« first day (1598 days earlier)      last day (3240 days later) » 

7:15 AM
Hi everyone
 
7:35 AM
@Katenkyo , Hi
 
@CoolGuy Has a nice week end?
 
Yea
 
 
1 hour later…
9:05 AM
0
A: Sandbox for Proposed Challenges

Martin BüttnerRefined Partitions code-golfarray-manipulationset-partitions Consider an array of integers: [1, 0, 9, 1, 3, 8] There are a lot of ways to partition this list into consecutive sublists. Here are three: A: [[1, 0, 9], [1, 3, 8]] B: [[1], [0, 9], [1, 3], [8]] C: [[1, 0], [9, 1], [3, 8]] We w...

 
@NewSandboxedPosts Interesting
I guess the bets way to procede is to determine if Y and X contains the same suite of numbers, everything else is in fact useless. The biggest part will be parsing the input
 
9:30 AM
@Katenkyo parsing the input should be trivial, because you can take your native array format (or even an actual array literal). however, just checking the numbers contained isn't the full story as the A <-> C example shows.
 
Ho, yeah, I was only taking into account the first list, not the ABCs
 
@MartinBüttner so we only check if the second one is refinement of first ?
 
Nevermind what I said so
 
@Optimizer yes, what else?
 
vice versa ?
 
9:34 AM
if they both were refinements of each other, they'd be identical
 
I meant, it can be that first is refinement of second
do we output true or false ?
 
ok
 
(there is a test case for that)
 
didn't see the update yet
 
9:40 AM
I'll probably post this later today. It's nice and simple for a start-of-the-week-golf, and I probably won't have much time over the next three days.
 
last week was a bad week for me .. got ill, lost my streak ..
 
I take it the latter was the worse experience for you :P
 
totally
 
I'll probably lose my streak next month... I'm already mentally preparing myself for the loss ;)
 
10:09 AM
-1
Q: Print Input Sideways

Kody KingYour program must take an alpha/numeric input string and print it sideways (rotated 90 degrees clockwise). So for example, Hi, would become: ######### # ######### ####### # Rules: The shortest code which produces legible output wins.

 
10:51 AM
@NewSandboxedPosts Sounds like another case of There'sAMathematicaBuiltinForThisButTheNameIsTooLong
@MartinBüttner Omitting delimiters means [[109][138]] is okay, right?
 
Guys, there's already a challenge about creating nonograms
 
Oh... actually you might want [[1 0 9]] [[1 0 9] [1 3 8]] (the other way around) to also be a test case
 
Would it be wrong if I post a challenge for solving nonograms?
 
11:08 AM
Creating's different from solving so as long as there's no challenge about solving nonograms I think it should be fine
 
didn't find anything for the words nonogram,hanjie or picross except this one. So i'll start working on it :)
 
@Sp3000 yes
@Sp3000 will add that
 
k :) (if it wasn't for the "You must not assume that X and Y are partitions of the same list" part I think it could be 31 in Mathematica)
 
whoa, 31?
I can do 40
 
LongestCommonSequence[#,#2]==#& works for the test cases except for the one I posted above, assuming input are strings of the form [[109][138]]
Still trying to convince myself it actually does the right thing though
 
11:22 AM
that seems to give False for some test cases
actually, I think it only works if #==#2
 
Did you test like f["[[109][138]]", "[[1][09][13][8]]"]? It doesn't work if there's spaces
 
wait .. [[109][138]] [[1][09][13][8]] is true ?
 
Basically I thought refinements would just have extra ][s in their string reps
 
11:26 AM
then this makes it more of a string-man than array-man
 
@Sp3000 I don't really understand why this works though
 
Why what works?
 
[[10 9][138]] [[1][0 9][13][8]] - true ?
 
@Sp3000 why LongestCommonSubsequence gives #
@Optimizer why would you include spaces in your input?
 
because these are two numbers in an array, 10 and 9 ?
and 0 and 9
 
11:28 AM
@Optimizer numbers can't be greater than 9
 
Well assuming they are from the same list, then you just need the second to be the first with extra ][, which implies that the first is a subsequence of the second
Since the longest common subsequence can only be as long as the first string, it must be the longest
 
but will that work under 5 seconds ?
 
(but [[109]] [[109][138]] breaks it due to extra elements, if we don't have the assumption)
 
ohhhh, subsequence, right
 
Yeah, Mathematica uses LongestCommonSubsequence for... longest common substrings :/
Saves 3 bytes though, if only this worked :P
 
11:29 AM
Mathematica's naming is a bit weird
@Sp3000 that's fine to assume though, I think
 
But you said "You must not assume that X and Y are partitions of the same list" :P
(I meant f["[[109]]","[[109][138]]"], which would incorrectly return True)
Hmm how do you even replace "][" with "" in CJam. This sounds like it'd be 10 bytes or something
 
"]["/:+
 
Ah, I keep forgetting that CJam replace is just split and rejoin :P
 
"]["- ?
that should serve the purpose here
 
Hmm depends how you take the input format I guess
I was thinking ll"]["/:+=
 
11:40 AM
well, that is not the final solution, is it ? :P
 
Isn't it?
 
[[123][456]] [[12][34][56]]
vs
[[123][456]] [[12][3][4][56]]
 
... oh right, oops
 
-__-
also, it will never ever give 1, as the first l will also have ][ somewhere
 
Now I remember why I was looking for a language with builtin common subsequence, forgot that just removing ][ doesn't work
But yeah I guess checking the other condition (that they originate from the same list of digits) is too long, so I suppose that idea's out
 
11:52 AM
37
scratch that. incorrect algorithm.
 
61 in Mathematica
 
this is actually longer if you try via string man.
 
That's why I said "If it wasn't for..." :P
 
I might have a 25 byte solution. array. man.
 
You're comparing with CJam though :/
 
12:03 PM
I think I'll post it now, unless you guys can see anything that needs fixing
 
Can I get one more test case in? (give me a sec)
 
Yeah I don't think you have [[1 0 9] [1 3 8]] [[1 0 9 1 3 8]] yet, just for good measure
 
false
yes, the algorithm works!
 
@Sp3000 added. let me know if you come across any other important ones.
 
12:09 PM
I think the main bases have been covered, but we'll see :)
 
@Sp3000 you have something ?
 
Nope, was just testing different approaches to see what people might mess up on
 
posted
 
(sometimes I think Optimizer posts a little too fast)
 
(too obvious?)
I was afraid Martin would post his first :P
 
12:12 PM
1
Q: Refined Partitions

Martin BüttnerConsider an array of integers: [1, 0, 9, 1, 3, 8] There are a lot of ways to partition this list into consecutive sublists. Here are three: A: [[1, 0, 9], [1, 3, 8]] B: [[1], [0, 9], [1, 3], [8]] C: [[1, 0], [9, 1], [3, 8]] We will call a partition Y and refinement of another partition X if...

 
Er... 2 min?
 
in my defense, the code got completed only after the program was posted
 
@Optimizer I won't post my Mathematica solution for at least two days I think
oh wow, xkcd page numbering is getting gold
 
how did that get that popular ?
 
I have no clue
 
12:21 PM
That's what I'd like to ask too :P
 
maybe "xkcd" in the title and 4 days of HNQ
I was also really surprised by CH's 7-slash display getting over 70 upvotes (and the answers being nowhere near that)
 
CH and magic go along. Not you.
 
CH seems to get mentioned a lot in chat. Maybe we should golf the initials down to the greek letter X
 
khi?
 
12:29 PM
A bit of a silly test case, but [[1 0] [1 0] [1 0]] vs [[1 0 1] [0] [1 0]] if you want to test examples where the same list is in both
 
@Katenkyo often spelt chi
 
@Sp3000 I think I've got that covered with the large test cases, but I'm not entirely sure
 
@trichoplax Easier to remember it as Khi, because it is write as it sounds ^^'
it is sooooooo long to parse a 30x30 nonogram to a list of array by hand >_<
 
@Katenkyo Indeed. I sometimes suspect English is deliberately confusing
 
@trichoplax Not only english, in french too we pronounce it "she" if it is written chi, and our language is more impacted by latin/greek than english ^^'
 
12:41 PM
Hmm I wonder in Martin's 61 is like what I've got right now... but I can't seem to golf the list indexing down :/
 
I am sure cjam can go to 20
waits for Dennis or Jimmy
 
30 columns of a nonagram : [[[10,6,9][4,9,8][5,11,7][7,14,7][7,15,5][8,17,5][9,18,4][25,4][4,2,9,1,3][4,1,8‌​,3][3,8,2][3,10,2][3,9,2][3,16,2][3,16,2][3,17,2][3,16,2][3,16,2][3,9,2][3,10,2][‌​3,8,2][4,2,8,2,3][4,19,3][4,18,3][5,17,4][5,16,4][5,14,5][6,12,6][7,9,6][8,7,8]]
 
Probably. Lemme see if I can figure out recursion in CJam...
 
Started to be hard to not make mistakes :/
 
12:53 PM
10 bytes in Pyth
But doesn't work for the large test-cases :-(
 
doesnt work at all ?
 
I should rename my challenges "xkcd math"
 
Well, too slow.
And also has memory problems.
 
you simply generate all possible sublists, don't you ?
 
Yeah.
 
1:05 PM
@Sp3000 59 now
@Jakube ]:)
 
0
Q: Deciding if x-y is 0 or negative without using conditionals

max0005So, My friend has come up with this interestng question, that is, given two numbers x and y deciding whether their difference: x-yis 0 or less without using any conditionals. The code should return 0 if they are 0 or less, or the result of the subtraction otherwise. My attempt (In Pseudo-Java):...

 
Hello :)
 
@Vioz- Hi
 
What's up?
 
Nothing special, everything is fine. And you ?
 
1:16 PM
People are actually at work today, so that's good
 
Haha, so what was the reason you were alone?
(I don't remember if you said it when I was here friday)
 
One guy didn't come in, and the other called in sick, so my boss just never came
He sent me a text about 3 hours in saying I could just go home
 
Nooooo every time I talk to a tester my name shows up in an email "Our current testing status is blah blah. [...] Rainbolt said not to write a bug on these three items:"
Then I get a desk visit from operations manager
 
then they reply to the mail saying "Rainbolt still emphasizes on not filing the three bug reports"
 
I'm just going to let them write all the bugs. Then I'm going to close them as duplicates of open tasks.
Then I will find the first opportunity to mention gently that they are wasting a lot of time.
 
1:40 PM
@Vioz- So it wasn't because of a sacrifice :'(
 
2:03 PM
No, it was not :P
 
2:26 PM
@Sp3000 any luck with golfing Mathematica?
 
I was doing it in Python but it's very pattern matchy and could be ported
But with the amount of indexing I do it looks a lot more than 59
 
I don't think I'm doing any indexing at all
and I've still got some annoying duplication I might be able to get rid of somehow
 
I keep dropping the first element of the second list, so I keep accessing the first elements of both
 
ah okay. I've got a completely different approach then
 
Probably better :P
 
2:31 PM
let me know if you want a hint :P
 
Once I'm happy with my Python one :P
 
hm, first attempt for porting the Mathematica solution to CJam is 27
oh and it doesn't even work for empty lists
 
:/
 
3:02 PM
0
Q: Stacking Pythagorean Triangles

BrainSteelBackground A Pythagorean triangle is a right triangle where each side length is an integer (that is, the side lengths form a Pythagorean triple): Using the sides of this triangle, we can attach two more non-congruent Pythagorean triangles as follows: We can continue with this pattern as we...

 
@Jakube I'm not entirely sure why you included the 10-byte version in your post, since it's the approach I've explicitly disallowed in the spec.
 
3:31 PM
0
Q: Simple Pazaak (Star Wars)

phantomjediPazaak is a card game from the Star Wars universe. It is similar to BlackJack, with two players pitted against each other trying to reach a total of twenty without going over. Each player has a "side deck" of four cards of their own that they can use to modify their score. Mechanics Gameplay is...

 
4:13 PM
@Dennis not trying refined partitions ?
 
4:24 PM
hi @MitchSchwartz
 
@Optimizer: Didn't have time to look at it yet.
 
4:57 PM
I suppose PPCG has already heard about TIS-100?
It's crazy addictive, at least to me, and I imagine it'd appeal to people who enjoy codegolf
 
@Calvin'sHobbies I haven't managed to finish the BBBF player and the 3 weeks are up. Once the winner is announced I'll keep working on it and post it as a late entry just for interest. It's a lot less straightforward than I expected :)
 
Jun 5 at 18:33, by Martin Büttner
ohhh, just got the Zachtronic Industries email newsletter. "TIS-100 is the assembly language programming game you never asked for!" I'm sold!
also this:
2 days ago, by Martin Büttner
In other news, there's yet another upcoming assembly-based game: http://tomorrowcorporation.com/humanresourcemachine
 
Ah :p
Oh, from the World of Goo developers.. that game was fun, too
 
I managed to solve sequence counter in 176 cycles, pretty happy about that one
What good scores have you gotten?
 
5:12 PM
0
Q: Bitstring Physics

ZgarbBackground Yes, bitstring physics is a real thing. The idea is to construct a new theory of physics using only strings of bits that evolve under a probabilistic rule... or something. Despite reading a couple of papers about it, I'm still pretty confused. However, the bitstring universe makes for...

 
I haven't started playing it yet. wanna wait until it leaves early access
 
Ah
I got tempted from seeing people fight over scores, and decided to join in
 
5:26 PM
@MartinBüttner I simply wanted to show off the ./ method. (I added this method to Pyth, and sadly wasn't able to use it since then).
But your right. I removed the (invalid) second answer and also added some extended explanations.
 
@FireFly I think if I played it now I won't replay it once it's finished
 
@MartinBüttner Ah, I see. I think I'm safe when it comes to that; I tend to replay games I enjoy a lot, so if it's gotten improved, that gives me even more reason to do so
 
5:48 PM
We've got programming puzzle games, programming survival-like games in an RTS style... It seems to me the next obvious choice is a racing programming game.
 
you mean like the two vector racing challenges or lab rat race?
 
I do mean like that! Perhaps at a larger scale.
 
well a racing KotH would be interesting for sure
we originally considered making Lab Rat Race like that, but then figured to simplify it to a single-player game in favour of focussing on the genetic algorithms
 
6:09 PM
@Lembik hi
 
The Pyth war is on!
 
i don't have any good ideas for j >= 2 so far, but i think i'll play with speeding up j = 1 a bit more
 
xbox 360 games playable natively on xbox one
 
Maybe now Sony will up their game. PS Now might have seemed okay before (not really), but it's ridiculous compared to backward compatibility.
 
didn't Sony announce backward compat last E3 itself ?
 
6:20 PM
Didn't really follow last year, didn't have a PS4 then. But I know that as of now, they don't have any option except through the PS Now (subscription) service, which is really really limited.
I can almost understand if they're just saying it's hard to emulate a PS3 (it probably is), but they're not fooling anyone by not allowing the old PS1/2 classics from the store on PS4.
 
6:32 PM
@MitchSchwartz but it's too fast for words already :)
@MitchSchwartz I know this isn't your main interest, but do you think the j=2 method can be parallelised?
I mean your j = 2 solution
@MitchSchwartz emailing Min_25 was so successful I am (perhaps not sensibly) wondering whether other world beating mathematical puzzlers might be interested too
grechnik?
 
In CJam, what is the comment character?
 
@NathanMerrill e#
(as of 0.6.5)
 
why e?
 
because comments aren't worth wasting a precious single-character command on
and two-character commands are prefixed by e or m, but m is for math-related ones
 
true, but of all of the letters, why e?
 
6:42 PM
"extended" commands
 
@MartinBüttner Maybe I should suggest the Joy of X as an Euler problem? :)
 
I don't know? I'm not even aware of a process for users to suggest new PE problems.
 
7:02 PM
@MartinBüttner I found this page forum.projecteuler.net/viewtopic.php?f=4&t=510
I don't know anything about it except what it says there
the post is 8 years old so who knows
@MitchSchwartz do you know how to suggest new project Euler questions by any chance? Is that pagehttp://forum.projecteuler.net/viewtopic.php?f=4&t=510 still correct?
 
as i recall, you're not supposed to suggest a problem unless you have solved it and not published it anywhere else?
 
It used to be possible to suggest PE problems. I did it once, but my problem was ruled too easy for Mathematica users.
 
@PeterTaylor do you remember what it was?
 
It was something to do with Galois fields and roots of quintics.
 
What problem isn't too easy for Mathematica users?
 
7:13 PM
On an unrelated note, I've finally tracked down all the references for the various tiling symmetries. Now to tackle some diagrams.
 
@MitchSchwartz that makes sense
@PeterTaylor I love that a problem to do with Galois fields and roots of quintics is now regarded as too easy for an amateur coding competition :)
how things change
hi @feersum
@feersum thanks very much for your really nice and clear explanation of Min_25's solution
 
Hi
 
@feersum just because I like to chat about it... I did try and extend the idea to j = 2. My conclusion that it was doable by someone who could cope with math complications better than me but it would never extend usefully to j > 2
@feersum do you have a view on this?
 
@Lembik I don't see why j=2 should be special compared to j=3. My guess is that the amount of calculations required grows exponentially with j
 
7:26 PM
@feersum it's not so much that j = 2 is special but more that the complications get worse to get a closed form solution and I suspect are unbearable for j>2. When you say the amount of calculations, do you mean the calculations that your code would have to do or that you would have to do to come up with a closed form?
 
the code
 
ah ok
I was still hoping for a closed form solution!
even if the amount of computation doubles I suppose j = 3 is still possibly doable
 
7:52 PM
@Lembik j = 1 can be a lot faster
 
8:05 PM
@MitchSchwartz go on...
have you got a new formula?
 
@Lembik added faster version
no, same formula
 
i put it at the bottom of the answer
 
ah yes :)
it would be nice if you could add a little text explaining what you did
 
what's not clear from looking at the code itself?
 
8:11 PM
2188 ! :)
that is a lot faster
2761 :)
the printing was slowing it down
that's very impressive
do you think it's possible to get j = 2 up to 200 say?
it crashes my computer as it runs out of RAM
 
wait, so the previous scores were including the printing, but for this one you disabled printing? or did i misunderstand
 
@MitchSchwartz the previous scores the printing didn't make much difference
but now it is printing huge numbers
I just run it with |tail
so it is still printing, just not scrolling
 
oh, got it
 
It would be great to get my harder puzzles in front of grechnik etc but that seems very hard to do
I think they might enjoy them!
I suppose I could ask Min_25 but it just seems all a little pushy and I don't want to annoy anyone
 
Let's say I was working on a golfing language, how 'developed' should it be before I start using it for questions? I'm sort of just adding stuff on the go, but I don't want it to seem like I'm just creating functions per the challenges requirements and putting them in short-form.
I imagine linking the Github page to keep track of changes would be sufficient, but I'm not sure.
 
8:29 PM
@Lembik i think you accidentally added an "n" tag to the question?
 
8:50 PM
@MitchSchwartz thanks! :)
@MitchSchwartz do you know what your ranking on project euler is out of interest?
@Min_25 Do you think any of the other top Eulerians (or equivalent mathematical puzzle solvers) would be interested in the Joy of X or my other other puzzle for j>2 ?
@Min_25 At least they couldn't claim they were too trivial :)
 
@Lembik How do you come up with these particular problems and what makes you so interested in them?
 
@Lembik eulerian rank is based on most recently published problems, so you have to be very active on the site to get / keep a high rank
my peak rank was 5th
that was 5 years ago
i haven't solved a problem there for a very long time
 
9:06 PM
0
Q: Make a creative image upsampler

Joe Z.People who have seen me around a lot know that my profile picture has been a greyscale mosaic of the AACS key for the longest time now. Sometimes when I look at it, it looks like it should be a blurred (or just severely downsampled) image of something else as seen when shrunk down to a 4x4 image,...

 
9:21 PM
rank computation changed a bit; when i got 5th it was based on how many of the most recent 25 problems you solved; then there was another calculation based on making the fastest 20 solvers, and my peak with that is 12th; i suppose now they go based on fastest 100 but i haven't kept up
so, they store the fastest 100, but eulerian is based on fastest 50 of the 10 most recent problems; and btw Min_25 is 1st again
 
9:47 PM
I know it's a common opinion here that challenges should not simply be art contests with a bit of programming involved, but I'm having a hard time trying to find an "official stance" on meta of what such challenges should entail. The above question, by the definition on the tag, is a perfectly valid challenge.
 
Think of it this way: Close votes are to popularity contests as downvotes are to every other challenge.
Because pop contests are primarily opinion based, so too are their close votes.
 
17
Q: The line between art and programming

trichoplaxI really like this question but I know there are complaints of it being more about art than programming. I'd like to hear both sides of this and see what is in the best interest of the community and the site's public beta. I'm posting my own view as an answer so that the votes will show where it...

 
10:03 PM
Just got a bug along the lines of "When I export a file to .csv it looks extremely crowded and column widths need to be modified to make it look presentable."
I walked over and said "Try opening it in Notepad. Now try fiddling with the columns."
Then I had to dig myself out of that hole immediately afterwards
 
 
1 hour later…
11:24 PM
0
A: Sandbox for Proposed Challenges

trichoplax5-a-side Toroidal Bot Soccer king-of-the-hill Randomised teams This is a team game. Rather than being assigned permanently to one team or the other, each bot will play in a number of games, each for a different randomly composed team, and score a point for each game in which its team wins. Thi...

 
@MartinBüttner From trichoplax's (top voted) answer, "...we are in danger of closing questions based on how much art is in them, rather than based on how much programming is in them." I feel like Joe's challenge requires adequate programming skill. It is not "just draw a pretty picture," to my mind. I will concede, however, that it is too broad as it stands.
 

« first day (1598 days earlier)      last day (3240 days later) »