The Nineteenth Byte

lyxal

14:02

@allxy you good gamer?

is 8 days

is week + 1

Radvylf Programs

@lyxal During the day, are you primarily in TNB on mobile or on a desktop/laptop?

lyxal

depends on the day

Sandbox Posts

0

A: Sandbox for Proposed Challenges

Collatz Encoding The Collatz Conjecture The famous Collatz Conjecture (which we will assume to be true for the challenge) defines a sequence for each natural number, and hypothesizes that every such sequence will ultimately reach 1. For a given starting number N, the following rules are repeatedl...

Radvylf Programs

I need someone in Australia or some timezone near there to run a userscript for me for a few weeks

lyxal

when

Radvylf Programs

14:04

As soon as possible

lyxal

and how often

Radvylf Programs

It just runs in the background when you're in TNB

lyxal

constantly?

Radvylf Programs

It's gathering information about who's in the room when single-star messages are starred

pxeger

Is there a reason this has to be a userscript and can't be run server-side?

Radvylf Programs

14:05

@pxeger Oh true, I'll just do that

lyxal

excellent idea

back to updating my bios for the week

Radvylf Programs

Installs Tampermonkey Server™

Ayy I think I just fixed a major memory leak in sock_2

I'll have to test it for a little while to see, but it was using like 4 GB of RAM a few days ago ._.

EphraimRuttenberg

@SandboxPosts if people could check out my sandbox proposal and give feedback, it would be greatly appreciated!

im pretty sure it hasnt been done before, but there are quite a few "collatz" themed code golf challenges

Radvylf Programs

Looks like a cool challenge

Bgil Midol

^

Radvylf Programs

14:09

You could possibly make it a standard sequence challenge, although I think it's fine as not being one of those

EphraimRuttenberg

thanks!

i thought about making it a sequence challenge, but i dont think it would function any differently

it doesnt seem as though generating the first N collatz encodings, or all of them would get any shorter answers than the current specification

Radvylf Programs

Yeah, I think in this case it's just "a function that turns a number into a number", which could be a sequence but doesn't really have any need to be

EphraimRuttenberg

i even considered submitting the sequence to OEIS, but i dont have a preexisting account

Radvylf Programs

There's probably not really going to be a cool recursive solution that generates the first N terms either

Bgil Midol

uh, i submitted a comment that was >15 chars but it said it was <15 chars

ok, the message length now excludes the ping

EphraimRuttenberg

14:16

should i bold the condition that n>0?

if you missed it im sure others will as well

Bgil Midol

maybe

EphraimRuttenberg

no good reason not to as far as i can tell

New Posts

14:38

0

Q: Rotate a string half a character

To rotate a string by one character we take the first character off the front and append it to the end. For example: \$ \DeclareMathOperator{\rot}{rot} \rot_1([1,2,3,4]) = [2,3,4,1] \$ To rotate a string by two characters you simply apply this function twice. For example \$ \DeclareMathOperator{...

graffe

14:49

it would be fun to have a challenge to rotate a character by a given angle

lyxal

@RadvylfPrograms I've just realised that your plan is useless

Transcript starting exists

graffe

is that trivial in any language?

lyxal

Meaning you don't have to be in the room to star a message

graffe

CMC Given a letter from a-z, output the character rotated by 45 degrees.

EphraimRuttenberg

is this problem possible for all string lengths? if the function represents a permutation of N characters, then the problem (i think) is equivalent to finding an element of the symmetric group S_N whos order is 2N. however, not every symmetric group has such an element

i dont have the deepest knowledge of group theory so please correct me if im wrong

Neil

14:53

bah, I wanted to include some @s in a comment but SO thought I was trying multiple pings

pxeger

insert a zwsp? here's one (between the quotes) ""

Neil

too late, comment is over 2 min old now, but thanks for the idea though

user

15:09

@RadvylfPrograms does the chat api also tell you when exactly someone entered a room?

Radvylf Programs

@lyxal Sure, but whoever's doing this 1-star-spam presumably isn't doing that

I mean, in theory you could even use a POST request directly

@user You get a message whenever someone enters or leaves a room. I'm just being lazy and using the data-users thingy in the HTML.

pxeger

You can also check by scraping the HTML of chat.stackexchange.com/rooms/info/240/…

Adám

15:34

@graffe HTML, 37 bytes: <input style=transform:rotate(90deg)>

Radvylf Programs

@DLosc I stumbled across this challenge with my random question script:

50

Q: Language Design: 2-D Pattern Matching

This is Fortnightly Challenge #6. Theme: Language Design There's a chatroom for this challenge. Come and join us if you want to discuss ideas! And now for something completely different... This fortnight, we want to experiment with a new type of challenge. In this challenge, you will be de...

pxeger

> 45 degrees

Adám

90 45 then

Bgil Midol

I solved Semantle #90 in 22 guesses. My first guess had a similarity of -1.89. My first word in the top 1000 was at guess #8. My penultimate guess had a similarity of 26.50. semantle.novalis.org

hyper-neutrino

oh since when did SO have a "trending" sort option (recent votes count more so it's like sort-by-votes but new answers have a chance to be visible)

this looks pretty nice

EphraimRuttenberg

15:45

@SandboxPosts should i go ahead and post this challenge? i know its early, but 4 votes in such a timeframe seems quite good

Radvylf Programs

@hyper-neutrino Recently

Bgil Midol

@EphraimRuttenberg the recommended minimum is a week

EphraimRuttenberg

im fine with waiting. however, i see a lot of challenges that are posted very soon after their sandbox, so maybe this should be standardized a little more rigidly?

there are various opinions on the topic it seems. in the sandbox faq the top answer recommends 72 hours

Radvylf Programs

A week without feedback is the normal recommendation

It's changed over time, and IIRC the sandbox FAQ is pretty old

EphraimRuttenberg

yeah lol 7 years

Codidact Posts

15:50

New Posts - Challenges - Code Golf: Create a range grid

posted on April 29, 2022 by Razetime‭

This challenge is based on the i. verb from J. A range grid is basically a range of m × n numbers fit into a rectangular grid with m rows and

EphraimRuttenberg

@RadvylfPrograms so what about with feedback? and do up votes combined with a lack of criticism constitute positive feedback?

im still not going to post today. just wondering

Radvylf Programs

There's not really a hard requirement, since the sandbox mostly for your benefit, but a week without any comments asking for clarification or things like that is the typical recommendation I think.

EphraimRuttenberg

i have posted several challenges before, and it seems pretty highly variable how much feedback you can get in X days. should i be linking it in here every day or so to request feedback until its ready to post?

alright. good to know

caird coinheringaahing

8

A: Is OK to post a challenge for a sandbox post with 0 votes but is 1 week old

Post Sandbox challenges when they're ready to be posted, not based on score or time The reason score/time are recommended (e.g. wait a week, have a positive score etc.) is because these are usually good ways of measuring if your challenge idea will do well. For example, a negatively scored Sandbo...

But yeah, generally, upvotes without comments means that people have read it, think its a good challenge and don't have anything to add to improve it

Radvylf Programs

@EphraimRuttenberg Posting it here for extra feedback is always a good idea, yeah, if you don't think you have enough feedback

EphraimRuttenberg

15:56

still im not quite clear on this. how can i determine if i have enough feedback? and should i post multiple times asking for feedback?

maybe my question would be better as a more coherent meta post. let me know

caird coinheringaahing

@EphraimRuttenberg Unfortunately, there isn't a hard and fast way. However, as a general rule of thumb, I like to check off 3 things: is it positively scored? have I addressed all received feedback? have I asked at least twice in TNB?

If all 3 of those are checked off, and you're happy with the challenge, then it's usually ready to be posted to main

EphraimRuttenberg

alright. those criteria are concrete enough for me. ill check back in a few days for suggestions

i suppose the question of "is this ready to post to main?" cannot definitively be answered without posting it, unfortunately

caird coinheringaahing

@EphraimRuttenberg I've left posts in the Sandbox for weeks, gotten a bunch of feedback, plenty of upvotes, then posted, and someone found a duplicate, or a glaring hole etc. Or, I've completely skipped the Sandbox, and it went on to score 25+. The Sandbox is a useful litmus test, but isn't a guarantee

Radvylf Programs

16:16

Y'know what word I hate? "Individual". it's so hard to type those is in the right spots.

"Indivudual" is always what I end up with

That or "Individial"

DLosc

@RadvylfPrograms Yep, that's one of my favorite questions here, and I occasionally answer in one of the languages that resulted from it.

Radvylf Programs

Oh cool. Also I'd meant to describe why I was bringing up that I'd stumbled across that question, but I got distracted doing something else lol

DLosc

I thought there was something missing. ^_^ My guess, after some reflection, is that you saw a connection to the Language Golf idea?

Radvylf Programs

Yeah

It brings up the overoptimizing-for-test-cases problem, which IIRC we hadn't come up with a solution to

DLosc

Right, which is why that one's a popularity-contest :/

How about this: "Here's a set of sample problems. The actual problems on which your submission will be scored are similar but not the same. Here is their MD5 hash. You have [N] weeks to design and implement your language, at which time the hidden problems will be revealed and you'll write your solutions to them. No new submissions or changes to existing submissions will be accepted after the problem set is revealed."

Radvylf Programs

16:25

I was thinking about posting the general concept the languages were meant to optimize for, then posting the challenge list after some due date, but that has two problems. It means you can't answer after the due date, and that everyone has to come back after the due date and golf with the language.

DLosc

Yeah. The hard deadline is a bummer, although the same thing applies (either de jure or de facto) to a lot of KotH challenges.

Radvylf Programs

Hmm, what if it was like a CnR? Not sure if this would help, but you'd have new languages posted in one challenge, and then a few weeks later, the second challenge would be posted with a list of challenges.

DLosc

@DLosc If the person who implemented the language doesn't come back and do the golfing, I suppose we could allow someone else to edit in solutions to the challenges. The OP would still get the rep from the upvotes.

Radvylf Programs

@RadvylfPrograms That means that if you only want to participate in the making a language part, you don't have to come back a few weeks later and do a ton of work to find your score, other people have an incentive to try to learn your language themselves and get upvotes for golfing in it

DLosc

@RadvylfPrograms Interesting thought! Main objection I can think of is IDK what objective winning criterion the first challenge could have. Would it have to be popularity-contest again?

Radvylf Programs

16:29

So that way, if you go through a ton of effort to learn someone's language and outgolf their solution to some of the mini-challenges (holes?), you get rep as a reward, and they get rep too since it increases their score on the first challenge.

@DLosc It could be based on the sum of the byte counts in the second challenge, and in the second challenge your goal is just plain code golf.

DLosc

@RadvylfPrograms Ah, interesting. So your language's score in the first challenge is determined by the best code-golf score in that language on the second challenge.

Radvylf Programs

So that way both people's scores/rep benefits from a language doing better. The person who made the language gets a better score for their language, and the person golfing with the language (who may or may not be the same person), gets more rep

That doesn't solve the problem of languages made after the second challenge is posted, but then again we sort of have that problem site-wide and with the removal of non-competing we no longer really seem to care.

DLosc

@RadvylfPrograms I would like to care. :P For this challenge type, we could decide to disallow languages newer than the challenge. It wouldn't be as much of a problem (preventing people from participating in something fun) if we posted more such challenges regularly and/or reposted old ones sometimes (with slight variations in the tasks to prevent over-optimization).

Radvylf Programs

That's true, I was mostly just thinking about languages designed specifically and intentonally to do extremely well on the chosen mini-challenges, but I suppose it's still a big advantage to know what they are even if you don't consciously design the language with specific mini-challenges in mind.

How many mini-challenges would these challenges have? I was originally thinking like, 40-80 to make overoptimization harder, but even 10-20 would probably work if the mini-challenges aren't posted until after the due date.

DLosc

16:56

Yeah, my gut feeling is somewhere in the 15-25 range would be good. The 2D pattern-matching challenge has 16, this one has 20, this one has 12, this one 10, and these two 9 (tho the 10 and 9 numbers were chosen partly for symbolic reasons).

I guess the next question is, would the second challenge be closed as a multi-part golf challenge?

Radvylf Programs

@DLosc I'd lean toward the higher end of that range, just so that the scores are more accurate, but yeah. Should there be any sort of weighting to ensure that more difficult challenges don't contribute disproportionately to the final score, or is the sum of the byte counts fine?

@DLosc I'd say no, as most of the issues with those don't apply here.

DLosc

@RadvylfPrograms Ideally, the challenges would be similar in difficulty, tho of course how difficult something is depends on what tools your language has. Then again, that observation is also a counterargument to weighting the challenges for difficulty, since it's impossible to objectively determine a language-agnostic difficulty level for a challenge.

Radvylf Programs

The duplicate issue doesn't really apply I don't think, since while many of the mini-challenges might already exist on the site, it's not harmful in the same way a duplicate challenge is

> The main question is, why didn't you ask these as N separate challenges in the first place?

And in this case, there's a very good reason: We need lots of likely simple challenges in an organized place, and 12-25 individual challenges all at once would be cluttered and lead to all sorts of issues

I mean, the second challenge could even just be a list of links to existing challenges on the site, and I don't think there would be any reason to close it.

m90

@RadvylfPrograms An alternative way (to try to equalise the impacts of different challenges) could be to take the geometric mean.

DLosc

@RadvylfPrograms Yeah. The problem there is that it leaves the door slightly open for over-optimization, since the exact challenges are available from the start (although you have to guess which ones). Also, particularly for older challenges, you start running into inconsistencies with I/O formatting and other requirements (though maybe it would be possible to say "These challenges, but with updated I/O rules as follows").

Radvylf Programs

17:08

@m90 Only issue there is 0-length answers

DLosc

@RadvylfPrograms I also still like the idea of "Here are (N) challenges. Complete at least (N-2) of them. Your overall score is based on your (N-2) best scores on individual challenges."

Radvylf Programs

@DLosc Yeah. I'm definitely not suggesting that's how language golf should work, but taken to the extreme in terms of duplicate-ness I still don't think it'd cause problems

m90

@RadvylfPrograms Right. Perhaps add 1 first.

DLosc

@m90 Random idea, haven't thought through the implications: Maybe use the median score?

Radvylf Programs

I'd be worried a language that adds 255 1-byte built-ins for things likely to be in the challenge list, then uses the last byte for Unary, could win

Or any other high-risk-high-reward strategy that has a few very long answers but is really short at least half of the time

DLosc

17:16

I don't think that'll be a problem if the challenges have enough personality. --Which we want them to anyway. Ideally, if we design the challenges well, no language would include a builtin that does exactly what any of the challenges asks. Who's going to make a 1-byte builtin for translating integers to [insert random conlang here]?

Radvylf Programs

That's true, but I think the median has other flaws. It hides a lot of information, and even if the extreme 1-byte thing wouldn't work in practice, the same general issue of up to half of the answers being ridiculously long not being reflected is something I don't really like.

Sum and/or geometric mean, possibly excluding the two best and worst answers, would work better IMO

DLosc

@RadvylfPrograms Fair enough

What do we think about somebody answering in an existing language? There are two and a half cases: 1) a language they made (e.g. me answering in Pip); 1.5) a language they contributed to (e.g. user answering in Vyxal or Adám in Dyalog APL); 2) a language they had nothing to do with (e.g. any of us answering in Python).

Radvylf Programs

17:35

Not sure. On the one hand, they don't get the rep from the first challenge, which would ideally be the main source, and it provides a nice baseline for the languages to compete with, but on the other hand we don't want the part twos to turn into plain old multi-hole-code-golf if there aren't many languages made in the first part.

I think making a distinction between 1.5 and 2 in practice would be a problem, since I could see that turning into some rather unpleasant arguments about whether someone's ideas or bug fixes were enough to give them the right to answer with it.

DLosc

@CodidactPosts BQN, 5 bytes: ⋈⥊↕∘×

@RadvylfPrograms Yeah. But then does lyxal answering in Vyxal fall under 1 or 1.5?

Radvylf Programs

E.g., if I PR a single-line fix to Vyxal that doesn't add any new features and just fixes a small bug with some operator, I don't think that'd count. If I suggested a new operator to lyxal, which later was added, I'm not as sure, but it probably still doesn't count. What if I suggest 8 new operators? And what if I make a language collaboratively, who owns it vs. contributed to it? I think if we allow pre-existing languages, we'd have to allow all of them, regardless of who made them.

So perhaps restricting it to languages submitted in part 1 is a good idea.

And it also prevents someone from making a language after part 2 is posted, which is super overoptimized, not posting it to part 1, and easily winning part 2

Since if, e.g., Python was allowed, I'd think languages made after the challenge was posted would be too

How would we handle people submitting pre-existing languages to part 1?

Could I post Risky? What about a trivial modification of Risky that focuses more on math operations, but is forked from the same repo and all but a few lines are the same?

DLosc

Yeah, that. Or a language that basically works like Jelly and has a lot of the same builtins but has a somewhat different type system and a different codepage? Just hypothetically. ;)

@RadvylfPrograms I don't know that it would solve all the objections, but maybe languages that aren't eligible for part 1 could post community wiki solutions to part 2.

DLosc

17:58

@RadvylfPrograms One thought: If we're planning to post multiple challenges like this, we should do one of two things: either 1) allow pre-existing languages / slight modifications thereof, or 2) place limits on the languages submitted to require a certain level of minimalism. The reason is because I don't like the idea of having to come up with a new language with 200-odd builtins for every single Language Golf challenge.

Either make it so everybody only gets 16 builtins (or some similarly small number), or let me reuse most of my existing language's builtins.

Radvylf Programs

18:13

TIL about node --inspect

It lets you open a dev tools window for it...in Google Chrome

You just go to chrome://inspect, and it's there

Since they both use V8 and stuff

This should help me fix that memory leak in sock_2

DLosc

18:28

@CodidactPosts tinylisp, 95 bytes

tinylisp 2, 40 bytes: (\(M N)(m(\(Y)(r(* Y N)(*(+ Y 1)N)))(r M

Lexical scope is so very nice ^_^

Ginger

onioni? (I'd put this in the Tarpit but it's frozen lol)

Radvylf Programs

Given how well Ginger Bot went, I'd hope whoever makes this bot has more reliable testing set up

And maybe some kevlar armor :p

Ginger

*checks closet*

*checks server*

yup

DLosc

@Ginger Keeping lists up to date is a good thing, but I have no particular onions about a bot because I don't know what Esolangs' policy/culture on bots is.

Ginger

@DLosc they don't seem to have one

I checked

Actually, they don't seem to have any bots

so yall think this isn't a garbage idea that will blow up in my face ~~like my last bot did~~?

Radvylf Programs

18:43

Who said that

Ginger

nobody has yet said it's a garbage idea

perhaps I am being too hasty

Radvylf Programs

Because I think it's a good idea, that would blow up in your face

Ginger

why, because I have a bad track record with bots?

Radvylf Programs

...maybe :p

Ginger

this is simple. it's boring. it could not possibly go wrong.

Radvylf Programs

18:45

I'd challenge your assumption it'll be simple

DLosc

@Ginger I wouldn't say it's a garbage idea, just that it probably wouldn't get traction (because they haven't done anything like it before) or if it gets traction, it probably won't last long (because of unforeseen problems, lack of maintenance, etc.)

Ginger

the task is simple, the bot isn't

Radvylf Programs

Making a reliable bot is hard, especially when it involves network stuff

Ginger

assuming I can overcome that issue, is the idea itself good?

Radvylf Programs

And as I've learned from NPSP, maintaining the bot takes up a lot of time, and you never know when you'll wake up and realize you've failed the community for the last six hours because your server got hit by lightning and the bot went down

Ginger

18:47

clearly I need to make a bot to maintain the bot /s

Radvylf Programs

@Ginger I think so, although people remembering to do one of the things and forgetting the other is probably rare

Ginger

I can check that

one sec

DLosc

@Ginger Heh, taunting Murphy. Bold move. :P

Radvylf Programs

Okay, well frick. The memory leak appears to be in something called knownTimersById which has three results when you google it

DLosc

@DLosc 87 bytes

Or 104 bytes without the library

Ginger

19:01

well radvylf, to answer your question,

according to my janky-ass program there are >9000 languages that are not in the language list

which seems... a bit sus

but alright

Radvylf Programs

Please tell me you didn't make over 10k requests to the esolangs server in the last 15m lol

Ginger

wait

there are only 3000 pages in the category

is it counting subcategories???

dafuq

:|

is it counting ghosts or smth?

uhh

oh

I am dumbass

Unrelated String

talk pages?

Ginger

ok, figured it out

there are actually 141

the reverse is... harder to estimate

but it's probably lower

so there you go

Romanp

why such a big difference?

Ginger

19:09

I wasn't using the continue property correctly

so it was just counting the same 500 over and over

Romanp

ah

Ginger

and there you have it, question answered

issue: it doesn't look like any of the admins are very active

the bot could also potentially clear this page every so often

@ais523 not sure if you'll see this, but what are your thoughts?

Radvylf Programs

They're not pingable

Ginger

dorn

I'd need to contact an admin for the bot to be allowed but they all seem to be inactive

ok, I found their chat

Newly Featured Posts

19:30

19

Q: King of The Holster: 1 WEEK LEFT

Closing on May 6! You stand on a dusty stretch of code, watching the other programs stare at you. Will they shoot? How much ammunition do they have left? All you know is that you want to be the last one standing. Before the Game You will be given 10 points to distribute between HP, Dexterity, Arm...

king-of-the-hill java

graffe

19:55

@Adám picture please!

@Adám I mean for this

Radvylf Programs

DLosc

Retro CMC:

Sep 6, 2017 at 20:32, by DJMcMayhem

@MagicOctopusUrn CMC: consecutive differences between inputs. My favorite answer gets a cookie: (🍪)

Radvylf Programs

x=>x.slice(1).map((n,i)=>n-x[i])

I'd imagine most golfing languages have this as a built-in

DLosc

@RadvylfPrograms Pip, 6 bytes: B-_MPg :P

Alternately: Sg-:Hg

After a lot of trying different things, I found 5 bytes in BQN: 1↓⊢-»

Radvylf Programs

@DLosc Ooh that contains an emoticon

g-: is the kind of emoticon you use when you're trying to get people to realize there's something wrong but if you say anything specific they'll notice and bad things will happen

DLosc

20:11

@DLosc Or, rather amusingly: -Hg-Sg

which parses as -(H(g-(Sg)))

All but the first element of g, subtracted itemwise from g; take all but the last element of that, and negate.

DLosc

20:50

This challenge requires STDIN/STDOUT, but if function solutions were allowed, then--BQN, 7 bytes: ≠∾○(⥊˜)

Oh, whoops, that has 2 tiebreaker characters. Let me see...

Unrelated String

@RadvylfPrograms Brachylog (generator), 3 bytes :P

DLosc

@DLosc All right, 8 bytes with no tiebreakers: ≠∾○{⥊˜𝕩}

Unrelated String

and add two bytes if you actually need a list

actually took me a few minutes to think of how not to have to use s₂

DLosc

@UnrelatedString Ooh, that's fancy

Unrelated String

s₂ does tie with reversed signs ofc

it's basically "find any contiguous substring of the input, it's a pair of two elements and subtract the first from the second"
and then the header gets every result it can backtrack to

DLosc

20:57

Yep

Unrelated String

oh wait i forgot you're a brachylog user

but yeah it's weird using ₍ on something that can also just take a list :P

DLosc

Using zip (which is probably what I would've thought of first) comes out to 7 bytes, a couple of different ways: b;?z₀-ᵐ or ⟨bzk⟩-ᵐ

Too bad there doesn't seem to be a zip-with metapredicate

Ginger

I'd do this in makina if I had time

DLosc

21:14

tinylisp 2, 43 bytes: (d H(\(L)(?(t L)(c(-(a -(] 2 L)))(H(t L)))( (because zip and zip-with aren't implemented yet)

I believe it's 3.5 bytes in HBL: 2-(2.).

However I can't test that right now because the online version is out of date and doesn't have zip-with

Unrelated String

@DLosc reminds me, i wonder how many times anyone's used ᶻ

DLosc

@DLosc Also b;?zk-ᵐ

@UnrelatedString I don't think I ever have. It seems like it ought to be useful in some context, but...

Hmm: ≡ᶻ⁰ pairs each element in a list with a copy of the full list. Is there any better way to do that?

Unrelated String

21:30

probably not

DLosc

I think I've had to do something like that before

Unrelated String

ᶻ⁰ probably has a lot of weird clever uses that none of us have thought of because we forget it exists lmao

and yeah come to think of it i think i've had to do that too and gone with something like g;?z

which is both in the other order (which can also be achieved by ≡ᶻ²) and relies on it being the input variable

reminds me i feel like a general "pair input with result" metapredicate could go a long way

hell that could even be a worthwhile quick in jelly for when chaining with , won't cut it

i've used Ƭ for that way too much

DLosc

21:48

@UnrelatedString So basically a shortcut for ;?↔ ?

Unrelated String

specifically input to the predicate rather than overall but yeah pretty much

as in the predicate it's applied to

DLosc

Ah, okay

So a -> ⟨≡≡a⟩

Unrelated String

yeah

DLosc

Come to think of it, is there any other 2-byte way to do ≡ᶻ?

Unrelated String

...good question

three bytes is obvious ofc

DLosc

21:53

;?z but that doesn't work unless you're dealing with the predicate input

Unrelated String

gjz

DLosc

Ah, okay. Yeah, I was trying jᵐ earlier but that didn't quite do the same thing ^_^

DLosc

22:22

Well, I haven't found any Brachylog answers I can golf using ᶻ, but I did find one I could golf in a different way ;P

Unrelated String

ncie

lyxal

23:45

@RadvylfPrograms although I would argue that because of the organisational structure of vyxal internally, there's a clear cut line of who owns the language.

Organisation owners would count as class 1 and everyone else as 1.5 or 2

Radvylf Programs

Yeah, it was the 1.5 vs. 2 distinction I was talking about up until the "and what if I make a language collaboratively" part.

Transcript for