The Nineteenth Byte

Neil

00:53

apparently \sum_i=0^\infinity1=-\frac12 for certain meaning of sum but I can't find an explanation of that one (the \sum_i=1^\infinity i=-\frac1{12} one is everywhere)

DLosc

01:05

@lyxal Same

emanresu A

@Neil there are a lot of ways to reconcile divergent series, not all of them consistent

Unrelated String

@lyxal I have no idea how but I spent way too much time in middle school literally just reading esolangs wiki. Not even doing anything with it, just reading

@att and I know I found IOCCC off Wikipedia (somehow)

those numbers are like. WAY higher than I would have guessed lmao
like if I *think about it* 20% is actually not implausible considering there's no reason most of that 20% can't still be neurodivergent in other ways, granted, but it's a very specific kind of activity that self-selects in some pretty specific ways
Maybe not so much the case for Perl and APL golf since it's more ingrained into the culture of programming in those languages, but otherwise you definitely need something or another in your brain telling you "holy shit this is cool I literally need to do this" and continuing to do …

@user the classicccccccccc 🫂🫂🫂

I actually had a therapist for years whose main job was just giving me those kinds of tips and my dad would just shame me for not implementing them myself 🤔

emanresu A

01:23

🫂

Unrelated String

<3

att

02:14

9

Q: $1 + 1 + 1 +\cdots = -\frac{1}{2}$

The formal series $$ \sum_{n=1}^\infty 1 = 1+1+1+\dots=-\frac{1}{2} $$ comes from the analytical continuation of the Riemann zeta function $\zeta (s)$ at $s=0$ and it is used in String Theory. I am aware of formal proofs by Prof. Terry Tao and Wikipedia, but I did not fully understand them. Cou...

rydwolf

06:29

feel free to add me on linkedin

mousetail 'he-him'

I think they need you in the baltic sea

RubenVerg

08:33

is there any pretty way to make an arbitrary height angle bracket ⟨ with Unicode chars?

m90

08:54

@RubenVerg U+2571 and U+2572 could work (appearance depends on the font, of course).

RubenVerg

yeah, i was hoping to have one char wide versions

for four tall there's 🯖🯗🯔🯕 from legacy computing

RubenVerg

09:13

heights 1-4 look nice but then it gets messy

Neil

09:23

@att ah, so you can't magic it like you can with \sum{i=0}^\infty i= \frac16\sum_{i=0}^\infty(-1)^i

emanresu A

10:34

TIL that this SE feature exists

ahh it's an extremely recent browser feature

Weird Glyphs

13:34

Since I edited my previous Sandbox answer instead of making a new one, I might as well put it here.

0

A: Sandbox for Proposed Challenges

Card-Jitsu Part 1: Find all winning sets of three cards Available here Card-Jitsu Part 2: Make a Card-Jitsu clone code-golfcard-gamegame Part 1 is available here. Card-Jitsu was a mini card-game based on Rock, Paper, Scissors available on the children MMO Club Penguin. A lot of footage of this ga...

I will probably validate the shortest code before publishing Part 2

So probably in a few days

Ginger

@emanresuA don't think it's that recent? chromium-family browsers have supported that highlight syntax for ages, and it just doesn't work on my copy of FF

Weird Glyphs

@Ginger Weird, doesn't work for me and my version of Brave is not that old. But I'll definitively try it as soon as I have access to my computer with auto-updates. This is a cool feature, letting anyone add their own anchors...

Seggan

@rydwolf wat da

Weird Glyphs

13:49

@rydwolf Best use of Linkedin

Themoonisacheese

15:03

@att mine was the ioccc submission that was a small flight sim with the source code in the shape of a plane

@Ginger works on mine, ff 134

@WeirdGlyphs unsure about this, specifically with regards to testing

like the way you interact with a program in some languages means that you must provide the input ahead of time, but since the cards are random it feels kinda weird? i guess it's fine because submissions are supposed to work with every card combination that could be generated but it feels weird to me

Weird Glyphs

@Themoonisacheese I mean, it is weird because you can't really choose your cards, you have to choose numbers between 1 and 5 in advance, and then the game will "play itself", but these languages will have a hard time competing for challenges using the input and randomness at the same time either way. As long as they can compete, I think it's fine enough

Themoonisacheese

15:18

i guess that's fine yeah

the task isn't to perform optimally at the game so it's fine if your input is bad

(though part 3 idea?)

Weird Glyphs

@Themoonisacheese Maybe, not a bad idea :) I already had an idea for Part 3, and I'm not sure if I'm going to publish it right after Part 2, but it can be entertaining, so I'm at least going to try writing it (unless you want to).

Themoonisacheese

no go ahead this is your baby

Weird Glyphs

But wait, would this be a [king-of-the-hill]?

Themoonisacheese

i mean fresh challenges are always welcome it doesn't have to be part 3, though i'll admit you really ramped up the difficulty for this one lol

@WeirdGlyphs i think not, i think card-jutsu has trivial optimal play

i can't be the first to think of this, has anyone ever proposed making a codegolf.se king of the hill server?

the main problem with koth is that once the creator loses interest, scoring submissions is kinda dead. yo ucould fix this by creating essentially code.golf but specifically for hosting and scoring KOTH problems posted on main

Weird Glyphs

@Themoonisacheese Yeah, I was worried about the spike in difficulty, should still be fine but barely. The reason why I was probably not going to make a Part 3 right now was so I can make a fresh challenge in-between, so that's not a problem.

Themoonisacheese

15:27

agree

Weird Glyphs

@Themoonisacheese Probably not a fresh idea, not a bad one either, but someone would have to maintain the servers

Themoonisacheese

@WeirdGlyphs i do that already

i've been too burnt out to creat new things but the servers i have are running

Weird Glyphs

Nice

Unrelated String

🫂

Themoonisacheese

i feel bad about it too i've been hosting stuff for madeline but i can't manage to find time to help her figure out postgres replication and ipv6

eve though i thinkl my ipv6 problems are like 1 server install away from being resolved i'm still not doing it and that sucks, like why am i not able to do things

@Themoonisacheese i guess the main problem with this idea is that most people do not care about KOTH anymore

(but is that because KOTH challenges die in a month? hah! only one way to find out!)

Weird Glyphs

15:39

@Themoonisacheese tbh, I have a hard time thinking of an optimal play for Card-Jitsu

On top of that, would the opponent be able to choose randomly like in Part 2, or would it have its own hand to manage?

Themoonisacheese

@WeirdGlyphs i guess you're right it's not so trivial

actually since you never know what you opponent can play it might as well be random

Weird Glyphs

Well that's not completely right

Themoonisacheese

mmh but then you want to avoid playing cards that could give them the win

Weird Glyphs

If your opponent plays Fire multiple times for example, you can assume that they won't have Fire for a few terms

Themoonisacheese

i guess a koth would make sense then

Weird Glyphs

15:50

It's like Rock, Paper, Scissors if you had limited amount of each

Themoonisacheese

well that and also you win by stacking things you play, which potentially makes you make different choices about what you play

Weird Glyphs

Exactly

Unrelated String

16:12

Themoonisacheese

who knew that hit game from the 2010s Club Penguin distributed by The Walt Disney Corporation™ had such deep strategy

Unrelated String

wait did the reply not go through

@WeirdGlyphs I was specifically replying to this :p

Weird Glyphs

@Themoonisacheese Deep strategy indeed

@UnrelatedString I only saw the picture you posted, nothing else

Unrelated String

sad

Neil

16:49

@att I got to IOCCC from USENET, so there!

Unrelated String

Whoa!

nice

Neil

it used to be that everyone thought that the web is the internet. now they just think chrome is the internet

mousetail 'he-him'

17:32

@Ginger It works on my FF, though it doesn't scroll down to the place where the highlight happened

DLosc

^

@Themoonisacheese It's definitely been proposed before, probably circa 2017. I don't remember the details of what happened.

rydwolf

18:09

dammit, I made enough from my online work-from-home thing this year that the IRS is gonna know about it

and I probably cannot afford the taxes on $10k in income >:|

Seggan

18:34

@rydwolf ?

you have that?

rydwolf

yeah, did some stuff online over the summer to pay rent

AI-related, which is embarassing enough I pretend it was just an onlyfans

But I made well over the reporting threshold, so I can't hide it from the IRS

Seggan

noice

ai smh

rydwolf

Better than me and four friends being homeless, but yeah still not proud of it

Seggan

is it just me or does every decent programmer i have met absolutely hate the current ai hype

rydwolf

It's the latter

rydwolf

19:18

@Themoonisacheese RTO was going to have this

With a cool cross-language framework thing

But then I got busy with the eternal struggle against the forces of evil (homework)

Seggan

19:42

i had a project started that i got permission from pxeger to use ATO for

Seggan

19:53

*similar project

emanresu A

Any feedback?

Weird Glyphs

@emanresuA Looks good to me, seems like a simple fun challenge, I don't have anything else to say about it

hyperneutrino

20:42

@UnrelatedString :3

@Adám without reading any of the subsequent messages to avoid spoilers: from what i remember that should be dyadic uptack and monadic rho and i did have to go on tryapl to remember what either of those do but that looks like it takes the shape and converts it to base 0? that seems wrong but i messed around with uptack a bit and it seems like it's basically just "evaluate the polynomial defined by RHS at x=LHS" since 2 ⊥ 2 4 3 gives 19 even though that's not valid binary

so i'd guess 0⊥ is basically tail for a vector of numbers but i am now moderately confused lol

Adám

22:47

@hyperneutrino You nailed it. So the tail of the shape.

DLosc

23:46

CMQ: Create an arithmetic sequence of ten prime numbers (i.e. the difference between each consecutive pair of numbers is the same). The sequence with the smallest maximum number wins.

lyxal

curse you 63 for not being prime

Adám

@DLosc 199 + 210n; score 2089

DLosc

Nice! Do you know if that's optimal?

I think it is: it's the smallest one with a delta of 210, as checked by this Brachylog program, and since the next lowest delta that might work is 420, the end number would definitely be higher.

Transcript for