« first day (4169 days earlier)      last day (971 days later) » 
00:00 - 13:0013:00 - 00:00

13:00
@PyGamer0 im not sure if im understanding what im saying here, but; is there a way to pad out a 2d array to be rectangular here?
I'm really having fun here setting up Fig
real good
thank frick it's done
thank fig
Why implement your own stack instead of using the built in one?
13:14
is there a buiktin>
?
the built-in one is called a list lol
python lists have O(1) .append and .pop
alas, i feel that i'm probably going to add some stuff into it
@hyper-neutrino well, isn't it still O(1)
You can inherit from collections.UserList
although .pop errors if the list is empty with the built-in so I guess you might want that
So you get the build in list methods for free
Then you can add anything you need but don't need to reimplement things already there by default
13:15
I might want to rotate it or flip it anyhow... (not to say it's hard to do with builtin tho)
@NobodyNeedsNames well given that you use a python list and .append and .pop your stack's time complexity is the same as the built-in list's
also I am pretty sure python lists are O(1) append and pop
you should probably be using a length check for your pop function instead of catching the index error
Error checking is slow
anyways more replies and stuff do ping me because else I won't check everything
It needs to collect the entire backtrace
13:16
o/
@mousetail forgiveness>permission but bye anyways
@NobodyNeedsNames disagree there
@lyxal it took me 40 mins once to set up gradle 👀
by the way, Fig is a mess, so please dont mind too much
@hyper-neutrino does python have try catch? :P
oh im late, lol
dw i was kidding
@thejonymyster whython has an error handling operator lol
lolol
lang dev stuff: i was making a cellular automata / 2d language type thing, but in programming it i wrote myself a bunch of macros so that it was clearer what each rule / symbol was doing
but then i realized it might be more fun to make a lang for that instead
13:29
a lang for writing macros?
no for making 2d automata type things
3
interesting
a macro writing lang would be good but is well beyond my experience
lol ive said that 3 times already
you are very interested!
i'll have to make it more variable than what i have currently
/ flexible
13:30
C macros?
i'll share my code, i just added a huge comment that explains shit lol
(jk thats a link to my code)
this was for the specific 2d lang i was making
but the like, subturn command itself is what i kind of want to turn into a language of its own
where the syntax would be like the calls to that function
So you basically have javascript that generates more javascript code?
well it doesnt generate it, i dont think its a true macro or whatever
im using terms my friends said to me
im not a real programmer :P
Ah ok, a macro is usually code that generates more code
right that makes sense
13:33
Interesting concept though
ty :-) this is my cleanest code ever btw X_X
Hmm and I still can't really tell what anything does
Doesn't help that the repl.it IDE is kinda terrible
yeah lmao
it's funny my friend said it was "fine, other than the fact that it's almost completely unreadable"
(paraphrased)
I agree with your friend
yay! previously my code was regular bad
anyway if i end up getting tips for the actual coding ill just go to the firing range or whatever :P
13:37
If your code is unreadable enough nobody can spot any mistakes. I've been doing that for years
10
but yea in re: "javascript code creating more javascript code" thing,
i forgot what i was gonna say
bye :P
@mousetail They recently got an upgrade, went from Monaco to Codemirror
There wasn't much of a change with the Java autocompletion but at least it wasn't glitching anymore
Issue is that either the window is tiny or fullscreen. No inbetween
13:54
@mousetail wdym?
you can click and drag to scale
I don't think you can (?)
@thejonymyster wdym
@mousetail next to the scroll bar theres a thingy you can click and drag to make the window wider or thinner etc
if im understanding what you mean right :P
@PyGamer0 is there an element(?) for making all lists in a list be the same length via padding?
or is that not a relevant question
13:58
I don't see it
Ah yeah, it's annoying when you're looking at someone's repl and you want to see their files and console at the same time
But if you fork it, then you can resize it and stuff
@mousetail OH i always forget that exists
god im linking people to that all the time oops
im always assuming people look at it in the dev environment T_T
@mousetail WW made a haskell golfing library (hgl)
You shouldn't need to fork to read the code properly
that's just bad design
i guess what i was talking about was fullscreen mode yeah
14:00
@PyGamer0 Cool, I'll check it out for inspiration
yea repl is kinda terrible :P i'll install vscode or something soon i just like being able to start n scrap shit at the drop of a hat
There's better online IDEs for working but you probably want to publish to github/gitlab in the end to make it easier for people to see your code and submit issues and comments
For a javascript project you can make it interactive via github pages/gitlab pages/vercel/netlify
i'll look into that :D
@mousetail well, if you want to have a shorter time, better use python and not some golflang IMO, idc
o/
?
Golfing languages usually have no error handling since a error would be the option to do something else useful
14:05
@mousetail also cloudflare pages
@thejonymyster github.com/features/codespaces might be of interest
You can also rent my old laptop for 10x that price
>price
~_~;
but my m$ney....
@user I got confused for a second there because I was like "hey wait we never used Monaco" and then realised you mean replit
What's a linter?
14:17
Checks your code for common errors and formats it to match standards
think a PEP-8 format checker in the case of Python
For JS typical choice is ESlint + AirBnB
I use Prettier for JS/HTML/CSS lol
Prettier is more of a formatter than a linter
ah prettier ive heard of :P
idk ive got weird ideas about whats readable for me sometimes
i guess i could always step in and undo things it does that i dont like
like separating my code from my mid-code block-comments
14:31
CMQ: If you could change the name of any esolang, which esolang name would you change and what would you change it to?
@lyxal 05AB1E to "ReadableName"
@lyxal brainf**k to brainfrick
for obvious family friendly reasons :P
@lyxal Vyxal to Veggan
oh wow. i was looking through the selected technological data in the arctic code vault and apparently they decided Clojure and Erlang were important enough to warrant entire book-length specifications stored in the vault
5
lol they stuck Shakespeare in there too
14:46
@lyxal /// back to /// if the author changes the name :P
@Seggan the lang or the literature?
or the bard himself?
@thejonymyster all the guy's works
genuinely wasnt sure since you were talking about langs lol
that makes more sense
CMQ: what does the M in CMQ stand for
mini
what makes it minier than a regular question
is it the big bold CMQ label
it's by analogy with CMC, which is a mini challenge because it's not being posted as a challenge on the main site
14:55
ahh
0
Q: Compute the cophenetic correlation coefficient

not-a-featureCompute the cophenetic correlation coefficient Background The cophenetic correlation coefficient is used in bioinformatic to measure how close the length of the branches of a tree are to the distances of the original distance matrix. It is defined as follows: \${\displaystyle c={\frac {\sum _{i <...

@pxeger oh nice
15:43
@lyxal That makes sense, but I think it'd be better to do it within the language rather than a flag. Like APL's ⎕IO
@Seggan I mean, the arctic code vault also includes a specification for Unsure, which is of course the #1 most important language to save in the event humanity is at risk
Wait no it was plumber, even worse lol
is there any consensus on using languages that never halt by design?
You'd have to look at the I/O rules
If it outputs something then just continues whirring away for the rest of time it's probably still valid
well, programs must terminate by default seems to be winning as a general rule
but i feel like the rule might be different in a lang without halting capabilites :P
this isn't even hypothetical btw, i was thinking of using bytebeat for some audio output challenges
it gives you a variable t which represents the time state of the program (constantly ticking upward) and the "output" is the tone(s) produced by the expression
i dont know how to explain it good enough to be accurate but its basically that
it doesn't halt on error even, it just doesnt play audio whenever the expression is invalid
so like t%2 doesn't halt, it just isn't playing half of the time
you could definitely make it guaranteed to always be invalid after a certain t though
t>n?1/0:1
15:58
also Vim 9 has been released
16:13
Hello guys!
I have a question about K!
Does somebody knows how to "put parameters" when running a K program with ngn ?
Something like: rlwrap ./k example.k < example.txt
what do you mean by parameters? command-line arguments can just be used normally AFAIK
0
Q: What is our consensus on languages which do not halt by design?

thejonymysterCurrent consensus seems to be that programs must terminate by default. This makes sense in general. However, does this bar use of languages which are not designed with a halt state (other than manually stopping the program)? For example, take something like ByteBeat, which I believe is a family o...

Yes of course!
But for example, let's say I write a script in some language, and I want to run a K program that takes data from the STDIN. My question was; is it possible to make a K program run as if there was already STDIN.
For example in some languages you can just do ```
py main.py < input.txt
```
and normally it works. However I tried this in K and it doesn't seem to work.
I'm not very clear sorry
@TKirishima You might want to also ask in the k tree and The APL Orchard
Thank you ! :)!
16:23
the < file syntax should be independent of what command you're running on the left side - does the program work when you run it without the < file and just input into STDIN manually?
Yes it works whitout the < file
In what respect is it not working?
For example:
My basic program is:
a: 0:`
`0:a
So when I execute it, I must enter something and after it, it "prints" it.
But when I do the same thing with < file
It's still ask me for input, but the input should already be "given by the file".
And the file isn't empty
Have you tried without rlwrap?
Still doing the same thing :/
16:31
why not read the file directly
What do you mean ?
Like that I execute the K program than I put the content of the file ?
/example.k
contents:0:"example.txt"
Oh thank's a lot it works
17:18
I got third in some maths competition! (again lol somehow I can never get better than third place)
@mathcat why you get third place: 1 2
huh true
GG mathcat!!
thanks!
@PyGamer0 but this time it was due to my lack of knowledge in combinatorics
@PyGamer0 This!
17:24
@lyxal APL to Apple
@DLosc same. When I was talking about brainf*** with a teacher, I had to say "brain, and then the f-word"
Ironically brainfuck is a pretty normal language. My brain hurts more after javascript
i think javascript is more vulgar than brainfuck
yeah i think itd be worth changing, not so much for censorship reasons but because it's pretty tame lol
@mousetail Depends what you're trying to do, though, right? I mean console.log(42) is a lot simpler than whatever the equivalent is in BF. :P
Waiting for someone to use the fact that the last second of 1969 is falsey to golf something
17:31
I wouldn't say JS is more painstaking than BF, but I can't understand why it's so popular. I've met people who didn't know anything about esolangs but could program in bf. I'd say <><, Forth and even Japt are more "estoric"
BF is extremely estoric, but super popular for some reason
@DLosc At least in brainfuck you can easily visualize the entire state of the program at any point. In JS when there are 15 async promises pending and a complex network of objects it's hard to reason about. I effecitvely run out of mental RAM.
It's still language design workshop, right?
I have a new language to present: Chocolate
Not near done, but functional
It's written in Scala, but you can use the .jar file in the root
17:33
Is there a list of commands?
Not yet
Let me compile one now
It doesn't have very many yet
It's a similar syntax to Fig and Pip and Pyth, prefix functions
@mousetail True. BF is undeniably simpler.
@mathcat It's popular because of this little thing called the web browser :p
Node/Deno for back-end stuff and electron for desktop stuff is a relatively new phenomenon
Like, nobody's out there writing in Actionscript or Lua because they just have a deep love for those languages, and it was the same with JS until everybody got stockholm syndrome
oh right
But JS has both come a long way, and it's been around long enough that people have gotten used to working around its flaws
17:37
that's why SOs survey looks like this
Ironically just like its namesake
Too bad VBScript never caught on
@RadvylfPrograms I have nostalgia for ActionScript, does that count?
I doubt many people just truly enjoy using Java, but it's common enough that people accept the pitfalls of it and OOP as pretty normal, and so it seems like a baseline
@DLosc Depends, I guess. But you probably don't write desktop apps or server back-ends in Actionscript :p
No. I would like to write little games in it if I still could, though.
@mathcat I've been a python guy since I was born and to see that SQL's more popular than python is enraging me.
DROP SQL
17:40
SQL's not really in the same category, similar to HTML/CSS
Like, they don't really compete, SQL's rather specific in scope
@RadvylfPrograms I tried Node once and it just felt wrong to be writing non-browser JS.
But yeah SQL and burn and die
^^^
someday I'm going to make a better SQL
IDK why people dislike SQL
17:41
ALL CAPS.
I liked Node at first, but now it kinda bothers me ever since I realized just how fragile its good performance is
it looks weird
@mathcat Not required
You can use lowercase if you want
(in SQL)
yeah but most people do it
17:42
@mathcat That's a feature, not a bug. :P #qbasicgang
@DLosc I hate it because it's just not how I think. Complex SQL queries just feel so unnatural to me.
The syntax is also blegh and doing simple things requires way too much complexity and potential for human error
> For readability, all SQL commands should be written in uppercase letters.
All caps keywords with lowercase variables is the earliest version of syntax highlighting.
@mathcat - idiot dev
@DLosc hmm makes sense
Why not do it backwards select * from MYTABLE
lol
17:44
@lyxal you have heard of dihydrogen monoxide toxicity right?
^^^|^^
@Steffan I guess because a command should be shouted. :P
I guess it's called a command :P
But then SQL shouldn't error. It should do what it's told no matter what
"I'm sorry, Dave. I can't do that."
But if you don't shout, it will error at the tiniest mistake
And it will shout at you
17:47
@Neil what? My grandma kept saying brown apples don't have iron, so it has to be correct.
"My grandma said so, so it has to be correct."
exactly
Fun fact: You could buy a refurbished server that stores the entirety of Wikipedia, plus pictures, for about $1500
huh
somebody told me Wikipedia can be stored in an SD card.
18:02
Without images yeah. I actually have a Wikipedia SD card a few feet from me.
Jun 18, 2021 at 17:13, by Redwolf Programs
You can fit all of Wikipedia on a micro SD card. I doubt SO's running out of storage space any time soon :p
@RadvylfPrograms uff images
I'm starting to hate images/videos
It's about 4000 times bigger when you include images
^
I tried to design a sort-of-qr-code which can fit images/videos
Videos can compress surprisingly small, but yeah they're massive
without error correction my design can store 11.619 kb
18:05
@RadvylfPrograms Does this mean we need a new number for "A picture is worth 1000 words"?
Wouldn't a QR code that can fit a image just be a image?
You can't really compress it more
@mousetail ups lol video then
Still, you are not going to get any more compact that just pictures of every frame next to eachother
I used ffmpeg to get a somewhat watchable 1m video to fit in 1 MB once
unless you are going to require a microfishe scanner to read it
18:06
@mousetail QR codes aren't exactly intended for compression lol
Yes, QR codes are designed to be highly redundant and thus only store a very little bit of information over a large area
@mousetail I'd be somewhat surprised if compressing a video with something like MP4 and coverting it to a QR-code-like format wouldn't be smaller than putting every frame side by side
Thus are unsuitable for strong big things like images
yeah, a watchable video stored in a QR code would be nice
@RadvylfPrograms I can guarentee you the QR code would be bigger
Depends on the size you print the frames of course
18:08
@RadvylfPrograms what was the video resolution?
@mousetail With an actual QR code yeah, but "sort-of-qr-code" could include a lot of changes
You could get close
E.g., making every pixel one of 64 colors would encode a lot of additional information, but still not be too hard to parse
My goal would be to manage to fit a gif in a QR code, which could be possible
@RadvylfPrograms Yes, but at that point a phone wouldn't be able to scan in accuratly
18:09
@mathcat 90p, 10 fps
@mousetail It can't do that on a bunch of images next to one another either
@RadvylfPrograms ah okay
No, but the mistakes will be less notacable
Idea: a 2D image-based language like Piet, except instead of abstract art it's designed to look like a QR code
The thing with compression is that small mistakes can lead to unrecognizable results
@RadvylfPrograms that was my idea too, I wanted to start with 8 colors
18:11
We can assume somewhat optimal conditions though, and use some simple error correction which wouldn't cost much
Maybe
There's a trade off
accuracy vs compression vs how ideal the situation needs to be to scan it
If you're trying to stuff a video into a scannable 2d code you probably aren't focusing on practicality all that much
I'd imagine most phone cameras nowadays could handle a decently high res grid of 64 colors though
Remember a roll of film is tiny and also can hold a entire movie
with 64 colors you can get 17.718 kb
The Google Pixel 7 is supposed to have a 48 MP rear camera, that's a lot of pixels
18:13
I think the project would only be succesful if it can beat a roll of film for either quality or size
You'd waste most of them of course, but if we assume 10% contribute useful data, that's still a lot of data
and maybe shapes could work too
square+square with hole maybe?
Even a million pixels with 64 colors each is 0.75 MB, not quite enough for a minute long video, but easily sufficient for a GIF
@mathcat Not sure. You'd add a single bit of data doing that, but at the cost of it potentially being scanned as a darker shade instead of a different shape
So probably not worthwhile since you need at least quadruple the number of pixels to identify shapes vs. just colors
ah well
JAB code already exists, but it doesn't focus on storage space (that much)
btw the storage calculations I'm doing are with the size of QR code 40
 
4 hours later…
22:05
how does j-uby work? i have some ruby code that i want to convert to juby but its too confusing for me
J-uby is just Ruby with some extra methods defined
If you have Ruby code, it should just work
Converting J code to J-uby is harder
J-uby LoTM died quickly
Because bountyn't
ye ik but i want to make ruby code shorter
Oh ok
What's the code you have?
@emanresuA ye well im trying to get a juby answer up and going and i have some ruby code working, i just want to use jubys extra functionalities to make it shorter
22:08
You can look at the GitHub repo, it shows you the rules
@user literally cant understand it at all
also just learned the basics of ruby so my code suuuuuuuucks lol
puts ('!'..'~').group_by{|s|s.ord/16}.map{|h,*t|t*""}
gets.chomp()
@emanresuA J-uby has the same problem for me that Curry did: it feels like I need to know Ruby pretty well first, and I can't be bothered to first learn Ruby and then learn something else on top of it. It looks cool, but not that cool. :P
22:09
@DLosc yeah i feel same way
@AidenChow The docs say F * x == x.map(&F), so maybe you can use that to replace the map?
Oh wait nvm
what is &F ?
I've been doing some analysis on LoTMs and bounty usage, and my current results show that LoTMs with bounties on average get 40% more answers than those without.
A reference to that function? idk lol
Another overloaded operator?
22:11
No
i think the & has smth to do with: stackoverflow.com/questions/14881125/…
not sure though
att
att
I have no idea what the distinction between block/proc is useful for though
@att yeah perhaps
i feel like j-uby lotm just flopped becuz so little ppl understand ruby enough to be able to apply it to j-uby
Which is kinda sad because it must've taken a lot of effort to get Ruby to behave like J
I tried writing answers a couple times but I know neither Ruby nor J lol
att
att
come golf mathematica (:
22:27
@AidenChow You might want to try out each_slice btw
@AidenChow Bit shorter
att
att
@user aiden's first line has 15 characters
Oh, whoops
22:48
@att dont u have to pay to use that???
att
att
usually yeah
but it's free to use on tio and wolfram cloud
@user wait that works if its ' '..'~', my original implementation was a bit off
was trying to find a builtin that split arrays like that, but didnt manage to find one lol
@att oh yeah thats cool, but theres like a million builtins to keep track of right
seems daunting
att
att
that's part of the fun of it ;)
23:10
Does anyone know of bf programs that take extremely long (but not infinite) to run?
Isn't that most bf programs?
14
Q: Busy Brain Beaver

Anton GolovWrite a brainfuck program of no more than 256 characters that takes as many steps as possible, but does not loop infinitely. The program may not take any input. More specifically: Assume an infinite number of cells to the right. A < when at the leftmost cell does nothing. A - when cell value ...

As in, busy beavers?
ninja'd
The problem with that is that my interpreter has fixed width of 30000 cells and doesn't allow going out of the tape
32
Q: The cell at the end of the rainbow

Jo KingWe all know of, or at least have heard of, brainfuck, one of the most famous and influential esoteric languages. While these days most implementations use an infinite tape, the original compiler by Urban Müller had a tape of 30000 cells. A little known fact* is that there is actually a special va...

23:14
IIRC that one finishes almost instantly
23:50
yeah, i've got a better busy beaver somewhere on this site
the biggest number under 100 bytes question I think?
yup, that's it
definitely won't work with the fixed width cells though
@NobodyNeedsNames Please stop requesting access to the Bakery. The entire point is that the only people who have access are the ROs/mods, so that people don't get pinged when their messages are moved there
00:00 - 13:0013:00 - 00:00

« first day (4169 days earlier)      last day (971 days later) »