« first day (1544 days earlier)      last day (3301 days later) » 

12:05 AM
I have an idea for an auto-golfing program. I would take a known solution and work in order of levenshtein distance, rather than starting from nothing and finding an ideal solution.
I feel like somebody would have already created something like that.
 
I came in here, saw the spinning loading circles, and thought my internet was having a problem. I was about to get up and check my modem before I realized one of them was starred. ಠ_ಠ
 
@AlexA. did you not find it weird that one spins the opposite direction and one spins faster? ;)
 
I did think that was weird. I just figured my internet was really stupid.
@MartinBüttner: Also, Cave In - Luminance
 
You're probably not wrong.
 
Well, I know my internet is stupid, I guess I just figured it was excessively stupid.
In unrelated news, it turns out that finding answers to APL questions is difficult.
 
12:17 AM
 
@trichoplax: Great, now there's a bug crawling all over my screen.
 
I really should superimpose that on a code block so it isn't so obvious
 
@AlexA. Hm, it's not bad but I'm not a huge fan of the vocals.
 
@MartinBüttner: I would probably consider Stephen Brodsky's vocals an acquired taste...
 
@AlexA. I know what you mean. I like La Dispute a lot (partly becaue of the vocals), and I think it's probably an objective statement that Jordan Dreyer simply can't sing.
 
12:21 AM
 
@PhiNotPi why?
then again...
 
Nobody's found the Easter egg yet.
 
@trichoplax: Haha I actually rubbed my screen because I thought that was just some crap on the screen.
@PhiNotPi: Found it.
 
@PhiNotPi I see 5 orange pixels
 
12:26 AM
Will someone please resize that Easter egg?
 
@AlexA. Sometimes simple things are more effective :)
 
@trichoplax oh wow
 
@MartinBüttner I'm guessing they aren't the Easter Egg though...
 
They were.
 
Yay!
 
12:28 AM
Happy Easter! You found eggs!
 
Stack Exchange should make an egg-themed minigame.
 
Oh hey, that sounds familiar. Yes.... familiar. ಠ_ಠ
 
I still have it as my PPCG thumbnail in Chrome.
 
@MartinBüttner I had never heard La Dispute and I agree with you about the vocals. Reminds me a bit of Steve Sneer from Kill Sadie and These Arms Are Snakes (the latter of which I like).
@PhiNotPi Really? Haha why?
Can you manually set thumbnails in Chrome or is that a coincidence?
 
Coincidence.
 
12:35 AM
@AlexA. I don't know either. But yeah La Dispute is something to get used to. But Somewhere at the Bottom of the River... is an amazing concept album, and his vocals fit the emotional lyrical content perfectly. Also, I'm generally a fan of vocals that go in and out of talking/singing/screaming like that.
I don't know what song you listened to but Said The King To The River is a great example.
that record is also quite mathy
 
Cool, I'll take a look. I can't remember what I listened to either, even though it was like 10 minutes ago or whatever.
@MartinBüttner These Arms Are Snakes - Deer Lodge
 
that's quite nice
 
@MartinBüttner Listened to Said the King to the River. I like the instrumentation but the vocals are kind of off-putting for me. I would probably have to let it grow on me. ;)
Previously I listened to something from Wildlife.
 
fair enough ;)
yeah I'm not a huge fan of their other records for some reason
 
Other than Somewhere etc. etc.?
 
12:46 AM
yeah
(and yeah it's a ridiculously long record title)
 
Fiona Apple has a longer album title.
I think she currently holds the record for longest album title.
 
lol that's huge
 
Right?
Glad you like These Arms Are Snakes!
 
if you're on facebook, Post-Rock Essentials are an amazing source of new music. their praise for bands is sometime slightly exaggerated, but they do share a lot of really good stuff.
 
Oh nice, I'll check it out! Also, this Solkyri song is badass. I love it!
 
12:52 AM
uhhh, which one does it start with? (that link should be the entire record)
 
Yes, I'm Breathing
Clearly I can't be bothered to scroll down, where the entire album is listed.
 
it's far away
 
Well the song that plays when you press the magic triangle is fantastic! I fully intend to give the rest of the album a thorough listening as well.
I haven't had the chance to start making my way through postrock.md, but I bookmarked it. :)
 
I would have been very impressed if you had :D
 
1:12 AM
Ugh, I should go do things. Nice chatting with you as always. Talk to you again soon.
Have a good (your time of day).
 
@MartinBüttner Time to rewatch faulty pancake algorithms :)
 
:D
eww, there can be duplicates in the input which have to be retained in the output. that's annoying
 
Hmm I don't think that's too bad...
Unless you were planning to use set or something
 
at the end I just wanted to flatten/union/sort the entire thing.
 
Ah, right
Does summing lists not work?
 
1:20 AM
sure, but it's gonna be longer
 
(@Sp3000 added the 2, 1 10 testcase)
 
:) you should also put 5, [1, 21, 46]
 
done
 
1:36 AM
I took my pancake approach and I have a horrible 187 :(
Python keyed sorting is expensive
 
I just did some intense golfing of task #2 from languages through the years.
(Yes, I know it wasn't a golfing competition.)
 
Don't worry, you weren't the only one :)
 
Element is now tied with CJam at 39 bytes for that particular task (ASCII art N).
 
@PhiNotPi q~:I,_ff{[\0I(](#)'NS?}N*
looks like @Optimizer didn't try very hard ;)
(that's 25)
 
Okay
shrugs shoulders
_'1[y~+y;[#1+3:"2:'%2<y~=|\ [#N]`"]\
`]
That's the best I can do.
 
1:50 AM
@Optimizer q~:I,_ff{[\0I(](&'NS?}N* 24, actually. feel free to use it.
 
I guess it's going to be a while before Element wins anything against CJam.
:/
 
2:37 AM
@Sp3000 did you see xnor's solution? yay for simplicity :D
 
Yeah I did :P That's why I didn't bother posting mine
 
3:00 AM
in Python can you split a string in code into multiple lines with single quotes?
 
I don't think so
 
print('O\
nce whe')

this gives the unexpected

O\
nce whe
 
yeah I think you need triple quotes
 
but why isn't this a syntax error then?
 
ah wait, you're saying it actually works?
 
3:04 AM
kinda, but the \ and \n is part of the string too
 
ah right, I think then you actually need to escape both of them. like \\\n (literally)
otherwise \<linebreak> just ignores the linebreak entirely
 
that's what I would want
 
I probably still don't understand which result you're seeing and which result you want to see.
 
I would want "Once whe", but I get what I pasted in
 
right, that's weird, because I do get Once whe, both in Python 2 and 3
 
3:08 AM
hmm, right, Pypy says otherwise
it is hard for me to reason the Pypy output
but in offical Python it works, thanks :)
 
seems like a pypy bug then?
 
maybe, I barely know Python though
 
3:34 AM
"A backslash does not continue a token except for string literals", apparently
Oh, further down it says that \newline is ignored
@randomra I don't think not having repeats is comparable to Sudoku, which seems to refer to a much stronger property to me. Better would be Latin square, but even that's still stronger.
 
0
A: Sandbox for Proposed Challenges

randomraSudoku programming code-challenge restricted-source source-layout You should write a program or function which outputs as much of Chapter I of The Little Prince as it can. Your program or function has to be an M-by-N block of code containing only printable ascii characters (codepoint from 32 to...

 
Also, your example has two /s on the last line?
 
@Sp3000 I see, I don't have an alternative title yet.
Opps, edited the example too.
 
3:50 AM
It's okay :) What I mean is, even if you can't think of a better title (not that I can either, admittedly), I'm just not sure that "just like in a Sudoku puzzle, i.e ..." is the best phrasing (you can probably just drop the rest of the sentence starting from the i.e.)
 
you mean drop the part before i.e., rigth? the " just like in a sudoku puzzle " part
 
Hmm if you drop that then the title might not make sense (surely there's a word for a matrix/grid where each row and column contains no duplicates...)
 
@NathanMerrill weren't you looking for something like this for one of your recent challenges?
ah no, yours was about wrapping diagonals, and this one is about rows and columns
 
5:03 AM
@MartinBüttner I was doing both diagonals and rows and columns
 
 
1 hour later…
6:16 AM
@MartinBüttner this is for ... ?
Also, I didn't try hard for .. ?
 
6:31 AM
I want answers!
 
@Optimizer I think "Languages through the years"
 
oh, true that then.
 
7:26 AM
@Optimizer "Your solution has to solve any example test case under a minute" does your solution obey this?
 
trying right now
which is the most heaviest ?
 
depends on your algorithm
15, 156 888 2015
 
will have to download latest cjam java version. I'll delete till then as I have to go out for a while.
but pretty sure 1 minute should be enough time .. lets see
 
(I will only test it if it's close on your machine.)
 
 
5 hours later…
12:19 PM
@randomra well, I was wrong, its nowhere near to 1 minute
 
12:36 PM
@Zgarb for the shift interpreter, so chain(f,g) puts on the stack a function h that calls f (which alters the stack in some way), then g (which alters the stack in some other way)?
if anyone else wants to chime in on that, since Z isn't active, feel free
 
12:54 PM
Apparently I was active on Yahoo! Answers 8 years ago
My answers are alarmingly terrible
 
maybe you just wanted to fit in
 
0
A: Sandbox for Proposed Challenges

PhiNotPiASCII Robot Wars This idea is based off of the game "Besieged" (which I've never played) and a previous sandboxed idea of mine called "Epic Customizable Tank Battle." The main idea is that your program is the AI that controls a robot equipped with various weapons. In this challenge, however, y...

 
1:15 PM
@Rainbolt Links or it didn't happen
 
@Optimizer what randomra said ;)
 
@sirpercival Yes, that's essentially what happens, but if g has larger arity than the length of f's output list, then applying chain(f,g) is undefined (because it has the same arity as f).
 
@Zgarb ah ok so do i need to put in an arity checker? or do i assume that all input strings are valid?
i'm doing a python one. it's super long. XD
so many decorators!
 
@sirpercival You can assume that the inputs are valid, and crash if they are not.
 
@Zgarb what i meant is: currently, if the stack is large enough to support g's arity, it will work, even if f/h's arity is smaller
(in my interpreter)
can i leave it that way, or do i need to make it crash?
 
1:27 PM
@sirpercival I think the meaning of "undefined" is, do whatever you want ;)
including making a trout jump out of your monitor of making demons fly out of your nose
 
@sirpercival That's perfectly fine.
 
@MartinBüttner stop giving away mah secrets
 
@Zgarb awesome
 
1:29 PM
:D
To be fair, those are terrible questions, too.
> I'm crazy to see my city from the heaven!?
 
@Geobits Terrible is a relative term. This was eight years ago after all.
 
would the following be an interesting code golf:;
 
Well, my standards may have changed in those 8 years, but that's just because SE has spoiled me.
 
given a list of points that are on the perimeter of a square, identify the rotation of the square
where 0 degrees = lines are vertical and horizontal
 
Is the list in any particular order?
 
1:33 PM
yes, and corners are not guaranteed
 
so , lets say that the list is from points along clockwise direction, where starting point is not guaranteed to be any specific thing
right ?
 
not necessarily clockwise
 
plain code-golf or some time restrictions ?
 
How many points does it take to define a square if the points aren't corners?
 
code-golf
 
1:34 PM
What about ambiguous cases? Like four corners of an axis aligned square could also be midpoints of a rotated one, right?
 
Correct me if I am wrong, but you'll have to provide points on at least three different lines in order to answer with certainty?
 
@NathanMerrill you have to fix on one of them, right ?
 
Nvm. Even three lines wouldn't be enough
 
@Geobits @Rainbolt there are certainly ambiguous cases, in which you should print some error message
 
Three lines, and one of them has to contain two points.
I think then you can answer with certainty
 
1:35 PM
well, actually giving a single line with three points is enough
 
I think you need at least five points.
 
because you only have to identify the rotation of the square
 
But you could be off by 180 degrees
 
How many points it takes to identify a square = Total number of stars to this message
 
(with only one line known)
 
1:36 PM
@Rainbolt maximum answer is 89.999999 degrees
 
90 degree rotations are identical
 
Oh I see
@NathanMerrill I think you have your answer. Challenge is interesting.
 
I do too :P
 
Here's some logic: two squares (edit: of different rotations) can intersect at 8 points, so those 8 points cannot define a square, so it takes 9 points to guarantee a correct answer.
 
@NathanMerrill hey i have a mostly-working king of the hill game, and i totally stole some of your code from the DDOS koth
so thanks for posting that
 
1:38 PM
@PhiNotPi Uh... two squares can intersect at infinitely many points and still be two different squares.
 
@sirpercival absolutely
 
@Rainbolt wut ?
 
I should say, two squares of different rotations.
 
Oh, right.
 
1:39 PM
I'm going to allow having multiple points on the same spot
 
if two squares intersect at infinite number of points, do points even exist ?
 
I'm not sure if this is a pun that I don't understand ^
 
If what I said is a pun, are our mouth even real ?
 
[0, 90) means not including 90, correct?
 
Sounds like an interesting question for math.SE "How many points does it take to uniquely specify the orientation of a square?"
@NathanMerrill yes
 
1:43 PM
You could also write it as [0,89.9999...)
 
I think Phi is right with 9.
 
well 3 collinear ones suffice
actually I think that might be the answer
because with 9 you've also got 3 collinear ones
 
well, we have to consider minimum points that cover all cases, so it has to be 9, I think
 
but then again you can probably place 7 or 8 such that only 1 square fits
 
If you can do it with 3, then obv you can do it with 7 or 8
 
1:44 PM
0
Q: Are control characters and ANSI escape codes allowed in output?

Martin BüttnerRecently, I had people go crazy with escape codes on one of my ASCII art challenges. This x86 machine code answer was able to save some bytes by throwing a few colour codes at the console. At some point along the way the OP even had a version which left the output blinking. Both the above, as w...

 
@Rainbolt I meant 7 or 8 where no 3 points are collinear
 
Oh ok
 
Well if I posted it on math.SE, I wouldn't ask for the absolute or necessary minimum, but rather the minimum in different situations.
 
@MartinBüttner any suggestions for how to handle the infinite loop in the shift interpreter for python? i hit a max recursion depth...
 
I think it turns out to be doable in 4 if they aren't collinear.
 
1:46 PM
@NathanMerrill Does that really make it more interesting? All that adds is filtering the input for unique points first.
 
@MartinBüttner ah, I didn't think about that. I was just hoping to make solutions not just count points
 
@sirpercival I haven't started working on any code.
@NathanMerrill Well it's not that simple.
 
@MartinBüttner no worries
 
If I wrote my next challenge in rhythm and rhyme would that be annoying?
It would keep me up at night, trying to come up with rhymes for bugs people found after the fact
 
@Rainbolt No, but I'll downvote if the test cases and tags don't fit in with that rhythm and rhyme.
 
1:49 PM
@Rainbolt Probably not, but you would get people who try to write the code the same way.
 
Can we have X more than three?
Let's just say that cannot be.
 
If it's obscure, they'll close it as unclear.
But you'll also get answers in Shakespeare.
 
Most normal input methods are fine;
hardcoding it in would cross the line.
 
Now I was really thinking of it,
The idea? Well, I love it!
But if, say, a small poem will do it,
This challenge has beaten you to it.
54
Q: Factorial in haiku!

TheDoctorTask Create a program that calculates the factorial of a number using no built-in factorial functions. Easy? The catch is that you must write your entire program (including testing it) in haiku form. You can use as many haikus as you need, but when pronounced, they must follow the 5-7-5 syllab...

 
Thank you Geobits for ending the rhymes. (Edit: I see you've removed the line)
 
1:57 PM
Removing the line?
Well, that's all fine.
And now I continue this rhyme of mine.
 
I didn't want to break the rhythm; and that line didn't fit well with'em.
 
It didn't fit well? Of course, take it out!
We can't have non-rhyming lines about!
 
yes
 
@Optimizer My, no rhymes? What a mess!
 
and no and no and yes
orange
 
1:59 PM
Llamas aren't the rhyming sort. His plan's to make us all abort.
 

« first day (1544 days earlier)      last day (3301 days later) »