« first day (4990 days earlier)      last day (149 days later) » 

00:07
I think I'm about to get like a 30 on a chem exam. I'm switching to a creative writing major.
00:25
unique choice of switching subjects but you do you
lol
700 days is way too long
stackexchange loves resetting the counter
two years of code golfing sheesh
nervous sweating noises
00:27
...I am astounded
I once managed 1000 consecutive days
h o w
That's sick
don't you go on vacation or something
(in more ways than one)
I check this site almost every day on my way to school and stuff
00:28
wOah
covid killed the streak
I think I want to start golfing in more different languages. Like I really want to try Jelly for example
if anything I thought covid would make the streak continue
got up to 1072 days in a row
@Redz not checking your phone for a day will do that sometimes :p
(from being that sick that I couldn't do anything all day)
00:31
fair
@lyxal sigh this is why phones are inconvenient
I feel like the streak should be more days atm, but SE resets the counter for no reason
because my view streak on PLDI shouldn't be 3 days
given it's the first page I open every morning
I've been on that get-an-idea-for-a-programming-language/develop-its-syntax-a-little/write-half-a-parser/repeat wave recently
Maybe I should interact with PLDI more, I haven't written anything there since the original launch
(stares at the most recent post being in april) same
00:56
Anyone ever read the comment section on the New York Times Metropolitan Diary? It's bizarre
It's like a weird little subculture
There's drama
It's a very human place, very rarely has to do with the column itself, rather confusing to stumble across
Like that one random YT video where people post comments with "checkpoints" describing their lives
But more conversational
@noodleperson same
@noodleperson Ooooh is it unfreeze JHT time
01:42
@UnrelatedString Don't get ahead of yourself lol
I did read the Jelly tutorial at some point and I get the whole arity matching stuff and how the micro acts as like a pipe. So maybe I'll start using it
Though probably if I do I'll use the Jello thing code-report made
Oh yeah I forgot about that
How's that been going
Idk if it's been updated in a while but it's at a pretty cool state and seems fairly usable. I think I installed it at one point though idk if I've tried it out that much
idr, rather
02:38
Okay so apparently the fluoroelastomer used by the first party Apple bands is considered to be way more comfortable than actual silicone
But it's still so slick
And I think the spare band I remember seeing is just Not Here but surprisingly okay-looking Chinese options are like $7 on Amazon so may as well give that a shot
 
2 hours later…
04:18
@UnrelatedString normal silicon is mid compared to what you could have
at least, that's typical of the android watch market imo
e.g. the faux leather and fabric bands for galaxy watches are much better than the default band
 
2 hours later…
06:46
@Ginger done
07:04
Why there's no practical 2D programming language? (Esolang exist though)
The most simple advantage is that IF and ELSE can be almost equally treated
att
att
07:36
@UnrelatedString I've been meaning to learn it eventually™️
@UnrelatedString I know enough Jelly to mostly not use it because you and Jonathan almost always get there first and do it better
Most of my Jelly answers are ports of my Vyxal that happened to work out better anyway
(except for this where the Vyxal's a port of my Jelly)
07:57
@l4m2 because SBIS languages aren't popular for practical programming, for obvious reasons
the closest would be scratch and well... gestures vaguely at scratch
Is that because I used to do scratch golf? :p
no i have a long message to type hang on
@lyxal @Ginger i finally figured my biggest roadblock with learning vyxal and it's a big ask but i wanted to put it out there:

it'd be nice if the online interpreter had a good debugger, with the following features:
step-through mode: execute instruction by instruction at the press of a button, with the next (or previous?) element interpreted highlighted, maybe showing it's mouseover thingo
stack visualizer: when stepping, show the status of the stack after each instruction, and highlight what will be taken as arguments by the next (or again, previous?) instruction, if any. this should als
Topical :p
There was discussion about such a thing earlier today on the discord :p
08:11
oh lol
There's one in progress which @user was working on
Not finished yet, but would be integrated easily I imagine
att
att
08:26
@lyxal only because you don't anymore (:
 
2 hours later…
10:54
> Snapchat SQL.txt. Suggests a breach involving Snapchat, particularly data extracted via SQL queries.
I thought the whole point was that the snapchat developers didn't know how to sql
11:33
0
Q: Randomly color the words

JordanInspired by this X post. Write a program or function that takes a string as input and prints or otherwise renders the text with each word in a random color, e.g.: The input will be a string of printable ASCII characters. For our purposes a word is a contiguous sequence of non-whitespace charac...

12:03
@Themoonisacheese noted :p
the problem with implementing stuff like that is that @lyxal wants it to work on mobile and I'm kinda at the limit of how much stuff I can cram into the mobile UI
so it would probably be desktop-only
also: Vyxal is not designed in a way that makes a debugger easy, because it executes a program by recursively traversing the AST
it would be super cool if the interpreter could be restructured so it iterates over a list of instructions
@Ginger I feel like that shouldn't make it harder
Given tree transversal algorithms exist
@Ginger the debugger could be something I could live without on mobile
@lyxal sounds very GPT-ish
@lyxal honestly i'm unsure why an esolang for golfing even has a mobile UI in the first place but sure
12:27
@lyxal oh right lol
I forgor
extra bonus of doing that: it would make output work a bit better
@Themoonisacheese yeah I know. That's where I got the quote from :p
@Themoonisacheese *website
The online interpreter is responsive
for some definition of "responsive"
13:17
@emanresuA LMAO
you love to see it
you love to see it
jonathan's Jelly answer on 500 cigarettes is beautiful
I was also thinking of the overlap thing, but everything else about it is just, wow
which solve exactly the problem I was having
13:58
> sfr.fr.txt: appears to be related to a breach involving SFR, a major French telecommunications company.
> frsfr.txt. Unclear, may relate to a French service with the abbreviation “FRS.”
...or maybe "FR SFR"?
not even maybe
seems like the more obvious conclusion lol
SFR very commonly uses FRSFR for internal names
some of their 4G APNs are named like this, i think they even own and use frsfr.fr for internal use
I'm stupid
I just showed up to a class that I don't even have today
well this is a nice classroom so I'm just gonna stay here
14:04
we used to do this in uni
one of us had a switch and would bring their dock, and we would play SSBU in empty classrooms
I love how they just leave empty classrooms unlocked and you can kinda just do whatever
It's fun to go to the big lecture halls when they're empty and stand where the prof would and be like "hello class"
just got my "golf" to hit 1234 characters yippie
whoa, what was it originally?
ah beans, i golfed it down
14:16
opinions on this trick to put each switch block in its own scope?
0
A: Waiting girl's face

ThemoonisacheeseBash + tar, 1186 bytes echo "H4sIAFe99mYCA+1Za07bQBD2b5+CC7RKIAnnQZQeAFK1P2PjEDeijaMmQKsU7aNSUSOVoDaiRW0PMyfpzNhOA+ThEO1iVR6tTOzYmc/z+GZmeb6zu/e0/qruGJQSSq1Swb+b1Wpti843t2slvl4qlSvVcsUpV2rb1erW1nYVr5fLm7Was1FyLMiLg/rO/saGs3ewt/C+3Z1nT/ZfOv+bgGgYWD6ICGQECtcA1AiUBnUKuk9H2XDNaA1BNUCh7j7IHqg+L9Q3Bjk...

just data
i had generated the archive with defaults instead of GZIP=-9, so simply by adding that option i saved 50 chars
0
Q: BTD6 Kinetic Chaos (9/26/24 Odyssey)

pacman256This week's odyssey in Bloons Tower Defense 6 has a neat gimmick, but it's slightly confusing at first. Each custom round is actually two consecutive regular rounds at once, but this means the round number no longer helps you know what is coming! The rounds are as follows: 1 1,2 2 3,4 3 5,6 4 ...

14:36
(also please go upvote other answers than my own on 500 cigarettes i can't stand that i have the top spot with a bubblegum answer)
@rydwolf isn't this like, the canonical use-case for free-standing scope blocks?
oh is it?
i mean
the goal is to limit scope
fair enough
i actually can't tell what lanaguage this is
two brackets is all the secret service needed at that trump rally smh
4
@Themoonisacheese TypeScript
14:41
oh lol
this could almost be c#
i see now that you aren't declaring the types of your variables but at first glance it could be any of Java, C#, c++, python with braces and semicolons
@rydwolf that's how I've always done it :p
15:33
-1
A: BigNum Bakeoff Reboot

The Empty String PhotographerSimilar to my Silly answer Python 3, 512 bytes import sys from multiprocessing import* p=print t=sys.stdout sys.stdout=open("a","a") def b(l): a=[] for y in range(256**l): try: a.append(y.to_bytes((y.bit_length()+7)//8,'big').decode('utf-8')) except: pass return a for x in b(99**99*...

15:56
@RubenVerg seems like you forgot to commit server.py?
 
2 hours later…
17:40
Hot Milk is such a cursed band name
their song Glass Spiders is super good tho
18:07
just copying a link down for me to use later
psst: it spawns a total of nearly 10,000 sub processes using multiprocessing during the execution of the program.
 
2 hours later…
19:48
It never occurred to me that part of the advantage of having an advisor for research is that they help you with actually figuring out what to research
It's so nice just being told "this is a really neat gap in the research, you could look into it"
Meant void foo(void* p) { if(p) { auto [x,y] = bar(p); foo(x); foo(y); }} but save some stack from not using call
20:14
Fun fact: there are actually two US flag emoji
🇺🇸 and 🇺🇲
which I suppose means you could steganographically encode a small number of bits in a patriotic spamming of american flags
wait that could be an esolang
Every program is just AMERICA followed by zero or more of the two US flag emoji
20:44
@rydwolf the first one is United States and the second one is United States Minor Outlying Islands :p
yep :p
Unicode is a little silly
well, not Unicode
font vendors are a little silly
21:03
They need to make TX render as a texas flag
I'm pretty sure the chilean flag emoji gets more use as a janky texas flag than as an actual chilean flag
@rydwolf make your own font that does :p
@rydwolf My phone is therefore lying to me, saying they’re the exact same emojis. I wish my phone was more honest.
@Ginger yeah, apparently *.py is in .gitignore
7
i'll force-add it but i think you might want that changed
what
what was I thinking when I did that lmao
should i remove it?
21:16
yes :p
pushed
merged
thank you!
It's computationally hard to determine if two matrices are equivalent up to row and column permutation. But it's tempting to think you just have to sort by the rows and then by the columns to do that. . what is a counter example?
CMQ Its computationally hard to determine if two matrices are equivalent up to row and column permutation. But it's tempting to think you just have to sort by the rows and then by the columns to do that. . what is a counter example?
That doesn't work since you can end up with things in the same row and/or column that weren't orgiinally
E.g.:
2 1
0 3
will then appear the same as
1 2
0 3
21:32
If we start with the top matrix we get 0 3, 2 1 by sorting rows and then it is unchanged when we sort the columns
I am assuming lexicographic ordering
^
Not individually sorting the elements of each column
/row/whatever
ah mb
No sorry. Sort the rows, columns as if that are strings of digits
If things sort the same, they necessarily are related by row and column permutations, because the sort is a permutation
It's just the other way around that's ticky
@UnrelatedString yes. But do we miss some?
21:34
Given the computational difficulty, I assume so :P
when you sort by one of the dimensions you may unsort it by the other
We must do because it is hard to tell if two matrices are equivalent up to row and column permutation apparently
So yeah. What do we miss LMAO
Exactly!
OOOOH
Yeah that's probably key
21:35
@rydwolf yes
I just can't make an example. I was really thinking of binary matrices
@rydwolf what you say must be the key
server is still dead :(
mine?
yep
just texted my dad, I'll see if some time in the next week is doable for shipping the server
hooray!
this has been a very long four hours
@UnrelatedString ooh .. You just swapped the first and last row and column?
@UnrelatedString that's very cool!
Thanks
It helped to start with something that was already sorted, then think of how to unsort it in a way that wouldn't work
And the cool thing is it doesn't work no matter how many times you iterate it either
yeah you'll just get caught in a loop right
What you really need is a total ordering of matrices
@Bubbler My current project has some deps that are only on Anaconda and some that are only on PyPI. uv can't handle that, can it? Do I have to keep using an unholy mixture of conda and pip?
and an efficient way to find the lowest (or highest) matrix in that ordering which is equivalent up to row and column permutation
which sort of feels like kicking the can down the road
(For an example each way: cuML and mediapipe)
@rydwolf that would definitely be amazing
22:01
Not really, see ^^^
You can of course simplify some cases by sorting by something that doesn't care about the rows/columns being permuted, but that becomes very weak very fast
@Ginger bad news, it'll have to wait until October 13th
I'll be back in Austin then though, so I can get it taken care of pretty quickly instead of having to guide my dad through it
@rydwolf got it
23:07
0
A: Sandbox for Proposed Challenges

Daniel ScheplerImplement the bind operation from the list monad In the list monad, the bind operation is a function of polymorphic type list A -> (A -> list B) -> list B. The definition is that bind(x, f) takes each element of the list x, calls the function f on each one where f is expected to return a list, t...


« first day (4990 days earlier)      last day (149 days later) »