« first day (3815 days earlier)      last day (1031 days later) » 

12:10 AM
@cairdcoinheringaahing I hate to be a party pooper, but tbf this question already had lots of answers (and in the same lang, since FRACTRAN is the only one allowed). It's the golfiest yet, but it's far from the first. See this comment!
To be fair, it's much longer. But that answer also got a bounty of +500
Correction, that answer was posted on SO with the [code-golf] tag and has since been closed. It's odd that it was never ported to CGCC since it has the tag. Anyone know why?
 
they didn't just auto-port all code golf questions over here
any ones that were migrated were done so manually
 
Oh really? That's interesting
 
Interesting according to this question I would be XS, but whenever I buy clothes in Norway they are always M
 
@hyper-neutrino In case they didn't fit our community standards or quality?
 
well no, there just is no automated system to port a tag to another site
it's not too common for sites to come into existence because they were broken off another site
but also cuz those questions have historical context on SO and CG is meant to be its own site, not just "we dumped these things from SO over there because we felt like it" (even though that kinda is what happened originally)
 
 
1 hour later…
1:24 AM
My custom styling is almost finished and it looks absolutely amazing
I still have to steal take inspiration from a few more parts of GH's UI first, then I'll post some screenshots
 
1:48 AM
@RedwolfPrograms that's a rather unfortunate typo
 
shall I 11 it for you :p
 
Can you 11 that into the correct vowel please :p
Thanks
 
*changes it to an a*
 
I blame Christopher Latham Sholes
 
1:50 AM
who now
 
The inventor of qwerty, of course :p
 
ah :p
yeah qwerty kinda isn't great - i'm sure it served its purpose of not jamming typewriters fine, but yeah i just can't be bothered to learn like dvorak or colemak or whatever
also IBM type ball OP anyway :p
 
I mostly use laptops, so if I learned dvorak I'd have to deal with the annoyance of the keys not matching their functions
 
I generally feel like getting comfortable or at least used to out of the box defaults is pretty handy, so I don't bother with alternative kb layouts but also things like relying on specific ide features/plugins
 
1:53 AM
Vim has keybindings for most editors and IDEs, but...Vim
 
unrelated, but i'm actually quite proud of the search interface i'm been working on for looking up jelly built-ins and stuff (the data entries aren't complete, but hopefully the frontend html and js are)
 
ooh nice
 
^ Looks pretty professional
 
once i'm reasonably satisfied with the core of it overall i may port it over the vyxal too :p
@user heh. materialize css OP
 
Imma steal your source code next time I make something :P
 
1:55 AM
Okay, I'm not finished yet but I can't resist showing y'all a screenshot of my current Text styling:
 
lol. there isn't that much to it (although i appreciate the compliments :p) it's just that materialize makes things look really nice and you just need to add a few classes to your html and structure it right
 
the only problem is because i have so much content, on mobile it doesn't fit properly and overflows or wraps onto like 17 lines
but if you're golfing jelly on mobile, you have issues /s
 
Meh, only caird golfs Jelly on mobile
Ninja'd :P
 
@RedwolfPrograms why so much green? or is this green not the type of green you hate :p
 
1:57 AM
I can't (currently) pick how CM categorizes things, so if I say that one thing is green a lot of other things end up green too :p
But it's a nice green
 
I've made some massive changes to the appearance though, I'll try a screenshot without this enabled to show y'all how different it is
Whoa, I just looked at the JS and MD files and they look beautiful
Especially for markdown, the old styling had some questionable choices
 
tmw your neural network is about as useful as a coin toss
 
also what's with the obsession with dark red/pink text in dark themes like that's just barely within "readable"
 
And that shade of yellow that hurts your eyes
 
2:06 AM
i should try styling my atom sometime, i assume i can given how customizable it is
@user yeah i've never liked that color
 
If I can style Text then you can style Atom :p
(It's not easy, there's exactly three officially supported themes and you have no way to change them)
 
Atom's pretty much meant to be hackable, which is pretty great
CMQ: Easiest editor/IDE to customize?
 
yeah it's nice
@user one that you code yourself
actually knowing how bad my code is, probably untrue :p
 
@hyper-neutrino lol, this is the extent of my UX skills
 
My eyes need medicine
 
2:09 AM
@user codemirror
@RedwolfPrograms iMedicine... The latest breakthrough in the medical field by Apple. (pills not included because we assume you already have them)
 
@RedwolfPrograms Does your synesthesia make it better or worse?
 
@user Neither, it doesn't work like that :p
And I think all of the parts of my brain responsible for color processing have shriveled up now
 
I actually had an even uglier version of it, gimme a minute to dig it up
I remember working very hard to make the alignment ridiculous and make the letters just slightly tilted
 
@user what's your plan for Vyxamples? How do you suggest we go about filling out idioms?
 
thoughts on just merging idioms into a part of my search system?
 
2:14 AM
I was thinking of an array of objects like
{
code: "lyxal",
description: "rickroll yourself",
tags: "rickroll roll",
link: lyxal.pythonanywhere.com...,
}
@hyper-neutrino +1
 
this it what it currently looks like for my jelly thing; it could be adapted to another language without too much work probably
and including idioms is just a matter of creating another section/category
 
@hyper-neutrino yes
Also, TIL I'm 7 upvotes away from code golf gold
 
pog
how many do you plan to get
 
3
 
hi everyone
 
2:25 AM
Howdy!
Dang. I only got one gold. Oh well, you can't break a system twice lol
Still
Very epic and cool
:)
 
Nice!!!
 
;-)
 
Also, my ploy to get upvotes worked :p
 
yeah always when I say I am x votes away from something
i get those votes
 
It's a path to either many upvotes or many downvotes, depending on how much I like you
I think I've made all of the changes I can to Text without getting into the HTML and JS. It now looks 10 times better in my opinion.
Anyone want a copy of the styles?
(You have to manually apply them until I package it up)
 
2:40 AM
this is only a Text thing (and so only applies if i have a chromebook) right?
 
Yes
Although I think you can get Text on other OSes, not sure if that's been deprecated yet
Uh oh
It makes HTML look quite ugly
 
@user Notepad (Microsoft)
pretty easy to change the settings
 
I'm going to have to change some settings regarding how HTML is parsed :|
 
true, it's technically easy to customize notepad
you just have like nothing to customize :p
i'm glad notepad is finally able to backspace whole words now
i still remember back when ctrl-bksp would insert some weird rectangle
 
@hyper-neutrino thats what i wanted to say
 
2:45 AM
@hyper-neutrino it does something even weirder in the dyalog IDE :/
yay, dyalog, sane as always
 
uh oh
what does it do
 
but you can even enable dark mode in notepad
 
@wasif where?
 
lol wut
 
2:47 AM
@rak1507 can confirm that happens
 
@lyxal from windows registry
I saw how to do it
but cannot remember now
 
Text is very customizable, you just have to download it, completely rewrite the entire CSS file from scratch, and try to figure out how to package it back up and install it
 
@RedwolfPrograms what is Text?
 
The default editor on Chrome OS
 
oh
 
2:52 AM
It's basically notepad with syntax highlighting :p
 
gedit is same too
 
3:04 AM
"Jelly can be typed on the ENG-US INTL keyboard layout" they say.
Well they obviously didn't account for my alt-gr key not doing its job
 
i can't get the over/under dots to work for some reason
so i just removed US INTL cuz i already have four keyboard layouts and i can't remove the two i don't need because windows is stupid
 
@hyper-neutrino same here
Supposedly, alt-gr+. then any letter should work
 
yeah, that's what the thing i found online said, but ¯\_(ツ)_/¯
 
Hmm, it seems the contents of extensions are signed. I wonder if it'd not yell at me if I just deleted the signatures :p
 
... sounds sus
 
3:17 AM
Okay, it seems I actually can just change the manifest to not check :p
 
I think it's just for checking the integrity of installed extensions, so if I make a clone of it without the signatures it doesn't know to verify anything
Got to go for a half hour or so, when I get back I'll try installing it so the changes will be permanent
Very suspense!
 
4:12 AM
It works!
I called it TWCS ("Text With Custom Styling"), and made the icon the same but hue shifted to blue.
I'll pack it and put a download here, if anyone wants to (and can) try it out
 
I am an Atom user, but I don't have my Atom-capable computer at the moment :p
 
spacemacs user :)
 
1.1.0 will have some custom JS as well, to fix the styling issues with HTML
The inspiration for this project is Text's default regexp color being so horrifically ugly it was hindering my ability to use them
 
4:51 AM
@cairdcoinheringaahing Jelly, 3 bytes: ıæ*
 
yoo wow
nice site
 
thanks :D
 
@RedwolfPrograms flags are love, flags are life
@RedwolfPrograms if they weren't flags, they'd be different github branches
flags just automate the process
 
That just makes it so much worse
It takes it from "a byte saving feature" to "the closest I can get to metagolfscript without getting downvoted"
 
Except it isn't MGS. There are two kinds of flags: output modifying flags and flags that would otherwise require a major change in the Vyxal specs
e.g. the rflag doesn't add anything...it just makes dyads/triads take their arguments in reverse
 
5:05 AM
Which is something that cheats around a problem other golfing languages have to deal with
 
which flags do you have the most problem with?
have you ever even looked at the flag list?
 
For example, with one monad and one dyad, there's three possible combinations (f(g(x,y)), g(f(x),y), g(x,f(y)) but only two ways to arrange them, so you need three bytes for one of them. With r, you can do them all in two bytes.
@lyxal I have a problem with all of them
Any flag which isn't something like jelly's e/u/n is automatically bad in my opinion
 
except that flags make the submissions different languages
so Vyxal r is different to Vyxal
 
Sure, but that's only really a workaround to allow flags without overcomplicating things. Nobody would actually tell you Vyxal and Vyxal with 100 at the top of the stack are two totally different languages.
In my opinion, languages which use flags for added golfiness are abusing a rule added in good faith to allow more languages to compete without silly penalties.
 
@RedwolfPrograms but what if I create a separate branch for each and every flag combination?
then they are seperate versions
 
5:11 AM
And then it goes from "I dislike Vyxal" to "I will downvote every answer using one of the weird variants"
 
you really would waste 500 rep on downvoting?
 
No, because I trust you wouldn't actually do something as extreme as that
 
because there are roughly 500 Vyxal answers on the site
and really, flags are the weird variants
the good majority use flags
 
@lyxal Well I would be more or less fine with those because they weren't written with the knowledge it'd be cheating in the future
My opinion on flags (the way they're used for Vyxal) is that it's just a dirty trick to save bytes, in the same spirit as writing a language for a specific challenge. Sure, it's not technically any advantage because of the scoring rules, but it just feels cheap.
 
but the flags (except the H flag) aren't specifically for any challenge
they are for general purpose usage
 
5:17 AM
@RedwolfPrograms (My opinion on Vyxal minus the flags is positive; I can tell you put a lot of work into it and I'm glad it's gorwing in usage and inspiring people to become interesting in golfing languages)
@lyxal Sure, but you could say that about MGS too
I'm not saying the flags are meant for a specific challenge, I'm saying they're in the same spirit as doing so
 
MGS is specifically "any challenge in 0 bytes"
flags are "usually -1 or -2 for 99% of challenges"
 
MGS is a much more extreme example, yes. But it's the same sort of thing, abusing the information not part of the byte count in order to improve the score.
 
My personal opinion has always been that I/O flags are fine (I put "join on newlines" into the footer in most Jelly answers where that's required anyway, so I don't see a problem with that), and logic/implicit flags are alright (like r, since it flips all operators and still requires the right program logic but without being hindered by too many swaps), but I don't personally like postprocessing flags that change the actual nature of the data.
Based on my personal opinion and consensus I still maintain that Vyxal's usage of flags is legal though, just that there are a couple I'm less certain about and a couple I still dislike.
 
Things like the join on newlines are in a sort of gray area in my opinion; for challenges like FB I think it's cheating, but for challenges less strict about I/O it's 100% okay IMO
r initially seems like mostly a convenience thing, but there's a lot of situations I can think of where it saves bytes and removes a lot of the hard decisions a golfing language designed would ordinarily have to make
 
well, I suppose decisions about which way operations face would be reduced by r (for example, in Jelly, "x occurs in y" and "index of y in x" take their needle/haystack in inverse order for a specific reason), but since r flips everything it feels less OP than i initially feel
actually, idk if there are a lot of vyxal operators that autoswap their arguments based on type overloading; that might make it a bit stronger. ¯\_(ツ)_/¯
 
5:28 AM
It's not as good for longer programs, but it adds something like 50% more viable two byte programs
 
@hyper-neutrino generally, operators won't swap their arguments - they try and just use types in the way they come if possible
 
eh. fair enough. granted, if vyxal (or any language really) can solve a challenge in 2 bytes, it's probably not that interesting (exceptions of course exist)
 
But then, if the r flag really didn't save many bytes, then why have it at all? The point of it is obviously to save bytes, and doing so with flags is a bit unsportsmanlike in my opinion.
 
what's your opinion on flags that change the behavior of implicit range (0 or 1 up to n-1 or n)
on one hand it's saving bytes using flags, but on the other hand, which functionality a language chooses is arbitrary and wasting bytes to do it implicitly doesn't make a solution more or less interesting
 
Still mildly cheating imo, but doesn't seem particularly useful
Making the solution more interesting by removing boring stuff like zero vs. one indexing is the challenge author's job, not the language designer's
 
5:37 AM
well when it comes to indexing restrictions, sure. actually i can't think of an example where changing autoranging would've really helped. so i think you're right, it'd probably not actually be really useful
ok so yuno also has a good amount of flags but other than the aforementioned implicit range behavior flags, two indexing flags, and output formatters, the rest are all debugging or performance related flags :p
like legit most of my flags aren't useful for golfing, they're useful for me to develop the language
 
@hyper-neutrino imagine not having auto-run on your online interpreter
 
meh. i could implement that if i wanted to ¯\_(ツ)_/¯
the original reason i didn't was cuz back when my interpreter was written in JS, if it auto-ran, i would lose my program if i accidentally caused an infinite loop and needed to close the tab
 
is pythonanywhere.com not working for anyone else?
 
@lyxal reproduced
 
5:52 AM
:58581472 heh, stoopid. I know there's no status page
 
My site's status page is the same as the actual site. If the site is working correctly, it'll return 200.
 
6:50 AM
@RedwolfPrograms I can't remember. Did anything ever come of the settings?
 
Not yet, I forgot to finish writing the seaching code :p
Will try tomorrow
 
How late is it over there?
 
@RedwolfPrograms No worries & no rush at all, honestly! Just couldn't remember & checking in
 
7:59 AM
@hyper-neutrino your "explain this code" button opens a new page that is 404 not found
 
 
1 hour later…
9:11 AM
@lyxal ah at least you're there
The Vyxal chatroom's empty
 
9:24 AM
@math did you notice how I have a new gold badge?
 
@lyxal no
 
@math have you noticed now?
 
yes
did you get code-golf?
or Fanatic?
 
code golf
I already had fanatic
 
wow 200 code-golf posts?
 
9:26 AM
I'm such a fanatic that I've visited 694 days in a row
 
hyper-fanatic
 
@math I have 418
 
I had a challenge idea recently; A cops and robbers challenge, it's change a character the code will print something else (eg. "Code Golf"), but the cop should make the code hard to read that the robber can't find the character he should change. Is it good?
 
9:58 AM
can't they just bruteforce it
 
@rak1507 what do you mean by that?
 
just try changing all the characters
 
@rak1507 to every charcacter that every existed in unicode, ascii and ...?:P
 
fair point
 
But if you try to understand the code, you might get it
so, is it a good idea?
 
10:03 AM
I'm not convinced, could sandbox it and see what other people think
 
okay
but I'll post the cops and robbers thread combined?
@rak1507 I'm not convinced too...
 
who is 2x-1 (or was i think he left)
 
@2x-1 who are you?
 
@2x-1 yeah tell
 
What?
no chance
 
10:13 AM
did he go to codidact?
whatever it is called
cokidact?
codikact?
 
codidact
why is everyone going to codidact? because of the community ads or monica?
 
@math the latter probably
 
I'm staying in SE for now. I can't do multitasking... If nobody's here anymore, then I'm leaving too
 
@math no one is asking you to leave :p
 
yes, but like if we are 1 question per day
 
10:23 AM
jht -> jelly hyper training -> jelly hyper neutrino
 
@lyxal 1561 for me
 
10:48 AM
@math but actually no
2x-1 had an account on both here and codi
Their account here was last active a few months ago
While on codi, their last activity was 8 months ago
@PyGamer0 a code golf user that I had the privilege of creating several esolangs with
@math don't be so surprised...they deleted their account and made a new one about three or four times
In total, 2x-1 would probably have around 16k-20k rep
@math (and @PyGamer0) read chat.stackexchange.com/rooms/240/conversation/howdy-world for a glimpse of 2x-1
Oh actually don't
The messages in that bookmark seemed to be moved
Let me see if I can find them
Found it
Wonderful
 
Has anyone on this site used pure typescript before?
A guy on discord I was talking to sent me a gif of an XML parser he made using only the TS type system cdn.discordapp.com/attachments/458319146921230357/…
 
Fuck me
 
I see potential for this to become the TS version of JSfuck
what
 
yup
why would you do this to yourself
 
oh nice :)
imma go read that
 
11:02 AM
When designing a language, would you not pick things very carefully to ensure you don't let it become undecideable?
 
Evil idea of the day: Language has syntactic ambiguity built into it. When there are multiple parses that compile, one is chosen at random. At runtime.
Also, it's really hard to write something that's not ambiguous.
 
that's actually hilarious and I'm going to work on that immediately
 
Why bother making a language decidable?
 
...so you can type-check it?
 
why bother type-checking at compile time?
 
11:07 AM
@pxeger I don't think I will
 
Technically still type-checkable if it's undecidable, just not quickly.
 
@JohnDvorak that's a whole other kettle of fish
@ATaco I just think I wouldn't want people to be able to abuse my language like that
 
(Types are a hang-on anyway, real MVPs use typeless languages)
We have very different language design ideals.
I wrote Funky2 to be as abuse-able as possible.
 
@ATaco 4-byte words are all you need!
 
4 bytes? What sort of programming are you doing that needs more than 1 byte?
 
11:09 AM
@ATaco link?
 
Although the TIO interpreter is quite a bit behind.
And I still need to flesh out most of the wiki
 
@lyxal thanks for the 4 messages in my inbox
 
@math yes.
 
11:29 AM
CMC split a text file by lines into two files x% of the lines in file one and (100-x)% into file two.
rounding to an integer of course :)
 
 
1 hour later…
12:53 PM
My quick take on vyxal flags: H: blatant cheating d/C/l/G/g/Ṫ: cheating M/m: cheating if your program fails without them L: probably cheating s: cheating except to convert character array to string j/S/J/ṡ: cheating if the question requires formatted output e.g. for k-c a/r/W/Ṡ/R/V: OK
note for M/m: if there was a flag that switched everything between 0- and 1-indexed, that would be fine
 
1:26 PM
@AviFS But it's the first to be posted
That bounty is there to help reduce the number of unanswered questions, while also making people look at those that have fewer votes. If someone had posted any of those as answers (CW unless they made them), then the bounty wouldn't be necessary
And the bounty is "only" +300 because I limited those bounties to be +300
 
My take on the flags is the same as it has been for many years; Flags should be considered unique variations of a language. Although they add information, no more so than picking a different language would have.
This is of course not the case where a flag can contain additional information, such as C's -d flag.
 
Maybe I should pose it as a challenge?
 
@Anush The splitting the lines of text?
 
@cairdcoinheringaahing yes
Well splitting the text file
 
@Anush If you do: don't bother with the "file", just allow input as a multiline string.
 
1:34 PM
It's more interesting if you have to shuffle the lines first maybe
@cairdcoinheringaahing I guess standard input?
 
Yeah. Not every lang can read from files
 
Terrible languages :)
 
The best languages are terrible
 
I fear that if I ask for the lines to be shuffled first people will complain there are two things
@ATaco go on... :)
 
1:36 PM
> There are only two types of languages: the bad ones, and the ones no-one uses
 
I support the idea of "keeping your golfing in your code, not your I/O". File input doesn't do the first
 
- Bjarne Stroustrup
 
But I do agree, reading from a file and writing to another file is probably too much overhead for the challenge.
 
What is the most popular language that can't read a file?
 
@Anush When you add something to a challenge, ask yourself "What does this add to the challenge, beyond just making answers do another thing?"
@Anush Define "popular"
 
1:36 PM
@Anush JS (except Node)? Jelly? (except with Python eval which is cheating)
 
If you mean here, probably Jelly
 
@ATaco given you need to make two files it's not clear how to output to stdout
 
Output with a separator
 
Either an Array, or a Deliminator typically.
 
@pxeger JavaScript! I had no idea
 
1:37 PM
BF obviously isn't going to be able to produce an array, thus it's best to allow for a Deliminator.
 
Honestly, I'd discard the whole "lines of text" thing, and just make it into "split a list of positive integers into two parts" where the size of the parts is determined by an input percentage
 
@cairdcoinheringaahing :(
 
Eh, I don't think it being text is too much of an overcomplication
 
The whole point is that it looks like something people really want to do
 
It removes it from your original intent, but it makes the challenge much more accessible (and honestly interesting imo)
@ATaco Depends on the language
 
1:39 PM
You are making a train/test split from some input data
 
@Neil My opinions: definitely cheating: hjLsdSClGgṪJṡ kind-of cheating: MmaRṀ ok: farOKWDoṠ N/A: hcvE
 
i want to say my opinion on flags is they should be reserved for either i/o tweaks or major systematic alterations to language behavior, but honestly as much as i like vyxal r being a major systematic alteration it's hard for that not to feel a bit cheaty
 
Is it more Cheaty than just using a different language version would be?
 
I think that counting flags as a separate system is the worst way of doing it, aside from every other way we've done it
 
i think my real stance on flags is i would never design a language like that on the basis of that by different flag combinations being considered different languages you're not really making any one of those languages golfier but i don't find it problematic that vyxal does do that
 
1:47 PM
I find it more impressive if a Vyxal answer doesn't need to swap flags, much like I find a short solution in Lua more impressive than one in Japt.
 
people just dislike losing to flags when they're posting the exact same answer in a different golflang
 
Also, a lot of people object to Vyxal using flags to outgolf them. But, given that the only competitions between languages are ones that we have in our heads (rather than the actual competition we define here), just do what I do: I don't consider a vyxal answer as "beating" my Jelly answer if it has to use flags to do so
2
 
my plan with perhaps at the moment is flags for i/o only, but build into two different executables for two languages with completely different names where their behavior differs on the principle of failing less/more but not systematically
 
I am still impressed when a Stack Based language beats Jelly.
 
The issue arises for me with things like Perl/Ruby's -n and -p: they weren't designed for shaving bytes of the code (although they're for quicker command-line invocation which is kind of golfing), so they're never really considered cheating, but they're really on par with a lot of Vyxal's cheaty ones to me
 
1:48 PM
true
 
I really don't see how it could be considered "Cheating" for using flags.
If you're using it to store code, sure.
 
I do like the fact that flags give me more motivation to golf my answers :P
If I'm a byte longer than (or tied with) a Vyxal/Japt answer that uses a flag, I'll try even harder to remove that byte :P
 
@ATaco i'm not terribly impressed when it involves strings (especially parsing) but actually in the case of parsing stuff that would naturally be represented on a stack vyxal makes it surprisingly difficult because of how strictly reduce is implemented
 
Vyxal needs a better way to list what flags it has tho
 
My favourite, in terms of interesting, language flag is Zsh's -e (although I know no-one else here really knows what I'm on about): it's designed for a really useful real-world purpose (errors are no longer implicitly ignored); but because golfing in Zsh involves a lot of error-producing hackery it rarely shortens your code, but on the occasion it does (especially for s) it basically removes 50% boilerplate from your submission
 
1:53 PM
ooh
 
@cairdcoinheringaahing like, apart from the help flag?
 
@cairdcoinheringaahing what's wrong with the current h flag? I do wish that output was somewhere in the actual docs, but it's basically fine IMO
 
@Razetime I mean in the actual docs
 
Knowing how to make the best use of a flag, often by significantly changing my method or the structure of my solution, to me is part of the golfing process. But then, golfing exclusively in Japt, I am of course biased.
 
should probably add flag documentation
 
1:54 PM
The interpreter doesn't actually seem to work for me
 
that is actually like halfway what i'm planning the perhaps variant to be lmao
 
@cairdcoinheringaahing pythonanywhere got DDoS'd today
may be a bit wonky
 
2:46 PM
@LeakyNun oh, right, oops. I should make it clearer that i haven't actually implemented that part yet; that's . i'll make the webpage just so it doesn't 404 and just say it's still in dev
alright, pulled to live version
 
the ARM LotM seems to be going really well
 

« first day (3815 days earlier)      last day (1031 days later) »