« first day (3747 days earlier)      last day (1097 days later) » 

12:19 AM
Imagine speed running skyfactory 4(this comment was made by okie who can’t even complete a vanilla run)
 
12:32 AM
1
Q: Collatz, Sort, Repeat

BubblerBackground The Collatz (or 3x+1) map (A006370) is defined as the following: $$ a(n) = \begin{cases} n/2, & \text{if $n$ is even} \\ 3n+1, & \text{if $n$ is odd} \end{cases} $$ Now, let's define the sequence \$s_0\$ as the infinite sequence of positive integers \$1, 2, 3, 4, \cdots\$, and defin...

 
12:50 AM
 
Sandbox posts last active a week ago: Compute a real root of a cubic polynomial., xkcd 2385
 
1:42 AM
For my Hunter-Gatherer KotH controller: I can't decide if I should try to make a graphical controller.
The problem is, there's a lot of complexity in the terrain that would need to be represented, and the map is fairly large
How important would y'all say a graphical controller is?
 
It would definitely help with understanding and visualizing the 2D space like a game (and also eye-catching)
 
Maybe I can try shudders WebGL
 
That's... pretty low level technology, there must be some graphics framework that you can use at higher level
 
Hmm, good point.
 
Maybe check out some 2D game engines?
 
1:56 AM
use ascii art
 
It'd be 3d if I used WebGL
Otherwise I'd just use a canvas, but I can't think of a way to make it look good
 
(it's totally possible to 2D in WebGL btw)
 
Unless I made the scale really close, and allowed panning, which might actually be a good idea
 
If you do want 3D, then try 3D game engines?
 
I'll look into that
 
2:00 AM
godot ftw
 
Oh nice, after ten minutes of confused clicking, I got the keyboard layout I'm used to to work on my other laptop
 
2:19 AM
is it just me or does the new posts thing keep showing a number but doesn't have any posts when i tab over to it?
 
Open the console. Any errors?
 
> Failed to load resource: the server responded with a status of 404 ()
 
That's normal. Anything else?
 
Odd. What version do you have?
 
2:21 AM
I'm here at last
 
0.8 apparently
 
That's the newest one I think, so it's probably a bug I introduced
 
@rak1507 I am a husk of a person. Were you trying to summon me? :P
 
I'll have to do some bug fixing later, not sure what the problem is.
Have any posts showed up at all, or does it always do that?
 
they seem to show up sometimes
 
2:27 AM
Both Q and A?
 
i think so. i'll do some more testing myself to see if i can replicate this more consistent and see if i can get any console errors
there's a chance sometimes the title gets cached when i restart my computer into the other operating system but i'm pretty sure that doesn't entirely explain it
but it might be why i got no console errors this time
 
I wonder if my throw statements are being caught somewhere
 
anyway, just wanted to quickly see if anyone else had this. if not, might literally be just caching and my confirmation bias, if anyone else was, then there's a fix
@RedwolfPrograms ^
 
3:13 AM
@hyper-neutrino Oh, that's useful.
I know what happened
 
3:34 AM
Fixed (Download)
Sorry it took so long, it was a simple fix but I had to break up a fight between some neighborhood cats then spend a few minutes petting one of them in the middle of the night in the middle of the road then go inside and feed fish to my cats and watch them try to figure out how to eat it
3
Productivity and cats do not (usually) go well together
6
 
ah nice, thanks! (also that's alright; cats do be like that :P)
 
@hyper-neutrino are you okay if I borrow some of the branch interpreter stuff for my own vyxal interpreter
mostly the jquery stuff that allows sessions
 
@Lyxal bad idea, but go ahead :P
 
Eww, jQuery :p
 
well luckily i can always roll back to my single file non-jquery version if I screw it up horribly
 
3:48 AM
copying, being inspired by, or even merely looking at any code i write is unadvised for health and safety reasons
 
What sort of functionality are you trying to copy over? Vanilla JS actually does some stuff a bit easier IMO
 
It's so nice not having to support internet explorer...I still feel a bit guilty when I use ES6 though
 
4:09 AM
@RedwolfPrograms Web sessions
I also need the backend flask stuff
 
Like sessionStorage?
 
Like being able to stop code execution once you start
 
Oh, like tying the requests to a specific instance of the interpreter?
 
Something like that
 
I'd just have a random 32 or 64 bit key generated on the server side and sent with the page (or do it client side), then send that with the requests. Shouldn't really need anything complicated (unless I'm misunderstanding what the purpose is)
 
4:12 AM
@RedwolfPrograms HN's site works in a way I want my site to work, and it has a flask server doing the work (which is what I have). So I'm just copying what works
 
You might be interested in The Key :p
 
Here's a description of what I'm trying to achieve: currently, if you run a program and it runs for a while, there's no way of stopping it until it either finishes or times out. I want a way to stop the execution like how you can on TIO/the branch interpreter
 
so link a button to pkill
 
The obvious solution we're all missing is just to have an API to run arbitrary code on your server, then just have the button send the command to kill the process. That offloads work from the server, so it's definitely a good idea.
 
> API to run arbitrary code
shudders
 
4:23 AM
Just restrict it to SQL, anyone who understands it well enough to do something malicious deserves it
(Something malicious and non-obvious, like mining bitcoin, not dropping tables or something like that)
 
5:05 AM
Did you know that the page for declined flags has a link saying "attempting to rickroll" as a rickroll?
 
5:22 AM
3
Q: Is it a chess cycle?

RazetimeInspired by a discussion from the Esolangs.org unofficial Discord Server. Instructions A chess cycle is an arrangement of a chess board where each piece must attack one other piece, and only one other piece, and each piece is attacked exactly once. The effective idea to to "connect" pieces with e...

 
5:34 AM
0
A: Sandbox for Proposed Challenges

EnderShadow8Assassin KoTH Description Each bot is a point in the 2D plane. Each game, the list off bots is shuffled and each bot is a particular bot as their mark. This is done such that there is a cycle of assassins and targets. Each bot's goal is to kill their mark and avoid their assassin. They can kill t...

 
@LuisMendo Thank you for brinigng my attention to this. I've added both APL and Dyalog.
 
6:38 AM
@hyper-neutrino wow your code actually works for vyxal
 
6:51 AM
@Lyxal lol whats the code
 
7:25 AM
@UnrelatedString Lol
 
0
Q: Do sequence allow outputting more than \$k\$ elements?

l4m2Consider question Output sequence 1,2,2,3,3,3,4,4,4,4,... Usually we allow sequence outputting via outputting the whole sequence in infinite time given \$i\$ output \$a_i\$ given \$i\$ output \$a_{1..i}\$ but for languages that outputting function is long, and such question is somehow easier ...

 
7:44 AM
CMC x and y are positive integers. Round x down to the nearest integer multiple of y. I.e. rounddown(100,7) = 98
 
@Ausername Rob the King: Hexagonal Mazes => Rog: 144 I think the Rog: in the output is unintended (?)
 
@Delfad0r Oops yes, I copied-and-pasted the testcases from the previous and forgot to fix it. Good challenge otherwise though? Any other problems?
 
@Anush Jelly, 3 bytes: :¹× with x on the left and y on the right, ×:@ with y on the left and x on the right
 
@UnrelatedString can you explain the method please?
 
floor divide then multiply
 
7:57 AM
in 3 bytes :)
very nice
 
:¹× is specifically "floor divide x by y, do nothing to that result, then multiply it by y"
 
y*(x//y) in python 3
 
%- in 05AB1E, 2 bytes
 
@Ausername If I point out silly typos it's because I can't find any other issues ;) Looks like a very interesting challenge, maybe one thing I would do is add a few "heavier" testcases, where the answer is in the range 10^7~10^8. Optimized solutions written in C or similar low-level languages are likely to take less than a minute even on these.
 
x-(x%y) in python 3 too
 
7:58 AM
and ×:@ is "multiply y by x floor divided by y"
oh yeah how did i not think of modulo
 
the python/05AB1E ratio is 3.5 :)
 
don't you need lambda x,y: in python?
 
@Anush It seems like you don't need parenthesis here
@Anush Can't you do x//y*y here?
 
@Anush APL, 3 bytes: ⊢-|
(The right argument minus the division remainder when the right argument is divided by the left argument.)
 
8:54 AM
Here's a much more mini CMC based on the CMC I gave last night:
CMC: Create an ITTM that takes 2w or more steps to halt in only 2 states.
12 hours ago, by Wheat Wizard
Here's a decent intro to ITTMs: https://arxiv.org/pdf/math/0212047.pdf
 
@Delfad0r I know, but I can't handle these test cases with my naive node.js implementation. You're right though, I'll try to create some longer-period testcases and remove some of the shorter ones.
 
9:13 AM
Nvm it got deleted.
 
9:32 AM
@WheatWizard the pdf is refusing to load for me :/
 
thanks
the pdf loaded when i clicked the link to the pdf from there lmao
 
10:16 AM
@WheatWizard how do you generate these? for some reason i am reminded of metalevel.at/postscript/animations
 
I hand animated them frame by frame in Latex.
 
@WheatWizard so once it hits a limit ordinal it also goes back to the first cell?
 
Yes, it - Goes back to the first cell - Goes to the limit state - And sets the tap to the point wise limit suprememum.
 
@WheatWizard does this work: q0 is start, limit, final. q0: 0→1,q1. q1: 1→0,q0
 
That halts at w I believe.
 
10:20 AM
actually, from wikipedia, it seems 2⍵ = ⍵? en.wikipedia.org/wiki/Ordinal_arithmetic#Multiplication
 
It repeatedly toggles the first cell for the finite time, so then at time w the first cell is on and the machine halts.
Oh yeah, well I intended to write w*2 then.
 
start: q0. q0: 0→0,q0. limit, final: q1. q1: 0→1,R,q1 ?
btw cool stuff, thanks for sharing
 
Yeah I think that works.
I had:
start q0. q0: 0→0,q0;1→0,q1 limit, final: q1. q1: 0→1,q0
 
nice 1-cell solution
 
A modification of your solution can get w*2+1, just add q1: 1→1,q0
 
ngn
10:40 AM
very cool stuff indeed
 
Oh, I just found a 2 state machine that takes w*w time to halt.
New CMC: Find that machine :)
 
ngn
11:01 AM
@WheatWizard did you get rid of q1 from your 3state machine?
 
That machine can't halt, since every node has two transitions.
 
I want to host a dotcomma interpretator
I'll call it dotcomma.com(ma)
3
 
ba dum tss
 
.ma is a real domain ending
so dot.com.ma can actually happen
 
yes
whoever writes an interpreter is legally obligated to call it dot.com.ma
 
11:12 AM
dot.co.ma is available right now at 85USD/yr
 
petition to rename "determine if a dot(comma) program halts" to "does dot(comma) fall into a dot-coma"
 
11:33 AM
@WheatWizard start: q0. q0: 0→0,q0. limit, final: q1. q1: 0→1,R,q0 ?
 
I'm pretty sure that halts at w*2, since the tape head gets put back to the start at every limit ordinal.
 
ah right
 
@Ausername damn it SE
 
ngn
11:56 AM
@WheatWizard i'm watching that video again and i noticed the way the continuum hypothesis is presented is a bit misleading
 
@StackMeter What did you expect? Sportaflop?
 
@Lyxal no I expected a link to a question that rickrolled, not a rickroll
 
ngn
@ngn the guy says "we just don't know", yet CH has been proved to be independent from ZFC, so it's more like "we can choose"
 
@ngn Yeah it is. IIRC there is even a somewhat subtle contradiction in how it talks about the CH. I know they say that 2^aleph_0 could be smaller than aleph_1, and I believe they define aleph_1 as the second smallest infinite cardinal (which is the proper definition).
 
ngn
@WheatWizard here's the relevant part of the video
 
12:02 PM
Ah ok, so it is not as bad as I remembered.
 
@RedwolfPrograms wait, Americans always spell licence with an s and practise with a c? ugh...
 
Wait you spell it prastise?
5
 
isn't there something about like practice = noun, practise = verb
 
yeah, that's how it works here, also for licence / license
it's just weird how the US standardised on -ise for licence and -ice for practise
 
12:08 PM
(or standardized, should I say?)
 
1
Q: Time bomb (KoTH)

leo3065In this challenge, players need to temper with a time bomb as late as possible without getting blasted in the face. Mechanics Let N be the number of bots plus one. Each round, each bot receives the following arguments: numbers played by all bots last round in ascending order. number of points it...

 
can anyone do a Javascript function that returns a random number between 1 and N/2
because I want to enter
but me no understand JS
 
can, yes.
 
thanks]
I'll link you in the answer
 
may, no.
 
12:15 PM
ok
 
ngn
@StackMeter Math.random() returns between 0 and 1. use + and * to offset it and scale it.
@WheatWizard are you taking a set theory course at university or just doing this for fun?
 
I am just doing this for fun. I don't actually go to uni.
 
12:42 PM
XKCD KotH was fun, but it's encouraging a bunch of simpler KotHs with low skill ceilings. I've seen multiple posted to the sandbox already, and that one posted to main is obvioulsy just a copy-paste of the XKCD one with some changes.
I'm fairly sure that the strategy in Time Bomb is going to be almost the same as other ones, like lowest unique number. There's slightly different things you need to take into account, but it's still just "pick the lowest number possible that isn't too low"
 
i would've thought the challenge would be interactive given you have bots tampering with a time bomb - it's just return a number? this is literally just "return a high number without returning a number too much higher than everyone else" in this case lol
predicting at least 5 submissions that return a constant
 
@StackMeter Not sure if you actually do or if that's just the set-up for a joke, but there is a dotcomma interpreter: redwolfprograms.github.io/dotcomma
 
wondering if i should try to figure out a somewhat decent submission for this or just ignore it for being too trivial / similar to the other one
 
Considering VTCing it
It doesn't really add anything new
It's slightly simpler in what you need to optimize, but the strategies will be almost identical to the previous ones
 
Me: *tries to deploy app on heroku**
Heroku: sorry but you have an error
Me: uh no I don't this works on [other host that doesn't allow multithreading] *deploys again*
Heroku: sorry but still no
Me: wdym. It's literally still hosted on the old host and it works offline *deploys again*
Heroku: yeah that's a no from me
Me: why. why must you do this
Man I totally missed out on the xkcd koth
 
12:54 PM
I now have a lot of upvotes from that due to dumb joke answer
 
you know a koth is setting a bad precedent if i can write a submission for it
 
@UnrelatedString you can write a submission for anything though
 
You know a koth is relatively too easy if I write a submission for it
Because I generally dislike koths
 
koth is my favourite mode in tf2
somehow it translates here
 
lol
 
12:56 PM
i like koths but feel like i'm in way over my head when i consider actually participating
 
I should work on the KotH idea I have been workshoping for 3 years now. I kind of want to ask a KotH.
 
@UnrelatedString same but I don't do koths because by the time I see it everyone has already thought of the best strats and there's not much point (imo at least) spending time trying to beat the majority of existing entries
 
I have an idea for a graphical-output animation challenge but I have a feeling nobody will answer :P
 
is more fun than koth change my mind
 
@Neil IMO the worst thing about Americanised spelling is their past participles. "spelled"? "pleaded"? fuck off
@Lyxal it is objectively more fun, but there aren't as many opportunities for new challenges
 
1:01 PM
I don't think "Pleaded" is generally accepted American Englsh.
 
Google says both are accepted in both varieties
 
I actually prefer spelled, spelt sounds like it should be a type of flour or something
 
Wiktionary says in fact that "pleaded" is exclusive to parts of the UK, and some legal jargon in the states.
 
@Neil Same for learned and learnt here. Learnt kinda sounds like you haven't learned how to spell learned.
 
@Lyxal learned is an adjective, pronounced "learn-id". The past participle is learnt
 
1:05 PM
Says you.
 
(if you can't tell, that was my opinion not a fact)
 
they're all incorrect
 
@Wezl no they aren't
 
@Wezl indeed; we should all revert to olde aenglish
 
and this ^ is why i hate verb conjugations lol. particles > conjugations
 
1:06 PM
Gee learn how to spell
@Wezl learned is not the same as incorrect. Neither is learnt
You clearly can't read.
:p
 
@Lyxal neither isn't learnt! You fool!
 
Dang
I got played at my own game
gg
 
@Lyxal playt*
 
Well I'm going to bed
Y'all have fun having 9283 trillion conversations now
 
@pxeger something purely functional is the only correct solution
 
1:08 PM
o/
 
o/
 
I'd be in favour of bringing eþ and þorn back into English. Ðeir such pretty letters. Ðey have a nice aesþetic.
Maybe I should just learn Icelandic instead.
 
@WheatWizard þe capital eþ makes it look like it's written in a Jamaican accent
 
ƿait ƿhat ƿas ðe misspelling i actually looked at it before ðe edit and it did not look ƿrong for some reason
 
I was missing an e in aesþetic.
 
1:12 PM
ah ofc
 
I like how you two to are writing "the" differently reflecting your personal pronunciations. :)
 
@WheatWizard I don't think there are multiple pronunciations of "the". No-one says it with an unvoiced "th", right? It's just about which symbol refers to which sound
 
i feel like nobody pronounces it "þe" but it's ðe more historically appropriate spelling
i forget if english ever even had eð
 
I pronounce unvoiced sometimes.
 
considering it's not like ðere's a phonemic distinction
 
1:15 PM
English did have eth.
The uses of eth and thorn in English was somewhat arbitrary and was not as simple as voiced vs. unvoiced.
 
figures
 
Eth died out first too, leaving a period in which thorn was used but not eth.
 
@WheatWizard Really? What's your accent/dialect? I don't think I've ever heard anyone pronounce it like that except maybe at the start of a sentence when starting to speak quietly
 
North east United States.
 
Ugh, why on earth did people upvote this sandbox proposal?!?
If we get another one of these trivial KotHs I'll make a meta post about it, this one's just a downvote...for now
 
1:25 PM
what's bad about it? (as a KotH amateur)
 
It promotes trivial, almost identical KotHs
It promotes trivial, almost identical answers
It tells people "a KotH is just a trivial number optimizing game where joke answers get +20 and serious effort isn't rewarded"
I don't want KotH to end up like underhanded
 
@RedwolfPrograms hmm yeah, but I don't think that's exclusive to KotH at all
 
I think there are overall some problems with the way we currently conceive of duplicates.
 
2
Q: How many Faro shuffles for a cycle?

pxegerBackground Inspired by this video by Matt Parker A Faro shuffle is a perfect riffle shuffle where the deck is exactly interleaved with its other half. This is how to perform one: the deck of 52 cards is split into two piles of 26 cards each the two piles are interleaved. In the case of an "in" s...

 
@pxeger No, but if that's what people outside CGCC start associating (KotH) with, actually good KotHs that require effort will be ignored after a while
That's why I'm so interested in the idea of Hunter-Gatherer. It's very complicated, and trivial bots have no chance of winning simply because of the massive skill ceiling that should be possible with specialization and teamwork.
There are two factors that make a KotH good (and much more, but bear with me): entry barrier, and skill ceiling. These KotHs have a low entry barrier, but then don't follow up with the high skill ceiling needed to keep it interesting after the first few bots. It's why 69 can even dare to compete with the more complicated mathematical solutions.
 
1:39 PM
@RedwolfPrograms i think we have a few new users, and many people except you know how to gauge the quality of a koth
 
@RedwolfPrograms I've VTCed it. Aside from the range of numbers, I think they're basically the same challenge, and we don't need a whole bunch of "different but the same" KotHs. FYI, this is similar to what happened after EmoWolf. Pretty much every koth after that for a year had an emowolf style answer and it just got annoying
 
oh right, it's a koth so y'all can't hammer it lol
 
Review stalker hasn't been working lately for me
Can anyone repro?
 
Works for me
 
1:43 PM
nope, works for me. here is what i have
 
That's a newer version, I'll update
But then I'd lose my custom features :/
 
out of the 6 subs, there's:
- return 1
- average but change it by +/-1
- random in some range
- random in some other range
- random in a third range
- average
 
I'm going to restart my laptop, that works sometimes. brb.
 
so yeah, not producing very interesting ideas and i doubt it would. maybe a couple of people would think of a longer and smarter solution but it'd probably not beat "return 1" by that much anyway
 
It appears to be a bug with either Chrome or Tampermonkey :/
I'm just going to write my own review stalker script. Can't be that hard.
 
1:48 PM
I should use that script. I haven't been able to do many reviews lately.
 
caird is probably regretting giving me the link to stalker :p especially since my reviews on some things like close votes prevent other people from getting review progress
fortunately (not for me) i sometimes forget to check the tab or just don't notice it found a review for like 10 minutes and then someone else gets to it lol
 
My program is going to be like five lines long and it has all the functionality I need lol
 

« first day (3747 days earlier)      last day (1097 days later) »