« first day (3974 days earlier)      last day (876 days later) » 

6:00 PM
See the location of caird about
 
Oh lol
Huh, I've raised a total of one R/A flag here
Wonder what it was for
 
what is R/A?
 
Rude/abusive
Oh it was an answer from half a year ago that contained multiple racial slurs
 
@RedwolfPrograms the molar gas constant / ampere = 8.3144598 m2 kg s-2 A-1 K-1 mol-1
 
CMC: Find a situation where Fmbalbuena's unit (molar gas constant / ampere) is actually useful
Oh wait, is that a unit?
 
6:06 PM
i found google and i got "molar gas constant / ampere = 8.3144598 m2 kg s-2 A-1 K-1 mol-1"
 
@Fmbalbuena Idk, I keep trying to leave but they don't let me :P
3
 
isn't the molar gas constant only useful for PV=nRT
 
Hey, does anyone know of a BF division algorithm that takes input a b and returns a/b 0?
 
@JosephWalker not me
 
Using only two cells?
 
6:15 PM
It can use cells to the right of those as long as it sets them to zero when done.
Also, I saw someone talking about the ideal gas law, and I may be able to help because I constantly have to use it for my AP Chemistry class.
 
Anyone want to try attempting my answer to Hide a message in an image?
It's, in theory, low difficulty
 
hint
 
@RedwolfPrograms ...what is that second image? lol
 
@RedwolfPrograms The first thing I tried was googling the message...
 
@pxeger That's like...my trademark lol
A wheeled llama with PSIF compression
It's also the one with the most breadcrumbs
 
6:29 PM
I was looking at open.kattis.com/problems/dictionary but I don't even understand the task. Is it clear to anyone here?
 
You need the output the longest chunk of the dictionary you can in under 91000 bytes
But it must start at the beginning
 
This looks like a good question for here. But we could reduce 91000 to something much smaller
What do you think?
 
It might be a near-dupe of something, but I like that sort of challenge
Definitely sandbox it first, though
 
Cool. As long as you don't down vote it :)
 
@graffe maybe 1000
 
6:33 PM
Unless they chose 91000 for some special reason....
@grandBagel sounds good to me
 
it appears I have screwed up
APOL needs a codepage in order for each character to be one byte
 
Maybe that is chosen because it's possible to do the entire dictionary
 
I'm guessing there's roughly 91k words
 
but I chose the characters for each instruction indiscriminately, so I'm not sure how to make a codepage out of it
 
Just arrange them in a semi-random order
Choosing characters is the hardest part
 
6:35 PM
how to get christmas hat
 
@RedwolfPrograms So just slap em into a table a la github.com/DennisMitchell/jellylanguage/wiki/Code-page ?
 
@Fmbalbuena do you mean this?
 
@grandBagel yes
no, not starfish
 
@Fmbalbuena Just post two comments that both have score >= 5
 
6:37 PM
@RedwolfPrograms Jelly's table has 256 instruction slots, but I only have ~50 instructions. I assume I should leave unused ones empty
 
@Fmbalbuena this?
 
no, christmas hat
 
@GingerIndustries Yep. Although you'll probably want all of printable ASCII
 
@Fmbalbuena secret hat?
 
@grandBagel mmm... maybe
 
6:38 PM
@RedwolfPrograms Oh, a-zA-Z etc
 
@Fmbalbuena do you have it?
 
@Fmbalbuena like my hat?
 
@GingerIndustries no, caird or thou.
 
@GingerIndustries Your interpreter also needs to have a way to interpret a file composed of raw bytes as APOL code. That way, when you claim a score of 13 bytes, you can (if asked) provide an actual 13-byte file that runs the correct program.
 
@GingerIndustries Everything from (0x20) to ~ (0x7e) yeah
 
6:40 PM
@RedwolfPrograms That almost looks like a job for puzzling.se
 
@DLosc what happens if the last bit is 1?
@grandBagel WH47 D0 Y0U M34N?
 
@Fmbalbuena Why is that any different from if the last bit is 0?
 
UTF-8
 
An SBCS is a replacement for UTF-8
 
6:43 PM
@RedwolfPrograms i know but if the encoder from SBCS to UTF-8, and the last bit is 1, what happens?
 
You don't just copy the binary into UTF-8 code points, you map it to the relevant Unicode character
So if 10110011 mapped to ¬, it would get mapped to the UTF-8 sequence for the character ¬
(11000010 10101100)
Not being able to use special characters in google searches is so annoying
 
@RedwolfPrograms so true
Any feedback for this?
 
@RedwolfPrograms How long have you been working on this answer?
And, is the image metadata relevant? I thought the challenge banned that but it seems like it does not
 
@tjjfvi Took me about half a day at most
I had some useful tools already developed, like steganosaurus, and some programs I'd written to test it
No relevant metadata
It's all contained in the images themselves
 
Hmm
Nov 14 at 3:41, by Redwolf Programs
user image
 
6:57 PM
I used that image as the container, it doesn't contain any encoded information
 
^ afaict this has the exact same pixel data as the image in your answer, save some cropping
 
Uh oh
it does
Is imgur dropping some low bits or something?
Wait nvm
It's all there
 
Did you upload the wrong image?
 
Nope, I just had a filter set in steganosaurus and forgot about it
The message is there
(Downloaded it again, in incognito, directly from CGCC and it shows the correct message when fed to my decoder script)
 
Hmm ok
Is the original image required for decoding?
 
7:02 PM
No
But I suppose you could compare the two to find any slight differences, which could be a little bit helpful
But as far as I'm aware, the other two originals aren't publically available
 
Aug 17 at 19:07, by Browncat Programs
user image
 
Oh huh
Well, the snow one shouldn't be here
 
Yeah I don't see that one
 
@tjjfvi Aw...
 
That's Scully I think
 
7:11 PM
@RedwolfPrograms Can you confirm that the md5 sum of the downloaded png is 46c897e0588c289f52864a55f5d86871?
 
Yep, 46c897e0588c289f52864a55f5d86871
 
Thanks
(though I suppose for security I should've made you send me the hash first :P)
 
lol
And, for future reference, 4bf4130af67681416775706b9399002b is the sum for the cat, and 7766488bbc2f3a849aff5d97d555791c for the snow
 
@Fmbalbuena My Santa hat isn't a real hat, it's from Winter Bash 2019 and I took a screenshot and kept it
 
@cairdcoinheringaahing i can't believe you would deceive us like this
 
7:16 PM
My other profile picture also has a hat from Winter Bash 2019 :P
 
@cairdcoinheringaahing i know
 
my santa hat literally never existed
 
@RedwolfPrograms For future reference, I could make their hashes collide for a few dollars.
(As in server costs)
(md5 bad)
 
Yeah, but md5's what was requested
SHA-512 for the win otherwise
 
Yeah
 
7:19 PM
I'll also take SHA-512
MD5 just seemed to be the most likely to be easy for both parties
 
Yeah, the point of this wasn't security, it's just to make sure no low bit data was destroyed by compression or something
 
Yeah, imgur weird
Or, annoying if you want the original image
Is this from the spec of that old challenge from a few years back, or something else?
 
What?
If you mean the images and stuff, they're from this
 
Oh, I thought that was dead
I have a quite easy one if you want it (I can just undelete)
 
7:25 PM
I looked at it a little yesterday, I'll attempt it if you undelete it
 
Ok, there you go
(I forgot you could see deleted posts)
 
I'm so confused rn
 
et voila codepage complete!
 
There is a difference between the llama message with code and the llama message from nov, right?
 
Yep
What are you using to check?
 
7:27 PM
Photoshop
I only see two pixels of difference
And those are the two pixels I manually changed
...Maybe I put the same image in twice
 
@tjjfvi The llama was created with a special colour pallete that takes up one bit per pixel.
 
*1 byte
But the stegano llama doesn't use PSIF
 
@RedwolfPrograms Exactly!
 
^ is the diff I get
 
@tjjfvi a photo of caird's soul
 
7:32 PM
That one white pixel is one I manually changed in the llama's leg to a slighter lighter black
So either photoshop's diff is completely failing me
Or there is no difference between the two
 
The difference is not going to be visible
It's very subtle changes to the pixel values
I'd estimate maybe four times below what's easily noticable with the naked eye
Perhaps brightening would work?
 
This should blow out the pixel differences
 
Nov 14 at 2:00, by Redwolf Programs
user image
Any pixels not in there have been modified
 
Hmm ok
I'll check that later
For now I should stop procrastinating
 
... someone at a high school in my city got stabbed???
 
7:36 PM
@emanresuA That's likely a bugged color palette
 
apparently some students from two different high schools got into a fight by the marketplace near there and there was an axe, a machete, and a baseball bat involved. i'm so confused
@tjjfvi this is what I get if I replace the black in your image with white
 
I know, nice job
 
Ooh my rep is a permutation of 01234 now
@emanresuA I wasn't sure if I should post it in the comments or here so I did both, probably didn't need to lol
 
The challenge is really overengineered, being both really open and really strict.
 
7:49 PM
SE down?
 
Not for me
 
@RedwolfPrograms nice
 
@RedwolfPrograms Mine is a permutation of 139. (Counting by threes.)
 
Must be an outage in a more local datacenter or something
 
7:51 PM
@pxeger Only SE Chat.
 
@hyper-neutrino That's...really interesting
It's halfway what I expect, but halfway not
 
@pxeger Bug
in CGCC
because wrong logo
 
The fact there's a nice repeating pattern is very odd, since there shouldn't be
 
@pxeger Maybe a cache from the earlier outage?
 
back now
@Adám I don't think it can have been that, because it was taking like a minute loading before finally displaying that error
 
7:55 PM
@RedwolfPrograms I have another idea, but it requires reverse-engineering hash functions
(To get the final image)
 
Sounds like you probably need to change your idea a little then :p (or uses a "hash" like CRC or something XOR based that's more reversible)
 
0
A: Sandbox for Proposed Challenges

FmbalbuenaWhich title i should put? Challenge: Check if the Input is any shorter string (Call X) (len(input())>len(X)) X can be reshaped with the size bigger than the length of the string. Examples: The string "abba" returns Truthy becacause "abb" can be reshaped to the size of 4. The string "acc" returns ...

 
^ Any feedback?
Oh i forgot to add the testcases
 
I'm getting worryingly close to vote capping today
 
@RedwolfPrograms I already am at the vote cap.
Does anyone have an image where each pixel is a random color I can use?
 
8:01 PM
Sure, just a sec
How big?
 
@RedwolfPrograms 200x200
 
There ya go
 
Hey, I finally got a division algorithm together. I used it to answer the translate arithmetic to brainfuck question.
 
@grandBagel gist.github.com/RedwolfPrograms/… will generate as many as you need
 
in The APL Orchard, 29 mins ago, by Fmbalbuena
CMC: check if the list equals to one of "number⍴string" (number > ⍴string) even if i don't understand and recreate but i understand
@RedwolfPrograms ^
 
8:11 PM
@JosephWalker Nice :p
That was a seriously hard thing to answer
 
Used one of my last few precious upvotes for the day on it :p
Down to two
 
and we're done!
PyPOL now has the ability to compile APOL programs to an .abf bytecode file and back
 
@RedwolfPrograms How can I ensure its rgb value sum is a specific value?
 
Start with all pixels being 0x808080 and do a certain number of repetitions of increment one pixel then decrement another
Is this a modular sum or a normal sum?
And is it per-channel or overall?
 
I meant iterate over every pixel, getting its RGB color, (e. g (31, 27, 128)) sum it, and add it to the total.
 
8:17 PM
nvm
Hmm, so getting a fairly uniform distribution of colors would be the hard part
 
@RedwolfPrograms I don't need it to be uniform.
 
You could generate a random image, take the sum, divide by the expected sum, then multiply. That'll scale the colors to roughly the right amount, and you can apply random increments until the sum is exactly correct
That's much faster, and will result in a more uniform (but not perfect) result
 
Repeatedly add pairs of pixels with values adding up to (total sum / 2)
 
How to set a pixel color with Pillow?
 
@grandBagel find in SO.
 
8:21 PM
applause
 
@GingerIndustries use \* and wdym?
 
@Fmbalbuena sorry, was referring to my above statement
 
@hyper-neutrino Seems like that's just an artifact of the screenshot / imgur compression
 
@Fmbalbuena So can I declare one with size (1, 2). When I use image.putpixel((1, 1), <some color>), it gives me a error.
 
I'm thinking we should probably delete this, its answers contain a lot of (obviously) malicious code
Oh there's only three answers lol
Still, two out of three are intentionally harmful
 
8:31 PM
@RedwolfPrograms Agreed
 
10k users, your DVs would be appreciated
 
DV'd
 
@emanresuA What does DV stand for?
 
Delete vote
 
@emanresuA What should I flag it as? Custom?
 
8:36 PM
Don't bother flagging it
 
It'll either send it to a queue for deletion, when it already only need one more DV, or it'll bother a mod
 
There should really be a delete vote review queue
 
10k tools basically are that
 
@grandBagel ask to SO
 
8:44 PM
@tjjfvi ah. that makes sense
also delete vote should be VTD (vote to delete) cuz DV is downvote
 
No, VTD is Vote That Down :p
 
ಠ_ಠ
3
 
Great, I have to go away for the next three days
 
AoC is going to go horribly
 
8:47 PM
:/
TNB won't be the same without you
 
@GingerIndustries Cats and trees
 
@hyper-neutrino I wonder how many times you've posted this exact message and it's gotten starred. I think it's happened at least twice in the past week or two. :P
 
and snow
 
@DLosc well, all I can say about that is: ಠ_ಠ
 
8:53 PM
ಠ_________________ಠ
 
ಠಠ_ಠಠ
they didn't call me four-eyes for nuthin
@RedwolfPrograms time to break out thiscatdoesnotexist.com
 
@hyper-neutrino Huh, those are the only two times. Insert "If I had a nickel..." joke
 
interesting
feels like it's been more
 
@DLosc Five times including it, just those two as a message on its own
And 11 pages total
ಠ_ಠ
 
8:57 PM
we can probably extrapolate from that to figure out when the next ಠ_ಠ will appear
 
lol
lel
lul
lal
lil
lyl
 
ಠ-ಠ
 
ಠ¯ಠ
3
 
ಠ=ಠ
 
9:01 PM
OK, let's not have those faces get out of hand.
 
lal lbl lcl ldl lel lfl lgl lhl lil ljl lkl LLL lml lnl lol lpl lql lrl lsl ltl lul lvl lwl lxl @lyxal lzl
 
Okay that's probably enough /l.l/s
 
@RedwolfPrograms why /l.l/s?
 
¯\_(ಠ_ಠ)_/¯
 
@GingerIndustries lyl lol lul lrl lal lrl lml lil lsl lcl lul ltl
 
3 mins ago, by Redwolf Programs
Okay that's probably enough /l.l/s
 
(╯ಠ_ಠ)╯︵ ┻━┻
 
9:07 PM
oh sorry for saying l.l
@GingerIndustries stop
 
@GingerIndustries In APL, you can write that as ⍉⍪
 
@Adám a prime example of golfposting
since when have I had 1091 reputation
 
That's your network-wide rep
 
@RedwolfPrograms ah
 
9:25 PM
hi everyone
 
@Iamaperson hi to you too
 
Hi!
Been a while
 
okey, I've posted a challenge to codegolf.stackexchange.com/questions/223158/…
good luck
:)
 
Exciting!
 
jazz hands
okay it's possible I screwed up
 
9:38 PM
@Iamaperson Hi Basketball.
 
nope nvm it worked
 
 
@Adám ಠಠಠಠ_ಠಠಠಠ
 
Doesn't look like there's anything in the image low bits, or the metadata...whatever you did must be pretty interesting
 
@RedwolfPrograms sir i don't even know what I did
@RedwolfPrograms i just kinda threw together an algorithm
i've g2g, good luck with those images
 
9:44 PM
o/
 
10:07 PM
@Iamaperson hi
 
@GingerIndustries he wide
 
0
A: Sandbox for Proposed Challenges

JulianOperator precedence is dead Calculate the result of some math expressions using the following constraints: Numbers will be between 0 and 9 Operators are + - / * Expressions will always have the format Number operator number operator number... Parenthesis have the highest precedence The order of ...

 
10:27 PM
@RedwolfPrograms so does that mean it's time for you to start development on your new secret golfing language?
 
10:49 PM
I think that was 2022
 
@lyxal No, I've decided this year's project will be RTO
 

« first day (3974 days earlier)      last day (876 days later) »