« first day (4262 days earlier)      last day (586 days later) » 

2:15 AM
 $#"
{=*.
/\|%
?,:;
rate my nybbles
 
How numbers
 
still #
 
how add numbers
 
* to access standard operators, and then whatever of those i decide completes the addition digraph
maybe i replace % with + because + looks nicer and the addition digraph is more obvious lol
*+
 
@thejonymyster wait what do you mean standard operators
 
2:20 AM
im separating them from like, assignment operators
 
So like most commands are standard operators?
 
here ill grab from my little sketch pad here lol
$var access
#number
"string

.functions
*operators
=assignment/comparison

/reduce*
\scan*
|map*
+idk*

{open
,com
?ternary
:ternary
;close
 nop+nopify
nothings "finalized" but this is sort of the idea im going for
 
I just came across the strangest thing
 
oh i forgot to expand "com" to be "comma"; in many cases it acts like a literal comma lol
oh do tell, radvylf
 
I made a pretty simple mistake while writing some stuff involving a 2d grid
 
2:23 AM
everyone point and laugh
 
Instead of array[x + y * width] >> 2, I just wrote array >> 2
And this array was about 100k items
And here's the weird part
It made the program extremely slow
Even though, as far as I'm aware, bitshifting an array is entirely meaningless
 
yea i was about to say what would that even do
 
Ohhhhhhhh wait
 
what lang, some c-like?
 
dumb type system strikes again
It's stringifying the whole list
 
2:25 AM
lmao
 
Then casting to a number and bitshifting it
Gotta love JS
 
LOL thats actually hilarious
 
Bruh what, why js do eat lolol
That*
 
wwjsd
 
Autocorrect smh
 
2:26 AM
that's just how js works
coerces everything to whatever it needs to be
so like [3] == 3
 
yeah, but coercing a typed array to a string then again to a number is a pretty weird behavior, even for JS
 
that's why you should always use ===
 
@Steffan That's kinda a bad example
 
i know
[] == 0
 
@Steffan how would that save you from bitshifting an array :P
 
2:28 AM
it wouldn't lol
why on earth would you want to bitshift an array though
 
array language brain took over
 
i guess it could vectorize
but mainstream languages never do that
 
I could see myself accidentally trying to bitshift an array lol
 
@Steffan The point is you wouldn't :p
I still managed to do it by accident though
 
clearly it should have the same effect as array.shift() twice
 
2:32 AM
What’s array. Shift()?
 
Pops from the front
 
array=[1,2,3] //hello im an array
console.log(array.shift()) // 1
console.log(array) // [2,3]
 
@thejonymyster My first thought was actually "did they add experimental syntax sugar for shifting" lol
 
lmao
verbose language where all the verbose keywords are still just combinations of symbols trying to convey their meaning to you
like emojicode but without the emojis
 
3:11 AM
@thejonymyster so basically everything useful is going to be *<command>?
 
not everything :P
but still, whatcha gettin at
 
so if I want to add two numbers and then subtract three I'd need to do something like *+#3;*-?
 
something of that sort, its not stack based though
 
 
correct response
 
3:23 AM
congratulations you've invented a SBCS language
 
lolol
 
If I need to use *<other half byte> for most things, you might as well use the full byte
because then you get more command options
 
its really more akin to a variable width encoding
 
yeah but you lose out on a lot of command options
 
since the increment for stuff is still half bytes and not full bytes
 
3:24 AM
*<other half byte> only gives you 16 operators
using the full byte gives you 256
minus syntax and literals
 
well "using the full byte" doesnt come from thin air, i have to change everything else too :P
 
that's my point
 
dont wanna :sob: dfsg
ill think about it. maybe ill make soemthing good for once :peace:
 
 
2 hours later…
5:08 AM
0
A: Sandbox for Proposed Challenges

Aiden4Find a word in the dictionary of all possible words Given an alphabet represented as a nonempty set of positive integers, and a word made up of symbols from that alphabet, find that word's position in the lexicographically ordered set of all words, assuming words can't contain duplicate symbols. ...

 
6:06 AM
@lyxal I don't know yet, still prototyping. Not even sure if this idea is viable yet, emanresu pointed out some issues with my approach
 
only your sampling approach
not the actual tree generation as far as we know
maybe you got unlucky and got all 45 answers with
which skewed the results maybe
 
I found a mistake, even though I requested a page size of 100 the api is returning only 30 responses at a time. Even though the docs state 100 is the max value
got 1672 responses now with a page size of 30 so that's a lot better
 
@mousetail well if you ever want the encoding to be officially vyxal recognised, just ping me and I'll make it happen
 
Thanks, I'll let you know
 
6:47 AM
@lyxal Am I correct that you can always replace line breaks with spaces and preserve meaning?
 
no
\
# comment here to preserve newline
that pushes the string "\n" to the stack
 
Hmm ok sad
 
and once #1424 is merged, newlines will have even more meaning
 
I won't mess with it in that case
 
 
1 hour later…
7:51 AM
                  origninal     compressed
min              0.00       0.00
max           6553.00    5346.00
mean            32.69      29.57
10% med          4.00       4.00
25% med          6.00       6.00
median          10.00      10.00
75% med         20.00      20.00
90% med         50.00      49.00
min mode         6.00       5.00
max mode         6.00       5.00
stdev          199.68     150.75
Seems to save around 3 bytes on average, much more so for longer answers though, barely does anything for short answers which is the majority of them
at least it doesn't make them any worse either
 
 
3 hours later…
11:10 AM
TIL you don't need parenthesis to define tuples in python.
x=1,2 works
 
You can also use x=1, to define a tuple with 1 element
note the comma
 
12:04 PM
0
Q: Inactivity on bounties with no deadlines?

DialFrostThe List of bounties with no deadline is a page where there are bounties that can be awarded anytime. However majority of these were created quite awhile ago and some comments don't receive feedback on saying "This answer deserves the bounty". How can be bring this back? There's also the problem ...

 
12:18 PM
@mousetail yeah i found that out when i had a stray comma andwas wondering why my code was not working :p
i fixed it that day, but today i realised that it was treating it as a tuple
 
(zoom way out)
 
@mousetail ಠ_ಠ
 
@mousetail ooh nice it seems to account for digraphs
 
what the hell is that
 
Yea I account for some digraphs
wanted to do all of them but don't think there is enough data and I'll just overfit
 
12:24 PM
frick it almost crashed my browser when I tried to zoom in
 
                 Orig       Comp       Diff       Ratio
mean            32.69      27.97      -4.72     0.9717
stdev          199.68     135.39      97.74     0.0762
min              0.00       0.00   -3680.00     0.1304
10% med          4.00       4.00      -3.00     0.8750
25% med          6.00       6.00      -1.00     0.9286
median          10.00      10.00       0.00     1.0000
75% med         20.00      19.00       0.00     1.0000
90% med         50.00      48.00       0.00     1.0000
max           6553.00    5055.00      39.00     1.3333
 
@Ginger vyι (or vyμ - idk what name was chosen in the end)
@mousetail dang that's pretty good
 
vyµ, pronounced "vim"?
 
does that mean there's a 4 byte saving on average?
 
Nearly 4 bytes yes
with 100 byte standard deviation lol so the average is a bit useless
 
12:28 PM
@RadvylfPrograms i think youre misunderstanding what im doing
the more common a digit, the higher ill stick it up on fig's huffman tree
im not talking abt digit runs here
 
@mousetail what's the average byte save on shorter programs?
 
Define short and I can get you the answer
 
< 30 bytes
also, when you have the length of the compressed program is that in raw bits or is further compression applied?
 
                     Orig       Comp       Diff       Ratio
    mean            10.05       9.72      -0.33     0.9796
    stdev            6.36       5.97       0.84     0.0688
    min              0.00       0.00      -5.00     0.6667
    10% med          3.00       3.00      -1.00     0.8889
    25% med          5.00       5.00      -1.00     0.9474
    median           8.00       8.00       0.00     1.0000
    75% med         14.00      13.00       0.00     1.0000
    90% med         20.00      19.00       0.00     1.0000
 
@lyxal thats short?
 
12:31 PM
@lyxal Length is bytes, including padding 0s at the end
 
@Seggan that's a range where huffman would probably give a slight edge over jelly/05ab1e/etc
@mousetail interesting. What are the results without the 0s?
because I'm sure base-255 encoding could get those byte counts down a bit more
 
base255 encoding?
 
like arn
 
@mousetail convert the bits to base 10, convert that to base 255 using the vyxal codepage as the alphabet
 
I already use every bit, base 256 encoding
including null bytes
I strip any null bytes from the end though, they are implied
 
12:34 PM
base-255 encoding/compression works a little differently to that
 
lower base will always be worse though right?
 
base-255 doesn't care about how many bits are used
it's more how many "digits" are in a number
 
Can you link me something about it? My google skills are failing me
 
say you have a number like 149238032. That requires 9 bytes to store as a program (one for each digit). If you convert it to base 16, you get 8e53110 which is 7 bytes, which is an improvement. If you convert it to base 36 (using 0-9A-Z as the alphabet) you get 2GUOSW which is 6 bytes - a 3 byte improvement over the original number
 
this is only needed because you need a terminator character right?
 
12:40 PM
converting it to base 255 gives ⟨ 9 | 0 | 22 | 47 ⟩ as a list of digits which, when indexing into the vyxal codepage gives «λæ0
 
so you need to reserve one code point
 
@mousetail no
huffman encoding returns a series of bits, yes?
 
yes
 
treat that series of bits as a base 10 number
and then convert that base 10 number to base 255
and that's the program
and that could be even shorter
 
that would cost one extra byte every 255 bytes right?
 
12:41 PM
tell me, what's the bit sequence for ₁ƛ₍₃₅kF½*∑∴ without padding?
 
Should be 88 bits in a SBCS
 
but what's the exact bit sequence
in your encoding
 
[1, 0, 1, 1, 1, 1, 1, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 1, 1, 1, 0, 1, 0, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 1, 0, 0, 0, 1, 0, 1, 0, 1, 1, 1, 1, 0, 1, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 1, 1, 0, 0, 1, 0, 1, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
Or in bytes b"\xbe\x92\x1dG\x03\xfc\x16+\xd9'+\xc0"
 
odd, that's somehow longer than the original program
 
That happens sometimes
Every compression method has this issue, it makes more common programs shorter at the expense of rarer programs
I do pad 8 extra 0 bits at the end
 
12:46 PM
so is that 13 or 14 bytes?
with padding
 
12 bytes
I remove 0s at the end when converting to bytes
 
so the list becomes [1, 0, 1, 1, 1, 1, 1, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 1, 1, 1, 0, 1, 0, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 1, 0, 0, 0, 1, 0, 1, 0, 1, 1, 1, 1, 0, 1, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 1, 1, 0, 0, 1, 0, 1, 0, 1, 1, 1, 1]?
 
Yes I think so
 
2 extra bits over the original :-)
 
my base 255 idea comes out to ¬⅛HuÞ†ǐX⁋dṫ⁋ which is also 12 bytes
 
12:49 PM
i still dont understand why 255 and not 256 :-0
 
@thejonymyster this
I don't understand how a lower base could ever be better
 
well you see, it's because I forgot that the encoding doesn't need to be surrounded by any characters like usual base-255 compression does
 
I do actually have a "end" character but it's purposefully made of only 0 bits and I imply a infinite number of 0 bits at the end
 
@lyxal of course if you use base 149238032 you could possibly store the number in 1 byte
ig
 
1:01 PM
@py3programmer me when i definitely know what a byte is :P
 
wait a minute
which is the lowest possible base (assuming we have enough characters) that can store the number 149238032 in a byte?
 
149238033 right?
 
there is no base which can store any number higher than 255 in one byte
 
if you have very wide bytes
 
Did you mean: character
 
1:03 PM
bytes are not always 8 bits on all systems
nowadays they mostly are but there used to be variation
 
crazy
imagine trying to interface with a machine that doesnt agree with how many bits are in a byte
 
quantum computer?
 
On CGCC though, "byte" always refers to exactly 8 bits of information
 
no wait that operates on qubits
 
1:04 PM
There are barely 8 qbits on the earth
 
make more then smh
 
is it "qubit" or "qbit"?
 
cube-it
 
cuebit
damn we made the same joke at the same time
 
disagree
 
1:05 PM
remember when some of us tried to pronounce "lyxal"
Bubbler wrote "licks-all"
 
lets all agree never to be creative pronounce things again
 
Is there another way to pronounce it?
 
@py3programmer That's still how I usually pronounce it lol
 
i dont think so
 
apparently its supposed to be "likes all" but like
 
1:06 PM
@mousetail The correct pronounciation is "like-sall"
 
its obviously licks all yeah
 
me too @thejonymyster
 
Likewise for Vyxal ("vike-sall")
 
That's the main reason I call it lyxial since it sounds less disgusting
 
if you pronounced vyxal similarly to lyxal it'd be vicks-all
 
1:07 PM
imagine just adding a letter into any word you dont like the pronounciation of
 
I do that subconsciously a lot
My brain is like "that can't be right" then makes something up without telling me about it
 
@py3programmer yeah, to me it makes sense, its like vixen
 
does anyone here use firefox?
 
@lyxal whenever possible :-)
 
Well I pronounce @lyxal as he wants so doesn't matter
nope
 
1:08 PM
I use FF
 
I use google
sometimes edge
 
Switch to FF
you'll never go back
 
...to manifest v3? :p
 
@mousetail (and @thejonymyster) could you tell me if tooltips on the keyboard here show for you when you hover over a character
@mousetail I already have. I'm asking for debugging purposes :p
 
@lyxal They look fine
 
1:09 PM
@lyxal they do but they also take up most of my screen :p
 
ah im at work at the moment though, no firefox here :cry:
 
tooltip covers the keyboard though
I'd move it down a bit
 
^^ you should make it clickthroughable
 
your're working right now @thejonymyster?
 
or even better put the info in that space above the keyboard next to the search box
 
1:10 PM
Or at least hoverthroughable
 
@py3programmer im at work, not the same thing :P
 
It's friday, the national slack off at work day
 
but yea the tip covering the keyboard seems annoying, since it means i cant move my mouse down into other keys
 
i dont work
 
dang, must be some weird af tracker blocking FF is doing then
 
1:12 PM
still have school
but i dont program much
other than here
 
@lyxal wdym :o
 
well if the tooltips are loading for y'all using FF but not me, it must be something FF is blocking
 
oh weird!
 
@lyxal Might still have some cached assets
try clearing cache and then hard reload
 
Hey @lyxal, what exactly do you feel when someone lets you know they pronounce your internet name "licks-all"
just curious
 
1:15 PM
@py3programmer well I suppose I think "what a nerd"
@mousetail already did
they work on FF mobile and chromium browsers
 
hmm weired
 
have for several months now too
 
clearly it's supposed to be /lyxal/
 
so you think @Bubbler and @RadvylfPrograms are nerds
 
Is a code golfer who is not a nerd possible?
 
1:16 PM
@mousetail I did
It was buggy
 
:/
 
maybe
 
The tracking protection is really good though
saves a lot of money on flight tickets
 
my friends are the opposite of nerds
yet their codes are shorter than mine by a factor of 10
they should join ppcg
 
what kinds of friends are these
 
1:18 PM
and with them, do you need enemies?
 
Imagine there is a different underground golfing society whose meta is so evolved they can beat the best of us by 10x. Yet we never know about them
 
that just sounds like the code golf discord server with extra steps
and the code.golf discord server too
 
what are they hiding from us
 
that's a very good question
 
If we knew they wouldn't be very good at hiding
 
1:20 PM
I've tried many a time to infiltrate their ranks to no avail
 
1:40 PM
@lyxal I made a repo for my work so far. https://github.com/mousetail/vyx-

Now to figure out how to make it work in the language itself
 
@mousetail visit github.com/Vyxal
 
What a honor
 
it makes permissions management much easier :p
you should be able to make new branches on the main repo without forking now
 
That's nice
 
and if you want, you can move the encoding repo into the organisation
 
1:52 PM
I will do that soon, gonna clean some stuff up first
 
cool cool
 
Then I'll make a PR to accept my encoding as a flag (?) is probably the easiest
 
yeah, that's how I would suggest doing it
encoding.py sounds like the file you want to focus on the most
 
Thanks I'll check it out
 
@mousetail oh and, hello, and welcome to the Vyxal family. My name is lyxal gaming, but you can call me lyxal. I am thrilled that you will be joining our team. Each and every day we serve our users exceptional bytes with impeccable service. We take pride in everything that we do.
 
2:01 PM
@lyxal multiple bytes per day? not very golfy :/
 
I only eat in nibbles
 
2:51 PM
@thejonymyster Multiple challenges solved per day, tho
 
@lyxal me
@mousetail hehe you had the same idea as me to store the trees as json :P
 
I wouldn't even know a practical alternative
 
ascii art :P
 
@mousetail something like polish notation where its either a "continue" character or a terminal character
 
@mousetail Lisp
 
3:03 PM
wow i did not communicate the thing i was saying good
 
@Seggan Is your tree going to handle digraphs the same as me? If so we should coordinate so that we need to change the vyxal code only once
 
im making a tree for Fig lol
i thought i said that
how are you handling digraphs?
yesterday, by Seggan
@thejonymyster this is what ive always been looking for
 
3:18 PM
@Seggan Oops my bad
I have a few alternate trees depending on the previous character
 
so digraphs are a single node?
 
no
the first char of a digraph starts a seperate tree
 
ah
 
I'll try single node digraph mode though to see if it's better
 
my plan is to have a "master tree" that has digraphs close to the top so they dont cost as much
as fig doesnt have too much monograph space
 
3:54 PM
1
Q: Is this a refinement?

Wheat WizardA partition of a list \$A\$ is a way of splitting \$A\$ up into smaller parts, concretely it is list of lists that when concatenated gives back \$A\$. For example [[1],[2,3],[51,6]] is a partition of [1,2,3,5,6]. The trivial partition is a partition that "splits" the list into only one piece, so ...

 
4:05 PM
unforch dupe, it is a good idea
 
4:21 PM
ouch since when do mods get hit by the dupe hammer :P
 
All shall be Mjolnir'd
 
4:46 PM
0
Q: Hexadecimal -> Binary Art

The ThonnuTake a list of 2-digit hexadecimal numbers as input, and output the binary value, replacing each 1 with an 'X', and each 0 with a space. For example Input = FF, 81, 47, 99. FF = 11111111 in binary, so print XXXXXXXX 81 = 10000001 in binary, so print X X 47 = 01000111 in binary, so print X ...

 
5:05 PM
Anyone else constantly get hamming codes and huffman coding mixed up?
 
cant say i have, no
i wonder how one might combine these two concepts however
some kind of ham-man code
 
@thejonymyster wait really
am I just bad
 
@RadvylfPrograms Nope, not even once
 
I can never remember more than one of those names at once
and it tends to be the one opposite to what I'm currently in need of
 
@thejonymyster That's my stone age name
 
5:09 PM
youre thinking of hammurabi
but make sure not to ham ur rabbi
 
Oh hey I just learned about that guy in art history class today
 
hammurabi code golf. "a byte for a byte"
5
 
@RadvylfPrograms i keep thinking its hoffman coding
 
5:39 PM
@RadvylfPrograms i was so close to asking "did you never hear about him in high school" until i remembered you're in high school and art history is just an actual ap class for some reason
 
6:06 PM
 
6:36 PM
Many APs are dumb
CS AP was free points for me though which was nice
 
cs ap is hopefully free points for anyone
 
6:49 PM
I'm taking 7 AP classes this year
Some are definitely dumb
 
 
2 hours later…
8:20 PM
CMM: should I edit out the in PHP (or anything else, for that matter) from the title of codegolf.stackexchange.com/q/9165/107299?
 
I don't think so
If it was just "in PHP", probably, but it's clear that it's language agnostic
 
yeah but imo it just adds noise
 
Our titles are rarely the succinct "summarise the problem" that the rest of the network has
It's been unedited for almost 10 years, I see no reason to edit it now
 
true
 
@cairdcoinheringaahing how about the noisy factor? every time ive looked at it i didnt like it :P
 
8:26 PM
Is the "in PHP (or anything else, for that matter)" the noisy bit to you, or is it the BASIC code?
 
only the (or anything else for that matter) , and maybe change in to from
theres gotta be a more terse way to say "by the way this is a regular challenge that doesnt restrict this to one language"
since thats basically the standard
or sure get rid of the whole in php ...
if youre wondering what language its from you can find out by like, clicking in
hold on ill clarify in a sec brb
 
@cairdcoinheringaahing Wouldn't the opposite make more sense?
 
@thejonymyster this
 
in PHP (or anything else, for that matter) is 100% noise
It is entirely meaningless
 
^
 
8:31 PM
Whereas in PHP communicates a meaningful restriction
 
@RadvylfPrograms yeah thats what i was about to get at
 
¯\_(ツ)_/¯ Sure I guess. Just think that it's an old question, no harm is leaving it that way
 
all "(or anything else for that matter)" serves as is the clarifier, for the "in php", but the "in php" is just trivia and not necessary
 
Y'all aren't even close to the most zealous editors the site has ever had, and even they left it alone :P
 
if it were phrased like "PHP's (code snippet)" itd probably work without losing information
^^ lol
CMC: given two pairs of points [(a,b), (c,d)] and [(e,f), (g,h)], determine if the line from (a,b) to (c,d) intersects the line from (e,f) to (g,h). you can assume they have different slopes, and that none of the four given points are equal, and that the intersection (if it exists) wont be at one of the points
 
8:41 PM
@thejonymyster Can we assume that no triple will be colinear?
 
hm, sure
yeah that makes sense, since the only times they would be intersecting would be illegal, making it an edge case
 
8:54 PM
@thejonymyster Geogebra 36 bytes, Intersect(Segment(A,B),Segment(C,D))
Returns the coordinates of the intersection else nothing.
 
9:13 PM
nice
 
9:31 PM
posted on September 30, 2022 by trichoplax‭

Given a chess board with some knights on it, say how many squares are neither attacked by a knight nor containing a knight. Input An 8 by 8 grid where each square is eit...

 
 
1 hour later…
11:00 PM
@thejonymyster how can a point be equal?
is the point (6, 2) equal?
 
I'm guessing the "to each other" is implied
 

« first day (4262 days earlier)      last day (586 days later) »