« first day (3664 days earlier)      last day (1169 days later) » 

1:55 AM
I see that [new sandboxed post] currently only post each sandboxed post once.
What do you think about the idea of making a bot to repost it 8 times, each separated 3 hours to account for different time zones?
 
Most people who are active in TNB will read through the previous chat logs, I'd think
Plus most sandbox feedback probably comes either from people who regularly browse the sandbox, or after people ask directly in chat
 
Except that sometimes some people talks too much and it's impossible to read them all.
There isn't...
we need some motivation for people to work on the sandbox.
 
Spamming it in chat won't help though
 
... a button to "go to your last unread message", too.
There are definitely more active users on main.
Perhaps just posting them to main but adding [SANDBOXING, DO NOT POST ANSWER] for a week is not a bad idea.
 
That is a bad idea though
 
1:59 AM
it spams the main page for a little, but as long as all of them are converted to real challenge after a while (or dismissed as "bad idea" it isn't a problem.
Also require people to finish the challenge (don't include TODO in it).
What else could the problem be? That people post answers to it anyway?
 
The problem is that many users simply don't notice the sandbox or know how to contribute.
 
I think a similar thing was proposed before (maybe multiple times), including "a post would be closed right away before it gets enough feedback and it gets reopened"
 
Yes I read that somewhere too.
Can't find it though.
The close proposal have the problem that some mod must run a script all the time.
 
and IIRC it was rejected due to the SE-side implementation issue
 
Yes that's the issue. [SANDBOXING, DO NOT POST ANSWER] doesn't have that issue.
 
2:04 AM
Clogging main with drafts is worse than sandboxing in almost every way; you ruin the signal-to-noise ratio, take reputation away from bad sandbox posts (dicouraging its use), and wreck the site and user stats by having a ton of closed, potentially downvoted/deleted posts that never went anywhere
 
I said "require people to make the challenge complete (to their knowledge)".
 
Sure, a few more people will notice it, but I'm sure more people will be aware of drunk driving if you drive an 18 wheeler into their house
@user202729 Before sandboxing?
 
Currently the sandbox requires that too, but it's true that sometimes I don't really do that.
@RedwolfPrograms that's another SE implementation issue...
Can't be helped.
Another idea would be to have one single post on main that is like a "sandbox posts that needs review" notice... wait there's already the ... wait where did it go to?
 
Except all the sandbox posts kinda need review, right?
That's why they're there?
 
Increase popularity, I guess? Not all people read the sandbox?
 
2:08 AM
The main problem from what I can tell is that the majority of users don't know or care about the sandbox (from a reviewing perspective). Any efforts made should foxus on that.
 
Do well still add the every week?
(To the sandbox)
 
There's links to the sandbox all over the sidebar, I doubt another will do anything at all
 
For now I add a [please review other sandbox posts] to my sandbox comments but...
I see that too.
 
I honestly don't think there's too much wrong with the sandbox. It's not perfect, but it's by no means broken.
 
Afternoon y'all!
 
2:28 AM
Guess I should find a simple recurrence for bubbler's challenge in the sandbox to get a fgitw.
 
Well, if you manage to find one, it must be a substantial improvement on the OEIS entry
 
Not that it's easy to prove such things anyway but...
 
2:44 AM
Looks like there's really no easy way.
 
3:24 AM
By the way, I proved that the [minimum cost matrix permutation] problem is NP-complete. Not sure what it's called.
And I'm scraping the site for BF programs, but perhaps scraping other sites is easier.
 
3:47 AM
Huh, I'm CAPTCHA-blocked now...
(as expected.)
 
4:08 AM
@user202729 Wait, are you not using the API and/or SEDE?
 
4:28 AM
0
A: Sandbox for Proposed Challenges

BubblerElias omega coding: encoding code-golf binary encode Background Elias omega coding is a universal code which can encode positive integers of any size into a stream of bits. Given a positive integer \$N\$, the encoding algorithm is as follows: Start with a single zero in the output. If \$N=1\$, s...

 
5:11 AM
@RedwolfPrograms Let's say that I'm lazy to learn new things...
 
 
1 hour later…
6:14 AM
SEDE seemed... somewhat easy to work with the last time I tried?
I think I was trying to count how many times some small sequence of characters had been used in Jelly
and I think it worked
 
SEDE's easy enough if you download the CSV and do the actual work with python or JS :p
 
I think the post body is given in the rendered HTML
Unless the Unicode chars are eagerly replaced with HTML escapes by the Markdown engine, searching for things like Jelly digraphs indeed should work
 
6:41 AM
Any feedback on this and this?
 
7:06 AM
(something I learned today: you can change the MathJax renderer to speed it up / reduce CPU usage.
 
Btw, you don't need to comment every sandbox post you review. You can simply upvote it if you don't see any issues and you think it's a good challenge
 
8:08 AM
I know that but other people upvote for interestingness too so...
Better make it clear.
I might actually make a approximation problem with dc z|P output.
It's terribly hard, to get the optimal solution.
Although we might end up with someone implement some nice heuristic that gives result instantly and nobody can prove why it's optimal.
 
8:34 AM
CMQ: How did you find the code golf stack exchange?
Was it through the HNQ list like me or by some other means?
 
@Lyxal Yeah, HNQ on judaism.SE
 
8:50 AM
@Lyxal I don't really remember. Maybe it was from a HNQ on Stack Overflow, maybe it was me finding problems to solve in various languages, maybe it was me searching specifically for code golf sites
And occasionally I feel deja vu in all three
 
I want to say it was HNQ from Stack Overflow, but my SO account is two months younger than my CG&CC account, so it definitely wasn't from when I asked a question
 
9:13 AM
@Bubbler How did you hear about code golf in general?
 
I know I first heard of it just browsing Wikipedia, but got the impression that it was mostly just something people do in Perl and never really looked to get in to it
 
@UnrelatedString Perl APL :-)
I must have learned the term from here, but I grew up with code golf.
 
When I was 12 or so, I out-golfed the shortest APL expression for "primes until N".
4
 
9:34 AM
@Adám I think I learned the term on Codewars, where one user was occasionally creating JS code golf challenges
At least it was definitely not Perl nor APL to me
 
10:04 AM
@Lyxal by mistake
2
 
1
Q: Create a Language of The Month answer scraper

RazetimeWhen I was posting answers for the Husk Language of The Month thread, I realized it took more time to enter them in the thread than it took to actually write the answers. So, why not automate it? Write a program/function which inputs a language name, and displays all answers in that language duri...

 
10:47 AM
@Lyxal pretty sure it was HNQ for me too
 
 
1 hour later…
12:03 PM
0
A: Sandbox for Proposed Challenges

SjoerdPenningsIs this a valid Mudra combination? code-golfstring In the MMORPG Final Fantasy XIV, the Ninja class has the ability to use combinations of up to three handsigns (Ten, Chi and Jin) to perform a variety of skills. The skill you cast depends on the last sign used, and using two or more of the same s...

 
 
1 hour later…
1:33 PM
2
Q: Which Ninjutsu am I casting?

SjoerdPenningsIn the MMORPG Final Fantasy XIV, the Ninja class has the ability to use combinations of up to three handsigns (Ten, Chi and Jin) to perform a variety of ninjutsu skills. The skill you cast depends on the last sign used, and using two or more of the same sign makes the ninjutsu skill fail and puts...

 
ngn
©NGN lol
 
2:11 PM
@Lyxal Probably an HNQ on Stack Overflow
CMP: How many of you think CGCC (or just code golf in general) had a good impact on your life?
4
 
@user definitely me, because it led me into being interested in "real" coding
 
2:29 PM
@ngn niceeee
@user Yep, definitely think so
 
@user Probably the only thing that keeps me sane (other than school)
 
@user I don't actually know if code golf is having a good impact on me. On the one hand, I enjoy it, on the other, my "real" code's quality seems to be getting worse and I feel like I'm addicted to it :P
 
@user I think it has.
 
@user I would say yes, just because it‘s a fun place to go to if I‘m bored, and the community‘s always been nice
 
 
1 hour later…
3:50 PM
@user I also think it has
 
It makes me happy to see all these comments saying code golf's been a good impact
Although this is probably a biased sample :)
2
 
are there any esolangs with lists/arrays/vectors as the only datatype?
@Wezl yes
 
@Wezl There's Stack Exchange, but it uses stacks, not lists.
 
@Wezl link?
 
@Wezl github.com/ysthakur/StackExchange (disclaimer: I started working on it a few days ago)
It's still in development, so it won't be too useful to you right now.
 
3:56 PM
@user survivor bias :)
 
@user useful? that's not what I'm looking for :)
 
@Wezl Pretty sure APL/J only use lists/arrays as the data type, but I'm not sure if you'd count them as "esolangs"
 
@cairdcoinheringaahing I mean, there's also numbers and stuff
 
I'm kinda looking for inspiration bc I'm brainstorming my own lang
 
@user Yeah, but aren't they treated as single element vectors?
 
3:58 PM
@cairdcoinheringaahing sometimes
 
@cairdcoinheringaahing Oh, I didn't know that
 
@user Not 100% on that, but I think that's the case
 
but I'm looking for something where the only datatype is type List = [] | List : List
 
@Wezl That's Stack Exchange, basically.
Here's the definition of the data type Stack:
```
enum Stack:
case -:(head: Stack, tail: Stack)
case SNil
```
Basically data Stack = SNil | Stack -: Stack
@Wezl Here's a room if you have ideas to share or you want to get ideas, although pretty much no one uses it
 
What's a good variation on a prime test such that it still identifies primes, but isn't very easy/nice to use? e.g. returns 1 for primes and squares, 0 otherwise
 
4:04 PM
@user SE looks different from what I'm thinking of because stacks are treated differently because you push and pop instead of index and concat
 
@Wezl Ah, that's too bad
 
well what I'm thinking of could probably be trivially added to or on SE
 
@cairdcoinheringaahing Maybe take the list of factors and get the second element?
So for composite numbers, it'd just be some factor other than 1, and for primes it'd be the number itself
 
@user So it returns n for primes, sqrt(n) for squares and a number between n and 1 for composite numbers? Very nice :P
 
@Wezl I don't know about indexing, but concat can certainly be implemented
@cairdcoinheringaahing "nice" isn't the word I'd use for it, but okay
 
4:12 PM
@user indexing wouldn't be with numbers, to: stack .index. i :do: if: i = [] :then: stack .top :else: stack .pop .index. i .pop ; ; where _ .pop represents _ after being popped from
 
@Wezl I'm not quite sure I understand - does the length of the stack represent a number?
 
@user yes, in what I was planning 0 is [], 1 is [[]], 2 is [[], [[]]], so addition can be union and subtraction can be difference
 
Ah, ok
Yeah, I could certainly add that to Stack Exchange later - it'd be pretty useful
@Wezl Sorry, I don't understand again. Is it a combination of length and depth?
 
should have said all this in dedicated SE chat
 
@Wezl Oops, you're right :)
 
4:20 PM

 Hatred

A room for the programming language Hatred, a language specifi...
Feel free to pop into this room and make suggestions for things to add to Hatred!
 
5:07 PM
1
Q: Can Whispers programs ignore a final output line?

caird coinheringaahingThis was first asked by Leo in the Whispers chat room: But I have a general question about Whispers submissions: couldn't we avoid counting the final Output line most of the time? We can consider the rest of the program a function returning the desired output, printing it to the screen with a fu...

 
 
2 hours later…
6:55 PM
That's an answer choice on a practice AP test lol
 
7:06 PM
-1
Q: How can i convert this from this language to java?

How Everthis is the code (idk what programming language is): rbind(cbind(M1,matrix(0,nrow=nrow(M1),ncol=ncol(M2))),cbind(matrix(0,nrow=nrow(M2),ncol=ncol(M1)),M2)) this is what i wanna do in java can someone help me to convert it?

 
 
1 hour later…
8:10 PM
@NewMainPosts If we delete that, the next question to main will be the 12000th
 
8:20 PM
I just looked at how many questions SO has, and apparently it's 20,809,331. That's a lot of questions
 
@cairdcoinheringaahing So wait, was that the 12000th question?
 
@RedwolfPrograms The 12000th non-deleted question
 
Kind of underwhelming I guess :p
 
That's why we gotta delete it :p
 
Hang on can I borrow 11550 rep to help out :p
 
8:31 PM
Don't you only need 10k to delete negative questions?
 
Oh, I didn't know that
Wait, why is 4 % 0.1 = 0.1? Isn't the remainder of 4 / 0.1 = 0?
 
.1 | 4 is 0 in APL
 
@RedwolfPrograms Yup ^
 
@RedwolfPrograms Because the behaviour of % for non-natural numbers is often language dependent
 
Weird, I wonder what JS and Python are doing...
 
8:40 PM
Lots of languages have badly implemented mods
 
8:50 PM
@RedwolfPrograms what do you mean by 0.1?
 
The usual meaning I guess :p
1 / 10
 
well, in that case, you would be right
but if you meant "the nearest floating point number to the rational 1/10", then you're actually getting 0.1000000000000000055511151231257827021181583404541015625
and obviously 4 % that isn't going to be 0
 
Oh, true
Hooray for floating point
Such accuracy, very decimal
 
Hm, even if I tell APL to use exact 64-bit float computations, I still get 0.
 
9:40 PM
Good 0 <= hour < 12 everyone!
 
Wait so you consider 04:00 to be day and 13:00 to be night?
 
4 am is definitely day
 
Okay 02:00
 
2am is late night imo
 
@RedwolfPrograms 4am is not day and 13 is not night :/
 
9:43 PM
Okay I'm definitely making a JS library for octuple precision floating point
It's like double precision floating point but more
I want all the precision
 
@RedwolfPrograms Power-of-2 precision is boring, quintuple precision is the newest fad :P
2
 
I actually think there was a computer at some point that used that
No wait that was 20 bits, not 20 bytes lol
 
20 bits is a little less precision than 20 bytes, I think.
 
I think once you get above quadruple precision there's no reason not to just use arbitrary precision :/
I want to make an "arbitrary precision" library that just chooses an arbitrary level of precision every time you run it and makes you use that
 
@RedwolfPrograms * once you get above 0 there's no reason
 
9:53 PM
@RedwolfPrograms You could use octuple precision to fool your users into thinking your language has arbitrary precision, and then trip them up when they come to rely on it
@RedwolfPrograms @cairdcoinheringaahing You reading this? More ideas for Hatred!
 
@user I'm not writing a whole library for a language :P
If Redwolf makes it in Python, I'll happily use it :P
 
@cairdcoinheringaahing You mean 80-bit precision? That one's pretty old actually
 
@cairdcoinheringaahing pypi.org/project/Js2Py (I don't know if it'll work for this, though)
 
Got to love having an O((N^2)!) solution :P
 
An On^2! solution to what?
 
10:01 PM
A challenge :P
 
@RedwolfPrograms 13:00 afternoon
04:00 morning
 
O((N^2)!) does mean that for N = 3, it takes around 10 minutes for a single test case :/
 
lol
 
Arnauld's answer (the only other answer) completes the N = 5 test case (and others) in less than 2 seconds ಠ_ಠ
 
which challenge?
 
10:14 PM
^
 
The one I'm about to post an answer to, which I'll keep a secret to not get sniped :P
Nevermind, just answered :P This one
 
11:08 PM
Dang, lost a 200-worth again
 

« first day (3664 days earlier)      last day (1169 days later) »