I needed this bookmark for discussion purposes.

hyper-neutrino

Jun 18, 2021 19:28

rip, vyxal once again beats jelly by moving the two bytes that would've made it tie into the flags...

rak1507

metagolfscript wins again!

caird coinheringaahing

@SlamJammington Pretty sure its 2.0000000000001 cookies, damn floating points :/

hyper-neutrino

outputting as a matrix of characters for ascii art is acceptable right?

i can take the Y out of my own code in theory

user

Well that was fast — SlamJammington 22 mins ago

Underslash

@hyper-neutrino assuming thats standard output, go ahead

user

Jun 18, 2021 19:30

Well that was fast :p — Redwolf Programs Apr 5 at 15:07

Do you always FGITW challenges, HN?

hyper-neutrino

well Jelly is able to output as a multiline string which is just a list of characters including \n

@user not intentionally it just doesn't take me long to solve easy challenges in jelly

the other reason is vyxal, like 05ab1e, has certain things as one-byte built-ins that Jelly has as two

also, the halve built-in in vyxal works on strings and splits them into two halves. Jelly's just errors. :/

(granted, it also wouldn't save bytes because Jelly has built-in string-list literals which vyxal doesn't)

Redwolf Programs

Jelly errors entirely too often for a golfing language, from what I've seen

hyper-neutrino

it doesn't overload enough IMO

i like that it vectorizes, but sometimes it can't and just errors, and I think overloading makes a whole lot more sense

unfortunately i cannot tie vyxal because of monadic 2,1 chain rule

caird coinheringaahing

I'd like y to not error on odd-length left arguments, and M and Z should cast to iterables. Otherwise, Jelly's erroring is pretty good

hyper-neutrino

i'd like the repeat built-in to accept the list on the right if the left argument is an integer, because you need @ or to mess up chaining a lot to handle that

caird coinheringaahing

Jun 18, 2021 19:36

I'd change X to error on 0/strings/floats and to return a random integer from x to -1 if x < 0

Aside from those 4, the errors mostly arise as parsing errors

hyper-neutrino

the only time i've wanted to be able to like, for example, put " with no chain before it is in polyglots :D

Catija

@cairdcoinheringaahing The NaR badge is for events - like conferences... it's mostly used by the marketing department. So if we don't go to an event (or do something that gives out a NaR badge) we don't get it..

Redwolf Programs

...is what a robot would say :p

Underslash

I wish the vyxal flags counted as bytes tbh

pxeger

booo

hyper-neutrino

Jun 18, 2021 19:41

I would be open to a new meta proposal revisiting byte-counting mechanisms

I think j is a legit flag though

Underslash

which one is j?

hyper-neutrino

here's the thing

cumbersome IO sucks, so i support j

the sum flag is pushing it too far IMO - you can't output [1, 1] and say "here, I printed 2"

@Underslash join on newline

Underslash

i think some of the flags are just extracted code

Redwolf Programs

But for things like FizzBuzz it's 100% cheating IMO

Underslash

@hyper-neutrino right

hyper-neutrino

Jun 18, 2021 19:43

I think wherever outputting a matrix of characters is acceptable, using j is acceptable. for example, I golfed my jelly code by removing the Y from the code itself because 2d list of chars is usually acceptable string output formatting

Underslash

^^^

if its just changing the output from one form to another, that shouldn't take a byte

caird coinheringaahing

@Catija Hmm, and now the Turing Test begins :P

Redwolf Programs

I think anywhere the output would be acceptable without j it's fine

Underslash

but initializing with 100 in the stack is a bit much imo

caird coinheringaahing

@hyper-neutrino I wouldn't

Catija

Jun 18, 2021 19:44

@cairdcoinheringaahing I have voice samples you can listen to. :D

pxeger

that's something no normal human would do lol

caird coinheringaahing

@Catija So does Siri :P

Catija

I'm pretty sure that Siri wouldn't be able to have a 15 minute conversation with someone about complex things related to Stack Exchange.

caird coinheringaahing

Given that none of the staff have the NaR badge, we can only conclude that SE is run by one neural network pretending to be multiple staff members :P

hyper-neutrino

there is another hivemind :o

Catija

Jun 18, 2021 19:45

And... y'all... seriously... if the company had the assets to create an AI, we'd be using it to answer questions, not to CM.

New Posts

0

Q: A multiple of n in every base!

In November 2019, Alon Ran published a particularly lovely sequence in the OEIS, A329126: \$a(n)\$ is the lexicographically earliest string of digits which yields a multiple of \$n\$ when read in any numeric base. 1, 110, 101010, 111100, 100010001000100010, 1111110, 10000010000010000010000010000...

hyper-neutrino

@cairdcoinheringaahing fair enough; it's just vyxal and if lyxal pushes it too far a) metagolfscript rule takes place and b) we can just go yell at them lol

and it doesn't seem to be a general issue

caird coinheringaahing

> Hey Siri, should Code Golf Stack Exchange have a custom new asker dialog?
I'm sorry, I didn't quite catch that. Could you repeat it?

Redwolf Programs

oops

caird coinheringaahing

Taken verbatim from my Siri :P

Redwolf Programs

Jun 18, 2021 19:48

Part of the reason I don't like Vyxal is how focused it is on one challenge (FizzBuzz)

hyper-neutrino

kF?

Redwolf Programs

Most of the features that help with FB were conventiently introduced right after I talked about the length of FizzBuzz in Ash

rak1507

flags absolutely should count towards the byte count and you can't change my mind :P

pxeger

yeah I agree, it's a little childish

Redwolf Programs

@rak1507 I don't like that as a rule either though, because it lets you encode a tiny amount of info for free

pxeger

Jun 18, 2021 19:49

@rak1507 rak1507 absolutely should count towards the list of people I disagree with and you can't change my mind :P

caird coinheringaahing

@rak1507 why?

Underslash

I wouldn't say its bad how its "focused" on one challenge, just has enough room for that

rak1507

@cairdcoinheringaahing because having a flag to sum the stack or whatever is just blatant abuse

Catija

@cairdcoinheringaahing status-inprogress

:P

pxeger

:)

hyper-neutrino

Jun 18, 2021 19:50

@rak1507 counting all flags is punishing for a couple of languages though (i think perl is one of those?)

caird coinheringaahing

What benefit does it offer to charge for command line flags besides making people feel better that their Jelly answer has the same score as the Vyxal answer that uses 3 flags?

Underslash

@RedwolfPrograms how does that let you encode a small amount of info for free?

hyper-neutrino

that too. languages don't compete with each other, anyway :p

rak1507

@cairdcoinheringaahing well why ban metagolfscript in the first place

hyper-neutrino

@rak1507 because it's boring

caird coinheringaahing

Jun 18, 2021 19:51

@Catija That's much better than the Siri response

@rak1507 Metagolfscript isn't banned IMO

rak1507

@hyper-neutrino as is using loads of flags in vyxal

hyper-neutrino

with S you still have to solve the body of the challenge. MGS literally requires 0 brain cells

Underslash

@hyper-neutrino they dont but they do, there is an element to golfing that is trying to beat other languages with optimal answers

@hyper-neutrino what is metagolfscript?

hyper-neutrino

well, inter-language competition is a competition not between the golfers but the language devs

6

Redwolf Programs

@Underslash Imagine a language where every program is two bytes. You can triple the number of variations by encoding it as two flags + empty, one flag + one byte, or two bytes and no flags

caird coinheringaahing

Jun 18, 2021 19:51

I don't see the Standard loopholes as "bans" - instead they are a list of things that if you do this, be warned that you'll get abunch of downvotes and people complaining

hyper-neutrino

@Underslash metagolfscript is a language with infinitely many versions - version X runs the golfscript program obtained from decoding X via some method

caird coinheringaahing

@hyper-neutrino It's both. Jelly may not have gotten off the ground if Dennis couldn't demonstrate its power very well

hyper-neutrino

that is true

Underslash

@RedwolfPrograms you could, but the alternative is a language that has minimal byte answers but a heavy abuse of flags

hyper-neutrino

but the competition to golf between them is still somewhat down to the devs - each language has an optimal solution (even if it isn't found by anyone) and it becomes a matter of language design and features

Underslash

Jun 18, 2021 19:53

is a program really "1 byte" if the majority of the program rests in the flags?

caird coinheringaahing

Additionally, Metagolfscript isn't a language, it's a family of languages. By our rules, Metagolfscript-1 doesn't compete with Metagolfscript-2 or any other language

hyper-neutrino

for example, thus far i've found vyxal better for many string challenges, but really really hard to use in many array challenges where the jelly answer is like 7 bytes

caird coinheringaahing

So all Metagolfscript-X does is be the shortest Metagolfscript-X answer, good for it

Redwolf Programs

Even if we officially say that answers in different languages don't compete, comparing Husk/Jelly/Vyxal still happens and it's still fun to have more competition

hyper-neutrino

@cairdcoinheringaahing fair enough. i apply most of them like bans - for example, "interpreting the challenge too literally" is more of a "you didn't even solve the challenge" not just "you're bending the rules", but i can understand your train of thought for that too

caird coinheringaahing

Jun 18, 2021 19:54

Yes, but you shouldn't break a system that works in order to have "more competition"

hyper-neutrino

and i guess that's what downvotes are for anyway - posting something against a loophole and getting 6 downvotes is literally worse than not posting it, so if they do, ¯\_(ツ)_/¯

Underslash

Because I will say, there isn't much posting of the same language on the same challenge, which is why you have to (at least a bit) have competition between languages

Redwolf Programs

^

Like, as a JS golfer, it's fun to try to beat Python answers.

Partly because there's usually not too many other JS ones, and partly because I don't like Python

hyper-neutrino

right, it's fun, but I don't see the need to restructure our rules for that

caird coinheringaahing

Yeah, and that's fun, but you shouldn't break the system because some guy is doing it a bit unconventionally

Hell, look at Pxem

hyper-neutrino

Jun 18, 2021 19:56

it's not causing a major issue with our site

Redwolf Programs

I think both systems work, I agree changing the current one is more effort than its worth

But I just don't like the "languages aren't competing with each other anyway" argument :p

Underslash

I mean, couldn't you metagolfscript with flags?

hyper-neutrino

and get 7 downvotes? sure ¯\_(ツ)_/¯ if you want

Underslash

I mean sure, but why even allow that stuff to exist in the first place, if normal metagolfscript is banned

caird coinheringaahing

The fact that we spent 6 years and hundreds of meta posts trying to get a conclusive definition of when to score specific flags and when not to demonstrates to me that we can live with a few users wanting to be able to tie/beat lyxal

Underslash

Jun 18, 2021 19:58

thats true, I haven't been around long enough to see any of that stuff

so I might be arguing a point that was made and refuted 5 years ago

hyper-neutrino

the current consensus does technically ban loopholes

because the definitive policy on invalid answers includes "violates a loophole" as an invalidating factor

however, most of the time when people violate loopholes, they get a lot of downvotes and a lot of comments and end up self-deleting, and i don't tend to actually use my mod delete on answers that break loopholes, just answers that didn't solve the task correctly and didn't get fixed

caird coinheringaahing

¯\_(ツ)_/¯ I interpret it differently. I think that, so long as an answer works and is scorable, it shouldn't be banned. It can be downvoted to hell because it abuses various rules, or people don't like it or whatever, but it shouldn't be banned

hyper-neutrino

fair enough

I guess the TL;DR is we don't really need to change anything :P

our system works fine. i agree with that. i don't really care that i got outgolfed by 1 byte because vyxal has flags (and tbh r isn't particularly outrageous, i'd actually consider it fairly okay)

The Nineteenth Byte

Participants