax (LaTeX syntax). Many Code-Golf challenges are mathematical in nature. Equations come up very often in discussions. Therefore, I think it would be very useful to enable support for this on here as well? Test to see...

 
Erm... not really
About puzzling, I can't wait @Doorknob
 
Me too :D
 
Are you interested in modding there?
 
... maybe. One site is probably enough to start off with. :P
I'll certainly participate quite a bit
 
@Quincunx Well, there you go. You just need to fix your having a windows 7 computer. — AJMansfield Apr 25 at 21:53
Haha
 
2:07 AM
Well, if you are a mode here...
 
@Doorknob I wonder though
Does creation of puzzles cover design of so-called "twisty puzzles"
 
2:39 AM
-3
Q: Reversing golfing - Ungolfer

hannes karppilaHere is many guestions tagged code-golf. Then code isn't usually commented and always "not human-readable". Now your mission is write The Ungolfer, program that ungolfs code. It must be written in language that it ungolfs It can ungolf code and comment it so good that even starting programmer c...

 
 
3 hours later…
6:04 AM
New post! Generate a Map!
 
This was a lot of fun:
1
Q: Why does time incorrectly report such a fast time?

QuincunxI was playing around with big numbers, and wrote the following code: import time def ispow2(n): return not n & n - 1 start = time.clock() ispow2(2**100000000) end = time.clock() print(end - start) Surprisingly, this outputs 0.016864107385627148, and insanely short amount of time. Howeve...

 
 
4 hours later…
11:22 AM
@TheDoctor Sounds like something Mensa should get behind sponsoring. ;-)
 
11:46 AM
hi
any ideas why the close vote for codegolf.stackexchange.com/questions/26855/… ?
 
11:57 AM
@user2179021 Too broad.
 
@Peter Interesting. Can you say more?
@PeterTaylor sorry.. I don't understand how this chat room works yet
but now I know you are peter taylor, may I say how impressed i have been with some of your answers!
 
@user2179021 I've expanded in a comment.
Thank you.
 
@PeterTaylor Thanks but I assumed the popularity contest flag at least overcame the first complaint
can you help refine/improve the question?
please
I think it is potentially a fun challenge
 
12:13 PM
I would send a close vote too, but as a mod, that would be binding, so I won't. But then again, I think too many people post popcons, so. ;-)
 
12:53 PM
@PeterTaylor Sorry I got cut off. Any help improving the question is gratefully received.
@ChrisJester-Young is that for the fake boot question you would send a close vote?
 
hi
 
1:06 PM
@Chris they're not as much of a problemo as code-trolling
 
1:25 PM
good morning
 
@user2179021 A good popcon question still needs a reasonably clear goal (example 1, example 2). As it stands, the range of possible answers to your question extends from simple one-liners like CLS:PRINT"OK"all the way up to full-screen animations with stereo soundtracks. It's way too broad.
 
1:40 PM
greets ;3
 
Hi
2
A: A completely fake reboot

TheDoctorBash + Coreutils (Linux and possibly Mac) echo "The system is going down for maintenance NOW. clear sleep 5 dmesg|while read i; do echo $i; sleep 0.1; done cat /etc/issue login

^^ wasn't too hard...
 
it's easy on a non-graphical os :P
i guess
;_;
 
Should we add a to that?
 
2:08 PM
@user2179021 Yes, but like I said, I won't, because 1. mod votes are binding, and I don't like to exercise that much power unless it's clearly egregious, and your post isn't; 2. I'm personally biased against popcons, but as a mod, I have to follow community consensus, so I'm not going to close a question just because it's a popcon unless, as a community, we decide to ban them.
Whereas, once we have a community consensus on code-trolling, I will enforce that with pleasure. :-)
 
2:37 PM
augh, why is java so bad
Changed my code to this:
import java.math.BigInteger;

public class Factorials {

    public static BigInteger fac;

    public static BigInteger two = BigInteger.valueOf(2);

    static BigInteger mul(BigInteger start, BigInteger end) {
        if(start == end) {
            return start;
        } else {
            BigInteger mid = start.add(end.subtract(start).divide(Factorials.two));
            return Factorials.mul(start, mid).multiply(Factorials.mul(mid, end));
        }
    }

    public static void main(String[] args) {
And I found the top of the error:
Exception in thread "main" java.lang.StackOverflowError
	at java.math.MutableBigInteger.divideKnuth(MutableBigInteger.java:1187)
	at java.math.BigInteger.divideKnuth(BigInteger.java:2049)
	at java.math.BigInteger.divide(BigInteger.java:2030)
	at factorials.Factorials.mul(Factorials.java:15)
	at factorials.Factorials.mul(Factorials.java:16)
	at factorials.Factorials.mul(Factorials.java:16)
	at factorials.Factorials.mul(Factorials.java:16)
	at factorials.Factorials.mul(Factorials.java:16)
	at factorials.Factorials.mul(Factorials.java:16)
@ChrisJester-Young Also, KoTHs. Seriously fun stuff.
 
@ChrisJester-Young We've already got a consensus on new code-trolling questions: the current discussion is about the fates of the old questions.
 
I very nearly missed Prisoner's Dilemma v.2, just remembered. For some reason I didn't get a message about Joe's reminder.
:c
 
divideKnuth() ???
 
@TheDoctor That's a part of BigInteger, nothing to do with my code. It's just that my infinite recursion is killing all the memory.
@TheDoctor No idea where the infinite recursion is coming from...
 
3:06 PM
hmm
 
3:23 PM
@PeterTaylor Good stuff. I will prepare to kill new ones with fire.
 
@Trimsty Have you traced/logged start and end to see what's going on? That's where I'd start.
If you're working with sufficiently large numbers, though, you might not have infinite recursion anyway, just "too much". I don't know what range it's supposed to be working on, but you're branching to mul twice on each call.
 
@Geobits Factorial function; I ran it with input 9. It throws this error on everything.
@Geobits It's supposed to act like merge sort, except multiplying instead of merging. Weird how it acts this way :/
Wait
it's so obvious
 
3:39 PM
On the second call mul(mid, end) it's getting stuck on 0,1
 
I forgot that in Java, == compares pointers
right?
 
wow. Just read that craziness from yesterday.
 
0
Q: Duct tape coding: new tag?

TheDoctorI have seen several (2) new questions that are supposed to be copied, not written, from StackOverflow. From the original question: Your mission is to make the most interesting program you can -- without ever writing a single line of code. You can only use code which is already written for you...

 
@Trimsty Changing it to equals() doesn't fix it. The second branch never has start = true
 
3:41 PM
@Geobits According to my analysis, it's getting stuck on the first call. See values of start, end:

0, 9
0, 4
0, 2
0, 1
0, 0
0, 1
0, 0
 
Right, in the (mid,end) call.
If you call it with 0,1, mid will be 0, end will still be 1. Every time.
 
@Geobits So 1/2 == 1?
 
No, 1/2=0, but you don't change end, so it's always 1 at that point.
So mid is 0, and it will never equal end
 
@Geobits Oh, it's rounding the wrong way
@Geobits Wait, no it isn't? :P
@Geobits The first branch is supposed to end on 0, 0 and the second on n, n
@Geobits The inf recursion is definitely in the first branch, as start is always 0 (see above)
although there is a case where end is also 0
why isn't it triggering then?
 
It is triggering then. Okay, you're doing a depth-first call. It goes all the way down the first branch until 0,0. That branch then ends. It goes up to the last call 0,1. The two paths it takes then are 0,0 and 0,1 (start,mid) and (mid,end). end is always 1 on the second call.
 
3:54 PM
@Geobits Oh, wow, I'm stupid. :P
@Geobits I get it now ;D
 
:D
 
@Geobits for the first branch, mid needs to round down. For the second, up
 
No
You need to ensure that the two ranges don't overlap at all.
 
@PeterTaylor Or I can do that.
@PeterTaylor I'll just add one to the mid for end.
 
The first argument also wants to be BigInteger.ONE rather than BigInteger.ZERO, or you won't get the factorial.
 
3:58 PM
@PeterTaylor Thanks :P
It works!
Thanks guys
wait, it gives the wrong answer ^^'
xD
For 16 is gives 2238138
(interesting but wrong nonetheless)
 
That's not what I get with those two mods (add one to mid for second branch and change ZERO to ONE)
 
@Geobits do you get 20922789888000?
 
Did you add a line to print the result rather than nanotime? Your original code pasted here doesn't print it.
 
And yes
 
4:02 PM
@Geobits I forgot I was printing the time and not the answer.
epic facepalm
it's correct!
yay!
Time to find the highest it can go in 10 seconds
WOAH
It just tackled 900000 in 11 seconds
Big speedup over 90000 @ 10
amirite?
 
Not necessarily. You need to time the naïve Python code on your machine to get a fair comparison.
 
@PeterTaylor yup
 
If you want another point of comparison, have a look at codegolf.stackexchange.com/a/1641/194
 
4:21 PM
so my mom put sports wrap on my legs
which have hair
bad idea
xD
 
Anybody here know WPF and Binding?
 
@PeterTaylor I can save 3 chars on that answer if you're interested
 
I am having a little trouble, and my question on SO didn't receive much attention.
 
@hosch250 I do, but I looked at your question and I can't help you because it is C++.
 
@ProgramFOX Thanks.
Right now, my problem is only a linking error.
 
4:32 PM
@PeterTaylor four now
 
@durron597 BigInteger.valueOf ? I might revisit it when I have time to thing deeply.
 
@PeterTaylor OP's system is faster than mine.
 
while(o.signum()>0){r=r.multiply(o);o=o.subtract(i);} -> for(;o.signum()>0;o=o.subtract(i))r=r.multiply(o);
and
if(n>1)g=new G(O.subtract(r),I,n-1); -> g=n>1?new G(O.subtract(r),I,n-1):g;
 
BigInteger is slow, Imma use GMP
nuu
no gmp wrapper for java
anybody know how to install github.com/tbuktu/bigint? xD
i'm stumped
 
5:06 PM
um upgrade to Java 8?
err OpenJDK 8, sorry
 
@durron597 :P
I'm on Windows, so
I'll just post as-is
0
A: How much can you quickly multiply?

TrimstyJava - 10.9 n = 885000 Mergesort-y. import java.math.BigInteger; public class Factorials { public static BigInteger fac; public static BigInteger two = BigInteger.valueOf(2); static BigInteger mul(BigInteger start, BigInteger end) { if(start.equals(end)) { r...

 
@Trimsty Try -Xbootclasspath
err google spat out the java 6 one. just change 6 to 7, it's the same option.
 
@durron597 Thanks! Too bad I get errors while compiling that. :P As at the bottom of the question, Recommendations for arbitrary-precision high-speed Java integer libraries? :P
 
5:44 PM
hi
 
Hi @TheDoctor
 
こんにちは
 
w'hai
 
hola
 
in duct tape, if i find a snippet that says for(int i = i; i < foo; i++) { // your code here } or something
in one single block
do you think i can put something inside that snippet?
 
5:50 PM
@durron597 Pretty sure you can cut duct tape ;3
 
As long as you copy the inner code from somewhere
 
@TheDoctor Well sure
 
so yes
sudo su -
sh-4.3 $ echo Muahahahah
sh-4.3 $ userdel Quincux
 
sh-4.3 $ usermod --add-reputation 5 iamkrillin
 
6:16 PM
oily doughnuts + icing sugar = irresistible = indigestion
:S
 
oily doughnuts + ((icing sugar = irresistible) = indigestion) ? I don't get it ;)
 
@Geobits I'm using = in the equality assertion sense, not the assignment sense
a = (oily doughnuts + icing sugar); a.irresistible = True; self.indigestion = a.irresistible
(can, of course, be optimized/golfed to this.d=0, but hey)
(and &{={d;0 in Ena!)
i really wanna make a DoorknobChatbot clone
:P
 
if(!self.hasIndigestion()) self.ingest(new Doughnut[self.stomach.getCapacity()+1]);
 
@Geobits aw.
 
@PeterTaylor hi.. are you still about?
 
6:30 PM
bazinga.
 
do{self.ingest(new Doughnut(Oil))}while(doughnutCount > 0 && !self.hasIndigestion)
 
0
A: Duct tape can fix mothersday

durron597Java Start with these: http://stackoverflow.com/a/19509316/1768232 String myStr = "hello"; String myStr1 = "hello"; http://stackoverflow.com/a/8341513/1768232 String[] history = new String[] { "hey", "you", "Mom" }; http://stackoverflow.com/a/23300004/1768232 String editednames = titleCa...

 
Hi ;D
@14mRh4X0r 3 years old!
 
It works though!
 
6:33 PM
o:
usrs?
 
I'm getting ~10 times faster
in comparison to java.math.BigInteger
 
:P
that's nice...
 
You need some gcj stuff though, so it's a bit of a hassle to set up
 
o:
What happens if you use that AND compile your code with gcj instead of using the jvm?
 
Well, then it's using a shit ton of memory
More than 12G for 100,000
 
6:36 PM
uh
Well.
 
At least on my machine :P
 
that's quite strange
> optimizing
 
Yup. But the same happens if I use the gmp-nogcj.jar
 
@14mRh4X0r it IS a heavily recursive function
 
I'm not recursing, I'm using a for-loop :P
 
6:39 PM
@14mRh4X0r Ah, interesting.
 
Because recursion is your enemy if you're using large numbers. You'll get StackOverflowErrors and such
 
@14mRh4X0r I'm doing it to keep the number multiplication complexity to a minimum with no care for memory consumption ;3
 
*shrug*
It's really not that hard, I'm just counting from 2 upto n, instead of recursively from n downto 2
 
@14mRh4X0r My function for 4'd be ((1)*2)*(3*(4))
= (1*2)+(3*4)
= 2*12
linear is 1*2*3*4 = 2*3*4 = 6*4
 
Ah, you're doing merge recursion, or whatever it's called.
 
6:45 PM
indeed
my favourite kind ;3
 
Yeah, I had a solution that divided the numbers into exponentially smaller chunks, and then calculated each chunk in a separate thread
But then I read it had to be single-threaded :<
 
:c
it sounds fast
I installed gcj and the original gmp and also the new gcc :P
 
Doorknob officially has a diamond next to his name now.
 
Congratulations @Doorknob
 
Only thing I see wrong now is that he's from Texas :p
 
6:52 PM
@Geobits xD
 
Yeah, the pure-java variant had a speedup of about 8 times (which is the number of threads my cpu has), imagine what it would've been with GMP :P
Ugh, train internet is not the best kind...
 
@14mRh4X0r

Windows sucks 3;

bash-3.1$ make
File not found - *.java
wget -c -SN "ftp://ftp.mirrorservice.org/sites/sourceware.org/pub/gcc/releases/g
cc-4.6.1/gcc-java-4.6.1.tar.bz2"
'wget' is not recognized as an internal or external command,
operable program or batch file.
make: *** [gcc-java-4.6.1.tar.bz2] Error 1
 
@Trimsty Maybe MinGW or cygwin work?
I'm not entirely sure
 
@14mRh4X0r That's MSYS' bash :P
@14mRh4X0r Nothing ever works on my computer anyway. :P I guess this is why I have my CentOS box.
 
7:03 PM
Haha, ah well. I'm not using Windows, so I wouldn't know :P
 
screen won't work :/
quite a sick joke this
computer is running fine, mouse is lighting up, but you don't have a display
and yet there's a working on right next to you
;_;
Okay
now I really wanna make a chatbot
i have ideas flowing around in my mind like thoughts in minds do usually o:
now i just have to figure out how...hm
wait, they haz api
:P
api no chat ;_;
 
8:04 PM
Idea: I'll write an API for the bot to interface with whatever ajax api i can conjure up (possibly stealing it from doorknob) (hey we can make a library!)
Time to be a jerk and ping @Doorknob? I only have 20 minutes left :c
 
9:05 PM
@ChrisJester-Young Why would you do that?
After all the discussion yesterday about how questions like that one belong on main, you community wiki'd it?
 
9:24 PM
@Rusher From what I saw in chat, he's operating on the assumption that there hasn't been sufficient community consensus about it yet, so the status quo holds. I wrote an answer on meta about this (among other things) earlier if you're interested. I agree that they should be allowed (and encouraged), and I'd like to see a consensus actually take hold.
 
@Geobits If I did anything yesterday, I proved that the rest of the community is solid on the issue. Trust me, there is a consensus
 
lol, I agree, and I was surprised (and a bit saddened) to see your post go CW.
I would say they're more solid on "tips don't belong on meta" than "we should be asking more non-challenge questions", though.
 
Oh well hold on, let me solidify the community on it real quick. I think tips questions should be CW.
I also got something out of all this. I learned how to alias member functions in Python
 
Yea, I was trying to put something together for it with a custom function rather than class, but the answer you got beat mine out. I'm not much good with python, though.
 
Was it shorter?
 
9:32 PM
@Geobits Congrats to @Doorknob!!
 
@Rusher It was a bit shorter than the original method you were using (assuming more than 6-7 uses, it still lost out on 4), but quite a bit longer than using a class. I didn't see the point of posting it then.
 
@Geobits Well, now that Doorknob is a mod, I'm sure they are within their powers to un-CW it.
 
Doorknob is a they now? No wonder their word is truth.
 
@Rusher I have a policy of using "they" for people with ungendered usernames, unless they specify a different pronoun preference.
@Rusher Just like I'd refer to you as "they" unless you have a different preference.
 
9:35 PM
I supposed it is more polite than "it"
 
@Rusher It means that we do not assume everyone is "he" by default.
 
It's a doorknob!
 
Is it weird that I associate genders with objects?
A Doorknob seems masculine to me
 
Now that Doorknob is a mod, it is within s/he/it/they/thing's powers.
 
@Rusher Depends on what your native language is. If it's French or Spanish, not surprising.
 
9:36 PM
What if we just use Doorknob as a pronoun
> Now that Doorknob is a mod, it is within Doorknob's powers.
 
@mniip You can, if you want. I'll still use "they" by default.
 
Or just shorten it to 'knob'. Then it's no longer than 'they' :)
 
I was being sarcastic
 
The knob's killing tags with fire today.
Yes, I like 'the knob' better.
 
@ChrisJester-Young Ah yes. Doorknob is pomo in Spanish. But I didn't know that 10 seconds ago
 
9:38 PM
@Geobits T. D. Knob
 
I like dk (dee kay)
 
Decay
 
It eliminates the need to use a pronoun, because dk is so short to type
 
Decay Country for Wii was okay.
 
9:40 PM
I just pictured a doorknob country on wii
where you spend all day turning handles
 
DK is too long to type :v
why don't we just call him The D
 
How about ?
 
Oh man I referred to like 50 times already in this sentence.
 
Not really no
It's a word of 0 letters. Words are separated by spaces, there has to be a double space
 
I actually did double space
It shortened it
 
9:42 PM
Separated-by-spaces?Maybe-in-your-puny-language.
 
Damn true
I did actually type a nonbreaking space this time
 
Orbetteryet,whydoweevenbothertospaceatall?I'msureeveryoneherecanreadthis;theJapa‌​nesedon'tusespacesandeverythingworksoutfineforthem.
 
Yeahthisiscodegolfanyway
 
Wow, I have to forcibly restrain my thumb from hitting the space bar when typing that much.
Muscle memory is a bitch.
 
It'slikehowAussiestalk
 
9:45 PM
thatescalatedquickly
 
(An Aussie said that to me - I'm not racist)
 
IusedtohangoutwithsomeAussiemilitaryguysoverinIraq.Thereweredefinitelysomequick-‌​talkingones,butsomehadamore"normal"cadencetotheirspeech.Theaccentwasalmostunintel‌​ligible,though.
Afterafewdaysyougetusedtoit,though.
 
@Geobits Have you listened to Jason Ellis on Sirius XM?
 
Don't think so. I don't have Sirius and the name doesn't ring a bell.
 
10:01 PM
 
10:26 PM
Oh wow, codegolf.stackexchange.com/q/26852/7162 this is something I had in my private mini-sandbox
 
11:12 PM
Happy moderation, @Doorknob
3
hi
 
how iz u?
 
not too bad
not staying for long though... wanna go to bed... just wanted to have a look what's going on (if anything)
probably gonna write up a quick answer to your meta post and then head off
 
Doorknob == mod, but i haven't been here for very long either...
 
yeah I saw that on meta earlier
(congrats)
 
11:24 PM
Yeaaaaaaaaaaaah! — TheDoctor 10 secs ago
 
11:40 PM
done
0
A: Duct tape coding: new tag?

m.buettnerI like this concept. I think it has the potential to spawn some really interesting new challenges. So let it thrive for now and see where this is going. I also agree with Quincunx that we can wait for another one or two questions before making this a tag. Maybe this remains a one-timer after all,...

good day/night/whateveryourtimezoneis
 

« first day (1196 days earlier)      last day (3647 days later) »