« first day (137 days earlier)      last day (592 days later) » 

3:32 AM
I feel like I should want to buy something really cool and expensive now that I have a job
But like...
I've got practically everything I want
(That's less than a million dollars)
 
yeah you kind of have to have something in mind
or else it's just pointless
 
Buy Vyxal
It costs $420.69
 
Ooh yeah, a hostile acquisition. I'll purchase it, run it into the ground, exit scam everyone, and walk away rich with a monopoly on the golfing language game
8
If I can't break anti-trust laws at work (according to the employee handbook), I'll do it on CGCC :p
Jelly and 05AB1E are nonprofits unfortunately, but I can still bribe government officials to strip their tax exempt status
 
@emanresuA really? I thought we were selling for $420.6969
@RadvylfPrograms sure, sure, do that
(no one tell him he'd be buying the most insecure/vulnerable to ace exploits golflang ever made)
 
 
8 hours later…
11:26 AM
@RadvylfPrograms a pi 4 8gb with a touchscreen
make a pi-tablet
 
11:38 AM
@mousetail well just .replace("line", "line segment")
 
@NumberBasher If there is one mistake I'm sure there are more
I can't trust the source anymore
 
@mousetail it isn't a mistake exactly
 
It's outdated information
 
First its how Euclid wrote it, hence historical significance
outdated? sort of
but useless? no way
 
Most of these rules still apply, you would need minor changes to make them relavent again, right now it's just confusing
Useless and misleading
 
11:41 AM
@mousetail what's the difference of lines and line segments anyways, most of the time
 
Lines do not have ends
Segments do
Rays have only one end
 
What difference does it make in proofs
gimme example
 
Anything involving concave shapes
or intersections
the famous proof that 2 lines that don't intersect are paralel
 
@mousetail "proof"?
 
It's a lemma yes
 
11:43 AM
anyways, anything Euclid ssays is good I understand your point
@mousetail It's more of a definition
if not, an axiom
 
@NumberBasher It's not a definition, it doesn't work in 3d for example
a definition must mean that paralel and not intersecting literally mean the same thing
which is not the case
 
@mousetail scratch that, add 2d, put it back again
@mousetail and then your proof wouldn't work
 
@NumberBasher Euclid is good but don't literally use his words in your proofs directly because the definitions of some terms have changes.
@NumberBasher The proof only works in 2d yes
 
What is the definition of parallel?
 
Angle of 0
 
11:45 AM
@mousetail true
@mousetail Huh? Explain
 
If you moved the lines to be on top of each-other, then measured the angle between them it would be 0 for paralel lines and anything else for non-parallel lines. This works in any number of dimensions. This also works for other things that can be parallel like planes, rays, and line segments
Alternatively, you can draw a line from a point on one line and a point on the other line, both parallel lines must have the same angle to the new third line.
 
“If you moved the lines to be on top of each-other,” seems bad fora definition
@mousetail i think you mean for all two points
true, let's wikipedia
 
@NumberBasher Any 2 points also works
If it works for any it works for all
 
> In geometry, parallel lines are coplanar straight lines that do not intersect at any point.
Ahh, definition.
@mousetail this
 
@NumberBasher Coplaner is the hint here though
you need to move them ontop of eachother to prove them coplanar
 
11:53 AM
why?
 
@NumberBasher the slope of one line is equal to the slope of the other
 
Maybe a better definition is that they must have the same distance at every point, that way the definion works for curves too
 
@mousetail it's harder to say "distance" for curves
 
The Wikipedia definition is dumb since a line intersects itself so that would make it not parallel to itself.
 
11:56 AM
yah, why not?
 
@NumberBasher A line is paralel for a given definition of distance
 
@mousetail give a definition
 
(a^2+b^2)**0.5
closest point to every other point
 
so a circle and its center are parallel?
 
Yes
They are not lines though
but curves
 
11:59 AM
@NumberBasher The locus of a point which is equidistant to the closest point in a line, is parallel.
idk if that is right
 
@mousetail AND A PLANE IS (sorry bad capslock) parallel to itself?
 
Yes
 
@PyGamer0 all of the points? or a selection?
@mousetail that's... counterintuitive
 
@NumberBasher all of them?
 
@NumberBasher how is this counterintitutavie? Everything is paralel to itself. Even me
 
12:02 PM
two lines are said to be parallel if their slopes are equal
3
 
Question though: Can planes be parallel to lines?
 
@mousetail i guess that makes sense, a line is parallel to itself
 
@mousetail the corresponding angle theorem?
 
@PyGamer0 so by definition, two lines in 3d space cannot be parallel b/c a line would have to be parallel to an infinite cylinder?
 
no the distance between each point of the lines should be same
or just use the slope definition
 
12:04 PM
@mousetail wait, in your definition a line can be // a plane but a plane cannot
 
Maybe the definition of paralel is not really important but what is important is that these properties all biimply eachother
 
// a line
 
Is biimply a word?
 
@mousetail equivalence?
@mousetail this is not true
 
@NumberBasher it is
 
12:05 PM
two non-coplanar lines, some selection of points
 
except if you choose the intersection point itself
@NumberBasher They can never be paralel
so if will be false for all combinations of points
 
@PyGamer0 <
 
Slope definition is good
 
@mousetail well true
 
@PyGamer0 жаит тчат ехистс
oops russain keyboard was on
@PyGamer0 wait that exists?
 
yes
its good
 
So they agree with me
 
@mousetail What was your hypothesis again?
 
12:11 PM
IDK anymore
I'm not sure what side I'm on and not sure what side anybody else is on either
I just know I disagree with everybody
 
If overlapping lines are not parallel,
@PyGamer0 this is false
@mousetail this is also false
@mousetail lol
 
@NumberBasher It's true, but use 3d angle not 1d angle
 
gotta go, will have argument later
@mousetail I mean, https://chat.stackexchange.com/transcript/message/62092078#62092078 this wouldn't be the definition for parallel b/c it judges overlapping lines as parallel
but gotta go
 
@NumberBasher Overlapping lines are parallel
you literally posted links confirming
 
12:36 PM
@mousetail wait
@mousetail how
 
> Two lines that coincide are the same line.
> A line is parallel to itself, so they are also parallel.
> However, it is clear enough to just say that they coincide.
 
> No, they are not parallel. If two things are said to be 'parallel' then they must not have any intersection(s).
 
It clearly says a line is parallel to itself
 
> OP: I think by convention repeated lines would not be considered parallel with the usual definition of parallel lines being lines in a plane that don't meet.
 
You can't cherry pick the answers you like
 
12:41 PM
Hmm, I didn't look at the quora one which agrees with your point
you know what, frick
 
the other one wouldn't load for me
I think it's obvious that a line is parelel to itself though since it has has all other required properties
 
Oh, OK then...
My math teacher says this: We usually seperate coplanar lines into three categories, overlapping, parallel, and intersecting
which implies that overlapping lines are not parallel
but yah well frick fuck it
Does two overlapping lines intersect itself
 
Yes they intersect
they are paralel too
 
> Parallel curves are curves that do not touch each other or intersect and keep a fixed minimum distance.
interesting
> Given parallel straight lines l and m in Euclidean space, the following properties are equivalent:

Every point on line m is located at exactly the same (minimum) distance from line l (equidistant lines).
Line m is in the same plane as line l but does not intersect l (recall that lines extend to infinity in either direction).
When lines m and l are both intersected by a third straight line (a transversal) in the same plane, the corresponding angles of intersection with the transversal are congruent.
 
4
Q: Do two equal lines also count as intersecting?

Janis PeisenieksSo, I asked a question about how to find if three lines are concurrent. I built the algorithm I needed, and it was working well, until I started doubting my power of judgement. So my question: Are two equal lines also intersecting? For an example, $9x+9y=9$; $7x+7y+7$ ? Because now, if I would...

 
12:46 PM
I love it when See full text does absolutely nothing and disappears
@Seggan nice
2
Q: Parallel lines, parallel to itself

LennartIn a lecture the professor drew two lines, a and b, parallel to each other. The professor then asked if line a is parallel to itself. The students said no because the definition of a parallel line is that it never cuts another line and a cuts itself all the way through. The professor said that ...

> Parallel straight lines are straight lines which, being in the same plane and being produced indefinitely in both directions, do not meet one another in either direction.
 
Again the answer clearly states that definition is outdated and inconvenient
 
What is going on here right now
 
I think it sort of agrees upon both are right
> But it would be wrong to say the student's point of view was wrong, indeed it is the classical point of view.
And nobody said anything was wrong
@lyxal or any other mod: MOVE?
 
152 messages moved from The Nineteenth Byte
 
OK, now what?
 
12:56 PM
eh?
 
I was looking for someone to review my sandbox proposal
 
well, this is off-topic tnb
 
mf bytecode scan algorithm bein foiled by Image Distortion (tm)
 
sheesh
 
12:59 PM
observe the problem:
the red circles are the pixel values that get checked to convert the image into binary
but they do not line up with the actual pixels
I'm currently computing the square width and height values by averaging the width and height of the tracking squares divided by 4
but it ain't workin
I am not sure how to fix this
do you have idea
I could maybe use the timing dots, but I'd have to detect them first and that is easynt
potential idea: draw a line out from the center of the bottom-right tracking dot upwards and assume any square lying along it is timing
 
1:29 PM
well I implemented that algorithm and it still doesn't work
so now I will try my Last Resort: Using the top timing dot as a 3rd corner to undistort the image
 
@NumberBasher how? prove it
 
 
1 hour later…
2:43 PM
@PyGamer0 overlapping lines
additionally, its harder to slope a 3d line
and take into account vertical lines, etc
 
@NumberBasher why, use quaternions?
 
@PyGamer0 5d?
generalize.
pin me when you reply, gotta go
 
3:02 PM
32
Q: slope of a line in 3D coordinate system

2amSuppose I have $2$ points in a 3D coordinate space. Say $p_1=(5,5,5)$, $p_2=(1,2,3)$. How do I find the slope of the line joining $p_1$ and $p_2$? After getting the slope (which I assume will be an integer) how do I get the coordinates of any other arbitrary point on this line? I do understan...

@NumberBasher i think the direction vector can be generalized to $n$ dimensions.
 
ok then... but that's way different from slope.
i agree with it then
 
3:41 PM
first test of distortion correction:
I do not think that is correct
 
4:01 PM
okay, finally did it
...and yet it's STILL screwed up
time to play my donald trump card
oh
oh dear
I have successfully sucked my bytecode into a BLACK HOLE
IT KEEPS GETTING WORSE
 
5:01 PM
okay, finally got it to work
 
5:17 PM
@Ginger pog
Picture?
 
Writing prompt idea: HNQ word salad
> It's Just Rocket Science, a short horror/sci-fi novella by Henry Kissinger: Can the princess escape non-English speakers with parallelograms while visiting Singapore?
 
5:37 PM
@RadvylfPrograms not the decoding, just the correction
but as of now I ALMOST have the decoding working, stay tuned
 
6:04 PM
well
it seems I have ducked up
 
6:28 PM
HAHAAAAAAAAAA
FIRST SUCCESSFUL DECODE
HELL YEA BABY
@RadvylfPrograms Ladies and Gentlemen and Other People, we done it
 
I like distorted stuff
 
okay so
bytecodes are not qr codes
they are a kind of matrix barcode that I invented
so like a qr code but I made it
and I just got one to scan correctly for the first time
here's an image of one (the one that scanned):
that bytecode contains the text "Hello World!"
 
nice
 
so now I can just generate arbitrary bytecodes
containing whatever the hell I want
power overwhelming :p
time to try making a 32x32 one!
 
@Ginger no error correction?
 
6:36 PM
@mathcat yes error correction!
 
parity blocks?
 
even better!
these codes use Hamming correction, which means that for a default bytecode it can detect and correct one error per 16 bits, and detect (but not correct) 2 errors per 16 bits
you will of course be able to change the correction level if you want
 
wow nice
 
nice!
 
time to propose this to ANSI and see it storm the world :P
 
6:40 PM
I wanted to do one to
(with different colors)
so that it can store videos/gifs
 
what would it do under different lighting
 
wat
 
talking to mathcat
 
so I may have underestimated the capacity of a size 2 bytecode:
lol
 
@Seggan I think I decided using 16 colors
It could store a gif
but it already exists so I gave up :/
 
6:46 PM
ginger you inspired me to write a challenge
 
oh dear
what is it
 
hey I have an idea
I'll implement a color QR code which can store videos and videos only
That'll be fun
 
you do not want to do that
 
I do
and I'll finish first
 
I have no words for how much of a Pain In My Ass it was to get this damn thing to scan
and even then it still has a lot of issues
 
6:48 PM
yeah well
idc
I have time
 
@Ginger pack the most data you can in a boolean matrix
 
uh
 
what
 
neat
 
neuro-evolutional-augumented-topology?
 
6:58 PM
ooh I love Neuro Evolutional Augmented Topology, saw them open for Correct Horse Battery Staple once
 
Fun fact: You can fit a 1m video in under a megabyte with usable quality
 
no way? really?
color qr codes it is!
 
for some reason with size 2 bytecodes my code only reads 29 characters before breaking
oh, nvm
am stupid
 
@mathcat If you did 64 colors, you'd still need a 1k×1k code, so it's not super viable
But for gifs and images it totally is
 
hm
 
7:04 PM
hmm
 
Idea: A video compression algorithm that treats the video as a 3d image
I'd love to see how (in)effective JPEG compression is with a 3d image
 
oooh
 
7:19 PM
okay so I think that there should be 4 levels of error correction, being able to fix 1 error per 8, 16, 24 and 32 bits respectively
 
makes sense
 
the error correction mode gets the first 2 metadata bits
and because I haven't given anything a confusing name yet, the error correction modes get to be named I, X, T, and R respectively
 
what if those bits are corrupted
 
It'll still work
hamming codes, right?
 
Well, if the error correction level was itself error corrected
 
7:33 PM
oh wait
@Seggan errors can be detected, but not corrected (i think)
 
Hamming codes allow correcting one bit flip, and detecting two
Seggan's talking about the error correction level itself though, which is part of the code
So if the error correction level gets flipped, it would break things
 
that's what I meant
 
8:04 PM
...oh shit
good point
I am not currently error-correcting the metadata, or for that matter the message length
lemme go do that rq
 
 
2 hours later…
10:29 PM
The bottom image is only 10 bits per pixel
And it looks so good
 
10:49 PM
wow
how did ya compress it
3 bit color? nah
 
YCbCr and some random fuzz added to everything so that you essentially get dithering
(The fuzz is added pre-rounding, so it won't ever decrease the quality, only increase it)
 
whats ycbcr
 
It's a different representation of color, it uses brightness and two chrome/hue components
You can compress the chroma a lot more than the brightness before it becomes noticeable
 
ah
 
@RadvylfPrograms Looks very nice as long as you don't zoom in
 
10:59 PM
Yeah. I'm hoping that with video it'll also have the sort of...time-dithering effect, where the pixels switch back and forth between values on either side and persistence of vision helps out
 
hmm
That could actually work
 
I've got a really neat idea for how to do the video compression
 
time to ANSI standardize it and patent it :P
ok thats kinda a contradiction
 
11:17 PM
do you mean ISO?
 

« first day (137 days earlier)      last day (592 days later) »