« first day (3785 days earlier)      last day (1045 days later) » 

9:02 PM
@SandboxPosts OP seems to have deleted this, so I doubt it;ll be main'ed. In which case: CMC: Output the following matrix, in any reasonable 2D format, without using any digits in your code
 1  8  9 16
 2  7 10 15
 3  6 11 14
 4  5 12 13
 
8 bytes in Jelly
 
What's your MD5?
 
e50e2e394a57130e881c86c6004f5178
 
Ooh, mine is fe0bfb470d3b55c37cd6327975a5cdd9 (also 8 bytes)
Link for mine
 
hold up mine is invalid lemme try again
(i had s4 lol)
is there a way to unflatten a matrix into a square?
oh found one. note to self: make (or equivalent) work in my language
 
9:08 PM
@cairdcoinheringaahing Sorry, after receiving notification of your comment I deleted mine to avoid clutter, and only then did I notice it was linked in the text. So, broken link now :-|
 
then you must be saving a byte over me elsewhere.
@LuisMendo if you want, i can try to undelete your comment
 
Hm let's see what @caird says. Maybe it's better for him to just delete the link?
 
@LuisMendo I can remove the link if you don't want HN to undelete it, no problem. Just thought I'd give you some credit :)
 
Anyway, I have ⁴sƽ¹Uƭ€Z... not sure where the extra byte is. I'm missing a quick that does this aren't I :P
 
9:09 PM
Ok, let's try undeleting then. It's a new experience to me :-) @hyper-neutrino
 
@hyper-neutrino Yes :P
 
alright, undeleted and fixed a typo :D
(had to use a userscript; the base interface actually doesn't allow me to delete comments that were deleted by anyone other than a current mod)
(including the user themselves, ex-mods, staff, etc)
@cairdcoinheringaahing wait since when did Ðe and Ðo exist
i have scrolled down to the bottom of the page many times for the maximal-by-link-value built-ins. am i blind?
 
@hyper-neutrino Like everything in Jelly, for at least 3 years :P
 
:'( yeah im just blind :D
 
@hyper-neutrino Haha, thanks!
 
9:13 PM
Hmm, apparently œ! is the newest Jelly command, and is 2 and a half years old :(
@hyper-neutrino 29th of May, 2018
 
writings.stephenwolfram.com/2021/06/… is this just asking if the s combinator is turing complete? bc wasn't there some challenge that relied on that not being true?
 
> the halting problem for SK combinatory logic is undecidable. However, neither K nor S alone has such property, and it turns out that the halting problem for each of them is decidable.
 
@cairdcoinheringaahing APL, 20: (g,≢+≢+g←⍪,≢+⌽)⍳⌈○≡⍬ Try it online!
 
9:44 PM
@Adám Similar to how this is the "canonical"/official page for J, is there an "official" page for APL (specifically APL in general, rather than Dyalog APL)?
 
@cairdcoinheringaahing J is easy because there's only one J implementation. I use apl.wiki as "the" website for APL in general.
 
@Adám Well, I was trying to decide between tryapl, apl.wiki and the Wikipedia page, so I think I'll use that, thanks :P
 
Is there a particular reason to require equal distribution? Just curious, I'm not too familiar with challenges.
 
@user Looks good to me.
 
9:53 PM
I'm sure what this comment says actually felt weird back then. Now it feels perfectly normal:
I am sort of waiting for a response like "f", 1 byte, in my math based golf language. — Zsolt Szilagy Aug 11 '20 at 11:57
 
@hyper-neutrino Not really, but I don't want people saying that the permutation it's already in has a 100% probability of being chosen and just doing id
 
@hyper-neutrino It's sort of the standard at this point. Allowing other distributions can be more interesting, but it also means you have to specify it clearly, or else someone says "Outputs 1 with p = 1 and every other integer with p = 0"
 
@user usually we use "every outcome must have a non-zero probability"
@cairdcoinheringaahing ^
@LuisMendo 2020 wasn't that long ago, was it? :P
 
@LuisMendo Doesn't even have to be a math-based golfing language nowadays :P
 
@LuisMendo Interesting that that was only last year
 
9:54 PM
@hyper-neutrino That makes more sense, actually
 
@user You can require nonzero probability for each (not necessarily equal probabilities)
 
Jelly had been only double that for 4 years before posting that
 
@cairdcoinheringaahing Oh. Sorry. I read that as year 2011 !!
 
@LuisMendo Yep, edited :)
 
9:55 PM
otherwise looks good :)
(proving each thing has equal probability can be difficult)
 
Great, lemme post to Codidact now :P
 
@hyper-neutrino I can't read :-D
 
@LuisMendo I read that as 11 can't read and started questioning my own ability to read
 
(or type)
 
(since if you're trying to choose a number from 1 to X, random.random() * X or whatever might not actually be even, and randrange(MAX_INT) % X isn't either)
 
9:56 PM
@user Because I mistyped, then corrected :-D
 
(so it's hard to necessarily prove your language's implementation is really random)
 
Fair enough
@hyper-neutrino Thank caird for that, I borrowed it from them :P
 
@hyper-neutrino Don't we have a standard for that? If your RNG says uniform, assume it really is
 
@LuisMendo probably. but it could still end up being annoying if it doesn't say that
jelly X doesn't say uniform but it probably uses python randrange, so ¯\_(ツ)_/¯
i don't think there's really a way to exploit this as long as everything must have non-zero probability?
 
9:58 PM
Anyway, requiring non-zero probability seems easier
 
Also just nicer for answerers, there's not much point in enforcing uniform randomness
Not sure exactly why I chose that in the first place
 
Yes, I meant easier for answerers
 
probably because it made sense to avoid "the initial permutation has 100% change" :P
 
with the inevitable XKCD link :-)
 
@LuisMendo Oh ok
 
10:00 PM
@LuisMendo Yes, we allow answers to assume their PRNGs are uniform
 
I remember the meta-post, I think it may be Doorknob's, but I can't find it
Ah, here
 
Honestly, that's probably an FAQ worthy post. I keep it favourited because I use the definitions quite a lot
 
0
Q: But, Is It Art?

caird coinheringaahingBut Is It Art? is an esolang created by ais523, where one step is break the program into orthogonally connected "tiles": A BBBB A B B AA CC A CC Each separate letter shows a different tile. Interestingly, programs in But Is It Art? only rely on the orientations of the tiles themselves, not thei...

 
Agreed. Which tag do we want to remove?
 
10:04 PM
probably
yeah this is only really specific to challenges
giving it proposed tag for the meantime
 
It's interesting that isn't mod-only. A new user could just add it to any and all questions they ask on meta
 
it isn't really an official tag right?
 
In fact, a few of the ones we rejected for the FAQ had the tag added by the OP
@hyper-neutrino No, but it's like semi-official
 
hm.
cuz the tag is able to die so it seems like something we just agreed to using or smth ¯\_(ツ)_/¯
 
Actually, that's slightly interesting about it
 
10:08 PM
@NewPosts i should add grid-graph-theory builtins to my language. maybe even as trigraphs or smth. :p
 
There are a few tags on meta that are unable to die (not just the status/required ones). For example: has never once been used (I mean that, there are no revisions where that tag is included), and yet is still in the list of tags
 
huh, and it's not a part of any synonyms either. interesting
 
, , as well, plus until I started adding it to relevant questions
Plus, if you go to any other SE meta site, they'll have those tags
@hyper-neutrino If we ever have that used here, I'll be shocked
 
i believe we have not :p
 
@hyper-neutrino Nope, never used
Although, apparently, I'm the only non-mod to add to a CGCC meta question :P
 
10:15 PM
what
ṁ doesn't work multidimensionally? :(
 
@hyper-neutrino Basically, the query looks at all tag edit revisions with [tagname] as one of the tags. I've edited the tags of a question after someone added [status-review] to it, so I got incorrectly credited with adding it :P
It's not perfect, but it gives a good metric of how often a tag has been used historically
 
@cairdcoinheringaahing oh i thought it was an SE glitch :p lol
does jelly have "vectorize but cut off at the shorter list"?
(like ")
 
Don't think so
 
:c
it's really annoying TBH
 
That could've been a good use for ' :P
 
10:22 PM
idek what that does
i don't think i've ever used that for either arity
 
It works for monads (stops them vectorising Ṛ = U'), but it's not great for dyads:
in Jelly, Sep 22 '17 at 21:06, by Dennis
@DJMcMayhem ' is useless. The original intention was the flat quick. For example, <' should compare two lists as Python would, not element by element. That works for monads, but it's rarely needed anymore. For example, ¬ already has the flat version . Because of a giant screw-up when I implemented it, ' is executed once for each pair of elements of the two arguments. This is a bug; þ should be used for this behavior.
But it stays for backwards-compatibility
 
oh ._. rip
 
That's a hack I am 100% using, as it's one link rather than ẋ2 :P
 
uhhhh
what sort of behavior is this?
 
10:28 PM
??
 
@hyper-neutrino 2,1 chaining: tio.run/##y0rNyan8/…
Add an ŒṘ - you've got some singleton lists there
 
doesn't 2,1 not really chain in a dyad?
like isn't +F supposed to be F(+(v, y))?
 
2,1 does the whole +(v, F(x)) thing in monads
 
Ah, I got it
The ," creates [[1, 5], [2, 6], [3, 7], [4, 8], 9, 0]
 
10:30 PM
oh i see where this is going
 
When you each, it removes the first of each element, except on 9 casts to range
 
And ṫ€1 doesn't work :/
 
anyway, is ,"LỊ$Ðḟ{ ... }/€ɗ really the shortest way to do " but without the extend thing?
 
@cairdcoinheringaahing But ṫ€2 does
 
10:32 PM
-1 byte with ,"L’$Ƈ{ ... }/€ɗ
anyway i gotta go for now. but i'll try to solve Is It Art in jelly when i get back if Nick hasn't gotten to it :P
 
@hyper-neutrino ,LÞḢṁ@<dyad>
disgusting, but it works
 
10:44 PM
@ophact As I said before, TIB requires reimplementing every single language in Rust, and there's no OS to back up the general-purpose languages. RustPython has a slightly more chance though
@user The entire point is not having the server, so it's a no starter
 
:(
 
11:20 PM
morning y'all
 
it's technically morning for me too I suppose :P
 
Same here I guess :P
 
@lyxal I would say "MORNING MOTHERF*****S," but it's evening here
 
CMQ: What is your definition of "morning"?
 
@Bubbler when I still remember that I'm tired
 
11:31 PM
From sunrise to just before noon
 
@Bubbler Anything from 00:01am to 11:59am
 
@cairdcoinheringaahing What about 00:00:01 am?
 
Or whenever I wake up - I will greet people with "Good morning" even if I wake up at 6pm
 
^
 
@user Thems the timeless lands
 
11:32 PM
@cairdcoinheringaahing Now that you mention it, I've never seen 00:00:01 on a clock before...
 
You're just not looking fast enough :P
 
Also, the clocks on the wall are analog, and it takes me half a minute to read them :P
 
imagine having walls
of course I do have walls but still
 
This post made by transcended to heaven gang?
 
no, welaze gang
fun fact: vgeetarian wealzes live on croutons alone
 
11:35 PM
Not for long
 
are waezles all bad at spleling?
 
No, just you two
 
@rak1507 they are sometimes bad at spleling Egnlish
 
0
Q: Can I license my controller for a KotH under the GNU GPL v3?

4D4850I've just finished making a controller for a KotH challenge, and I wanted to know: Can I use it for the KotH and license it under the GNU GPL v3, or does it have to be licensed with the MIT License to be used for a KotH on CGCC?

 
11:36 PM
@NewPosts Does it matter?
 
GPL is viral, but yeah, it won't matter
Also requires them to distribute source code, but again, it doesn't matter
 
Theoretically, you're the only person whose going to run the controller (you could even just not publish it), and unless you release it on SE (i.e. include it in the question), whatever license you choose will apply
 
The KotH code is not part of SE code so nothing will happen I guess
 
11:57 PM
ya know what'd be nice
if suggested edit review tasks showed who wrote the answer
I just overrode an accepted review task because someone accepted a golf via suggested edit
but only later realized it was the OP themselves which is fine
 
Ah yep, thought that might happen if you or WW saw it :/
 
and TBH it's only because I rejected and then realized the user who accepted it shouldn't have enough rep to access that queue so I realized it had to have been OP for that to work
 

« first day (3785 days earlier)      last day (1045 days later) »