« first day (1966 days earlier)      last day (3067 days later) » 

12:10 AM
literally want to strangle past self for deciding to use JS for embedding
Does Lua for Java even exist? Maybe it's not too late
@Upgoat You should check out Powershell then.
FiveWordCamelCaseNames for shell commands.
@Upgoat was going to say Cheddar website was 10/10 on point then relized u used a template.... I mean the site is 10/10 still just... you know...
you just don't get Bald Bantha's ten score. that ten score goes to the template designers
12:37 AM
man chat is slow I am gone for like 1/2 an hour and no new messages...
Slow? Back in my day, 10 messages would disappear from chat per hour, and we'd all have to each chip in a comment just to break even.
3 days later...
"Did anybody do anything good on the weekend? Feels like it went by in a flash"
-3 hours later...
Where's my next comic @Sp3000?
12:44 AM
You still remember that? D:
TIL: If you have a custom domain for one GitHub repo, you can use the apex domain instead of "user.github.io" for every other repo. E.g. 255.wf also works for 255.wf/FormulaCompiler
@quartata whar you go
@BaldBantha template was only layout. I did background, colors, buttons, titles, everything
GitHub finally implemented what I wanted:
May 31 at 18:57, by mınxomaτ
I'd like more customizability on the profile page.
@Upgoat ahh
Can anyone explain to me where the "lena" picture came from?
12:57 AM
@mınxomaτ two things: 1) your SO profile link in the footer doesn't work, it's your UserID for PPCG, not SO, 2) the link on your SO profile to your website should be updated from minxomat.github.io to 255.wf (even though it redirects)
I know. But I'm lazy
Lenna or Lena is the name given to a standard test image widely used in the field of image processing since 1973. It is a picture of Lena Söderberg, shot by photographer Dwight Hooker, cropped from the centerfold of the November 1972 issue of Playboy magazine. The spelling "Lenna" comes from the anglicisation used in the original Playboy article. == History == Before Lenna, the first use of a Playboy magazine image to illustrate image processing algorithms was in 1961. Lawrence G. Roberts used two cropped 6-bit grayscale facsimile scanned images from Playboy's July 1960 issue featuring Playmate...
@mınxomaτ screams internally... okay
@Sp3000 Gotcha covered ^
0/10 WTH is Calvin?
1:04 AM
I dunno. I didn't name the comic strip.
I figured changing room to yard was pushing it, and I didn't want to lose my guest artist privileges.
@calvinsho... @HelkaHomba What's your all time favorite C&H strip?
Cyanide and Happiness?
Calvin and Hobbies?
Calvin and Homba.
1:08 AM
@Bigtoes wat?
user image
4
@HelkaHomba
Awww :(
My feels. They hurt.
Yard still works though :P
Actually it's part of the secret plan to make a hi-res texture mod and then make millions from 4K Minecraft videos on Youtube.
1:11 AM
@DrGreenEggsandIronMan That's an impossible choice. Off the top of my head gocomics.com/calvinandhobbes/1986/09/13#mutable_51225 is always funny
Calvin and Homba:
user image
5
Isn't that Helka and Hobbes?
Oh shit, you're right. I'm backwards.
Ok, this is weird
When was TNB not weird?
2
Petition to un-pin the "Graduation Perks update" from the starboard because it's been there for 10 days.
12
And takes up way too much space. I wasn't kidding, I literally only see 2 actually starred messages max.
Pinned messages automatically unpin eventually.
But I wouldn't mind unpinning it.
Except that I'm not gonna do it since I'm not a PPCG mod. :P
@mınxomaτ Wait really? I see 8
same ^^
1:20 AM
@mınxomaτ Yeah, I have that issue too.
1366x768 is still a thing
Hmm. Does Cyanide and Happiness onebox? Let's find out.
No.
there's userscripts to hide roomlist, userlist and roominfo
@mınxomaτ Ey that's my resolution! Laptops ftw. :P
@Bigtoes That's probably a really, really bad idea.
1:21 AM
Yeah Cyanide and happiness can be pretty bad sometimes
Probably... not all of them are terrible though.
But many are NSWF as fck
Definitely
1:22 AM
they did a card game on Kickstarter
they started shipping last week \o/
@Quill Weren't they also the one that sold literal crap? Or was that The Oatmeal?
@PhiNotPi the amount of stars might just keep it there...
@El'endiaStarman I can't remember either
1:26 AM
@LeakyNun I haven't. Your turn.
Just casually throws out a quarter
@feersum the 4k option isn't even available yet :/ If it takes 24+ hours to process it may not be worth it
@HelkaHomba You can't really stop the processing though, can you?
@EᴀsᴛᴇʀʟʏIʀᴋ hello i have quetions how does juice an avocado i have try for thirtee minut and no juic
@El'endiaStarman No (apart from deleting the video and reuploading in 1080p)
1:41 AM
@mınxomaτ 10/10 weirdest yet most accurate representation of PPCG culture
@Upgoat Already preparing Vol. 2
1:47 AM
beautiful
:D Doorknob has as bad typing as me
nope, just a one off
btw I'm prepping v0.1-alpha, anyone wanna test it?
1:50 AM
sure
for cheddar?
"Paw and Order" would've been a better title for Zootopia.
@Upgoat sure, why not
@Upgoat you got a tests subfolder?
@Quill cool thanks :D, though you'll have to do a manual build. git clone -b version-0.1a https://github.com/cheddar-lang/Cheddar.git && npm install && grunt build && node ./dist/cli/cheddar.js
@Quill yeah, but I haven't figured out how to make the tests
I'll need to check if STDOUT is a speciifc string, idk how to do that
can I write the unit tests on your workspace, or do you want me to make my own?
1:55 AM
Do Youtube links that are still uploading onebox?
ok then
How about Youtube links that aren't anything?
@Dennis eadem mutata resurgo
Will mine update when it is uploaded and processed?
it won't update when it's uploaded
you can edit the message which updates the onebox though
@Quill I'm trying to git properly so you'll have to make a separate unit test branch, so you could do it locally or in zyabin101's c9
1:59 AM
@feersum Inspired by your URL: "Ten tenants' ten tents."
@HelkaHomba apparently that's private?
@Quill He is, unfortunately for him, a mere mortal, and as such is now unable to edit his message.
@El'endiaStarman More of an FYI, but yeah
I was gonna offer to edit it, but he can always just post a new link
Yeah. If you (or someone else) edit his old message, hardly anyone will notice.
http://i.imgur.com/FKk5NTX.gif
10/10 epic gif
2:05 AM
Someone give me two random arrays for test IO for next challenge.
Different topic: what do you guys think of the idea of having a dictionary that can have multiple keys for a value?
@DrGreenEggsandIronMan [38, 73, 68, 16, 2, 65, 69, 28, 66, 52] [41, 16, 7, 51, 63, 96, 28, 88, 71, 95]
@El'endiaStarman where's the usecase for that?
>>> from random import randint as r
>>> [r(0,99) for k in range(10)]
[75, 92, 73, 84, 26, 20, 10, 94, 53, 12]
>>> [r(0,99) for k in range(10)]
[41, 33, 53, 36, 44, 61, 98, 41, 6, 25]
@Maltysen Sorry, forgot to metion, they have to be the same length.
I can still use that though
2:07 AM
there
@EᴀsᴛᴇʀʟʏIʀᴋ Try now. I hadn't officially published it yet. (It's still uploading mind you.)
hey it auto updated
13 mins ago, by Helka Homba
no edit
@Quill A number of situations where a many-to-one relationship in databases is useful could be duplicated with such a many-keys structure.
@DrGreenEggsandIronMan Here. You can specify the parameters.
2:10 AM
@EᴀsᴛᴇʀʟʏIʀᴋ Onebox looks the same to me.
@El'endiaStarman It's called a hash.
Refreshed and now it's updated.
@Quill @DrGreenEggsandIronMan @EᴀsᴛᴇʀʟʏIʀᴋ @CᴏɴᴏʀO'Bʀɪᴇɴ @MᴀʀsUʟᴛᴏʀ I setup a Cheddar workspace for testing/playing around with/writing unit tests. Idk how to invite so if you'd like to help beta test Cheddar before I do the initial release, I'd appreciate that :3
@LeakyNun But most things hash to different things.
2:11 AM
Conor went to bed btw.
@El'endiaStarman Oh, right.
@Upgoat ;-; I just did the same thing
oh :|
@El'endiaStarman I guess so
Mini challenge: Given two ints a, b, output max(a, b, a+b, a*b)
2:21 AM
testcase?
@HelkaHomba Julia, 1 byte: *
@HelkaHomba eS+[sQ*F
eS+Q,sQ*F
lambda a,b:max(a,b,a+b,a*b)
2:22 AM
@EᴀsᴛᴇʀʟʏIʀᴋ wait what?
@Maltysen oops, ninja'd
eS+,sQ*F (8 bytes)
a=27,b=-1 -> 27
a=-7,b=-5 -> 35
a=-7,b=5 -> 5
a=0,b=1 -> 1
@HelkaHomba I know a->b->max(a,b,a+b,a*b) works.
@HelkaHomba Minkolang, 13 bytes: nn3$D*r+j$ZN.
@Lynn probly the simplest way :P
@EᴀsᴛᴇʀʟʏIʀᴋ even with only positives 1*2 < 1+2
Pyke: 7 bytes: BQs]+Se Try it here!
@El'@Leak are your solutions doing anything radically different than what Lynn did?
2:29 AM
Not really.
@HelkaHomba Nope.
just that I treated the input as a list
Half the bytes in mine come from stack manipulation.
so that would be equivalent to lambda Q:max([sum(Q),prod(Q)]+Q)
Jelly, 6 bytes: +,×;,Ṁ
same principle.
Input as list, 5 bytes: ;S;PṀ
2:33 AM
10/10
as dyad: [:>./+,*,,
as list: [:>./[,+/,*/
@Upgoat fixed
> fixed
Can sorting and traversing a list with only 0s and 1s be done in less than O(log n) time? I think so.
@El'endiaStarman O(log n)? How do you even know what the list is without going through every item?
2:46 AM
Oh, I meant O(n log n).
Counting sort can do that in O(n) time.
Sorting any list can be done in O(n log n) time
Well, okay, clarification needed: I actually have a list of objects, which are one thing or another. Order needs to be preserved, but the sorting key is binary.
I'm using Python. I'm just wondering if there's a more efficient method than .sort() and then traverse. Not that it's bad at all, I'm just curious.
@El'endiaStarman Go through every item. If they are 0, append them to list 1; if they are 1, append them to list 2; then concatenate the lists.
2:49 AM
That'd be O(2n). (Yes, technically, O(2n) == O(n), but I'm ignoring that for the moment.) Can it be done in less time? I doubt it.
O(n) is the limit, because you need to go through every item.
You can do it in place using a modified bubble-sort for O(n) time and O(1) extra memory.
Insertion sort, but ignore all the ones.
Wouldn't that basically be O(2n) in the worst case?
Wait, you're talking about performance in the ppcg chat? Now I understand what Doorknob meant with the signal:noise ratio...
@El'endiaStarman I think I've seen this called an AB sort before. Somewhere...
2:52 AM
@LeakyNun Hmm. How would that work? (I can't recall how bubble sort works at the moment and I don't want to read the Wikipedia page. :P)
signal: good, interesting conversation. noise: memes, avacado jokes, carets, ascii pictures
@Quill It was a joke quill, don't take it seriously
@El'endiaStarman Bubble sort is an O(n^2) algorithm. You go through the list n time, while swapping neighbouring items that have their order reversed.
And also, avocad, not avacado
@Bálint ;-; while(quill.doesntGetIt()) mug.fill('coffee')
2:54 AM
And carrot.avocad, if you really want see an angry Doorknob
I have an interesting function
@Quill You should get rid of the extraneous punctuation at the beginning of that line.
(review complete)
f = lambda n: abs(n - int(str(n)[::-1]))
@Bigtoes When did you change your name again?
Did I?
2:55 AM
0
Q: "Bit-borrow" two two arrays

Dr Green Eggs and Iron ManDid you know that a small number can borrow bits from a larger number? Here's an example. Let's say our two numbers 5 and 14. First, write them out in binary: 5 14 000101 001110 First we take the smallest on bit away from the larger number, and we give it to the smaller number. So Thi...

Umm, three or four days ago I think.
@LeakyNun Few days ago
@orlp not interesting.
@LeakyNun why?
@orlp Just something like the 1089 thing
Old.
2:56 AM
@LeakyNun what?
so how is it not interesting?
I gtg eat dinner, bye!
which arguments form cycles?
which go to 0?
why?
@orlp palindromes?
2:58 AM
f(45) -> f(9) -> 0
f(219978) -> f(659934) -> 219978
@orlp interesting.
I wouldn't call that old & boring
oh so now we agree!
@orlp yep.
since f(219978) grew, are there any numbers that grow without bound?
@orlp No, because the value of the function is bounded by the number of digits of the argument.
@orlp: Have you done any brute-forcing on this?
@El'endiaStarman no
so many questions
so few answers
@LeakyNun I guess that's true
so we know that each number must at some point cycle
(if we count 0 -> 0 a cycle)
@orlp yes
can a cycle length be bigger than 2?
@orlp why not?
18 -> 63 -> 27 -> 45 -> 9 -> 0
3:03 AM
that has cycle lenght 1
0 -> 0 -> 0
Oh, that wasn't an answer to the question
ah
That was just a proof that the number would not be 2-digit
>>> def steps_before_cycle(n):
...     seen = [n]
...     while f(n) not in seen:
...         n = f(n)
...         seen.append(n)
...     i = seen.index(f(n))
...     return i
>>> [steps_before_cycle(n) for n in range(25)]
[0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 2, 6, 4, 5, 3, 3, 5, 4, 6, 2, 1, 2, 6]
let's check oeis
@orlp -10/10 for not using tortoise and hare algorithm
@LeakyNun That's really overkill for a beginning investigation. :P
in Mos Eisley, 9 mins ago, by Wad Cheber
If you search for "quote", it just shows you all the quotes, not just the times "quote" was said.
> The first n with period length 2 and a nontrivial periodic part is 1012 (cf. A072140).
@El'endiaStarman joking :P
> The period length of the 'Reverse and Subtract' trajectory of n is greater than 1.
11436678 -> 76226733 -> 42464466 -> 23981958 -> 61936974 -> 13973058 -> 71064873 -> 33218856 -> 32662377 -> 44664246 -> 19582398 -> 69746193 -> 30581397 -> 48737106 -> 11436678
3:11 AM
>>> all(cycle_len(n) == 1 for n in range(1012))
True
>>> all(cycle_len(n) == 1 for n in range(1013))
all right
now off you guys go
make challenges
What's the simplest way to get a random integer 1 through 7 using only really simple operations? Context: am working on a tetris program for quest-for-tetris, and the current algorithm is a 16-bit xorshift.
@PhiNotPi where is your seed?
@PhiNotPi what kind of requirements do you have on unpredictability and uniform distribution?
also, what forms the primitive operations in your game of life?
e.g. would multiplication/modulo be hard?
@orlp distribution should be relatively uniform, no obvious pattern when looking at small groups of successive tetriminoes. It can fail all cryptographic tests badly, though.
multiplication and modulo are hard
@orlp We don't yet have an implementation of multiplication or modulo in circuitry, so we don't have those as commands. We do have shifts and addition/subtraction.
Multiplication and modulo can be done with loops/subroutines.
3:18 AM
we have addition, subtraction, shifts, all of the bitwise operations (and, or, xor, and-not, etc.)
1 to 8 would be easy...
@PhiNotPi Those are precisely the only four bitwise operations we have. :P
@El'endiaStarman NAND is functionally complete
Yes; my point is that the four he listed are the only bitwise operations we have as commands, besides the shifts.
Anything else needs at least two commands.
Although a NAND operation would not be that hard in circuitry, just not as a single gate.
3:20 AM
Yeah, it'd be a clock-gate pair.
@PhiNotPi: Would 0-6 suffice as well?
@El'endiaStarman of course
you put me in a giant room filled with wires and nand gates and two millenia later I'll be playing tetris on my PC
Maybe we could do x -> (x*13)%7. Yes, multiplication and modulo are hard, but this is at least conceptually simple...
...wait, it's also cyclic, which is not random. Nevermind.
@El'endiaStarman any bounded memory PRNG is cyclic
True, but this one has an especially short period.
3:25 AM
well, the biggest problem is that it's a generator of the field
so you get every element exactly once
Our period can probably be pretty short. Who's gonna play the game long enough to notice a repetition every 70 tiles?
what you want is a bigger state than the output field
@PhiNotPi My suggestion would have the whole sequence of tiles repeat every 7 tiles. :P
I propose cats.
3:27 AM
Hello
We're probably gonna have to use modulo to get anything interesting.
Introduce cats sitting on keyboards as your random source.
I don't have a cat, therefore no randomness.
I WANT A KITTY
Use a cloud cat service then.
3:28 AM
# output random values from 1 to 7, inclusive
my randval;
my rand = 1;
do {
  do {
    call rand = random(rand).state;
    randval = rand & 7;
  } while(randval < 1);
  display = randval;
} while (1);

sub random(state, shifted) {
  shifted = state << 1;
  state ^= shifted;
  shifted = state >>> 5;
  state ^= shifted;
  shifted = state << 2;
  state ^= shifted;
}
while loops in life?
pls
if you get 16 bits of state
no one will notice the bias of a mod 7
rawr I am lion
am also lion
not a pretty fluffy lion like me
@PhiNotPi That almost matches with your name
Good morning for everyone in timezones 0 - +5
I'm in timezone +3, so count me in!
3:38 AM
Good morning @zʏᴀʙiɴ101! Wait, where do you live again?
@Bálint how?
@PhiNotPi .
> Life of Phi
@zʏᴀʙiɴ101 Oh, yeah.
@Bálint You can make wiki one boxes inside replies with the same success as regular messages.
But messages with pings won't work.
Thanks, I didn't know thag
4:35 AM
Anyone know how I can check if a flag exists in a makefile? e.g. If I have a --production flag on the makefile can I run some special commands?
@Upgoat The easiest way: Use a target instead. :P /s
ಠ_ಠ
@Upgoat I was being sarcastic...
ಠ_ಠ
4:42 AM
I've seen that but that is ridiculousojlyr hacky
I'm thinking I should just have ./config --production && make && make install. That idiom is very common
@Upgoat Yes.
but that's annoying when devloping
4:55 AM
> And that's because, according to Brennan, there's no one else for people to turn to: if they don't want to use US-based technology because it's been forced to use weakened cryptography, they'll be out of luck because non-American solutions are simply "theoretical."
There are no words to express how stupid this is.
@mınxomaτ There are some, but they're usually censored encrypted.
5:16 AM
@mınxomaτ wat
> 'You're a theoretical f'king object!' ... CIA boss John Brennan
10/10 very professional
@PhiNotPi Petition to leave the update there for another 4 days.
5:55 AM
@Upgoat Is cheddar supported on windows?
yes but you must have: git, npm, and node already installed
Ok
I have git, but not the others
Do you recommend 4.4.5 or 6.2.1?

« first day (1966 days earlier)      last day (3067 days later) »