« first day (2865 days earlier)      last day (1994 days later) » 

12:53 AM
0
Q: "As we travel the universe..."

ArnauldYou'll be given the name of one of the 20 biggest objects in the Solar System. Your task is to return an approximation of its radius, expressed in kilometers. This is a code-challenge where your score consists of the size of your code (in bytes), increased by a penalty based on your worst approx...

 
1:22 AM
@Arnauld I really do like the challenge concept
 
@Arnauld What was wrong?
 
I think moving to sandbox was a good call
 
1:35 AM
0
A: Sandbox for Proposed Challenges

Arnauld"As we travel the universe..." You'll be given the name of one of the 20 biggest objects in the Solar System. Your task is to return an approximation of its radius, expressed in kilometers. This is a code-challenge where your score consists of the size of your code (in bytes), increased by a pe...

 
@xnor Done :) For now, my only change is that the maximum error is set to 95% (and the linked script is checking that).
 
1:48 AM
@Quintec With the original rules (without the bound on the maximum error), very small programs returning utterly bad approximations would get a penalty close to 100% (size * 2) and still win other better programs. (Even worse: an empty program would score 0.)
 
2:02 AM
Ah, I see.
 
But I'm still not sure about the current rules. Allowing a 95% error is probably much too high. I think 50% was actually a pretty good trade-off. Signing off for now. We'll see tomorrow...
 
 
1 hour later…
3:34 AM
codegolf.stackexchange.com/questions/176897/… Am I the only one not understanding what the question is about and wanting to close the question as unclear at all?
 
You're better off asking for clarification before voting as unclear
 
@H.PWiz I know I just had the impulse to do so only
I mean it seems to require quite a long time to fully understand what the challenge is doing
 
I agree that the challenge is too large
maybe following that EBNF and being fully modular should have been two separate challenges?
 
 
2 hours later…
6:01 AM
I like it when you write something that shouldn't work, and doesn't work when tested alone, but when you "fix" it, everything else that shouldn't have been working originally breaks.
 
 
1 hour later…
7:21 AM
@primo I'm ahead again on the Sudoku challenge. Your move :P
 
Is there anyone here who has used Clean before? I have to run the submissions for the Prime Containment Number (speed edition) challenge, and I keep getting the same error during compilation
 
Hello
@maxb What is the error?
 
@Οurous Oh, hey! I get an error that the modes can't be found
Though I'm thinking that maybe I had the wrong clm installed
 
@maxb Probably. It isn't in any package system I'm aware of, you'll want to get it from either ftp.cs.ru.nl/Clean/builds/linux-x64 or github.com/Ourous/curated-clean-linux
 
I tried doing clm ... and got a suggestion to install a package from ubuntu, I'm fairly sure that suggestion was something completely different
@Οurous Now I've got it installed and linked to usr/bin. Though I'm getting import errors
`Compiling main
Error [main.icl,3]: Data.List.dcl could not be imported
Error [main.icl,3]: Data.Func.dcl could not be imported
Error [main.icl,3]: Data.Maybe.dcl could not be imported
Error [main.icl,4]: Text.dcl could not be imported
Error [main.icl,4]: Text.GenJSON.dcl could not be imported`
 
7:39 AM
@maxb did you export CLEAN_HOME=/path/to/clean/directory?
If you grabbed it from the ftp server, did you get clean-bundle-complete... or something else?
 
Perhaps that was the wrong way?
 
No that's certainly one of the right ways to do it. Did you get clean3.0_64.tar.gz or clean-bundle-complete-linux-x64-latest.tgz or something else?
 
the first option
then i unzipped and ran make
 
Ah. Yeah you want the second one and then follow the `readme` it contains.
The first one is completely bare-bones, the second one comes with normal libraries.
 
7:54 AM
oh, okay. I'll try that
 
8:32 AM
@Οurous Now I'm getting a strange error: clm: can't start the clean compiler: No such file or directory
 
@maxb Do you get any output before the error?
 
it seems as if I'm failing to create the symlink in usr/bin
╭─[09:42:57][max][isabella][±][master U:1 ?:12 ✗][~/Documents/misc-scripts/codegolf/prime_containment_clean]
╰╼ which clm
/usr/bin/clm
╭─[09:43:00][max][isabella][±][master U:1 ?:12 ✗][~/Documents/misc-scripts/codegolf/prime_containment_clean]
╰╼ ls -lha /usr/bin/clm
lrwxrwxrwx 1 root root 39 Dec  3 09:41 /usr/bin/clm -> /home/max/clean-bundle-complete/bin/clm
╭─[09:43:06][max][isabella][±][master U:1 ?:12 ✗][~/Documents/misc-scripts/codegolf/prime_containment_clean]
╰╼ clm -fusion -b -IL Dynamics -IL StdEnv -IL Platform main
oh wait
I'm stupid
I literally had export CLEAN_HOME=/Users/username/clean in my .bashrc...
@Οurous Now it compiles without any errors, but I don't get any output files.
 
8:51 AM
@maxb Nothing at all? Doesn't even make a Clean System Files directory?
 
@Οurous I get that directory (I even got that before when the compilation failed), but nothing else
 
@maxb Try appending -o ~/main.out to the command?
 
@Οurous Still nothing, can't find the file
 
@maxb Make a new file test.icl and just put module test;Start="" in it, see if compiling that makes any output.
 
still nothing
 
9:05 AM
It's either broken or doing something silly. Check in clean-bundle-complete and /usr/bin for the output files?
 
@Οurous I think I found the problem
somehow clean/bin/clm got erased into a 0-byte file
so everything "ran" without errors, but did nothing
now I got the file!
 
Oh good now I don't have to remember how to do a native compiler build from source :)
 
@Οurous Haha, good thing. Thanks for the help! Now it's executing, currently a score of at least 24
@Οurous I'll probably do the real test later today, as I'll have to close down everything else to guarantee a suitable testing environment.
 
@maxb Awesome, thanks for letting me know! Will this be final scoring or can I poke you in a few days when I've gotten a chance to add multi-threading and better memory usage?
 
@Οurous I'm happy to allow improvements from all competitors, so feel free to ping me anytime. I haven't timed it, but a(25) definitely popped up before 5 minutes passed!
 
9:41 AM
0
Q: Computer Cipher

Kevin CruijssenIntroduction: I have loads of different ciphers stored in a document I once compiled as a kid, I picked a few of the ones I thought were best suitable for challenges (not too trivial, and not too hard) and transformed them into challenges. Most of them are still in the sandbox, and I'm not sure ...

 
 
1 hour later…
10:41 AM
@JoKing don't make me switch to Perl now :p
 
@primo noooooooo
it's depressing how far ahead your perl 5 solutions are compared to perl 6 for the longer challenges
 
your perl6 spelling numbers is quite impressive, though
...which you recently improved by 8 bytes o.O
 
11:00 AM
@primo Part of that was a new compression method. It does seem a bit futile to struggle against Lisp's builtin though :(.
uniname helps a lot
oh, and I realised that all the hexadecimal numbers were actually making my code longer, as opposed to just using base 10 lol
@primo I think you mean 11 bytes :P
 
11:30 AM
i concede defeat :p
back at 97 there was still a glimmer of hope
 
 
1 hour later…
12:48 PM
@Οurous I'm happy to announce that you're in the lead for the Prime containment numbers (speed edition) challenge! With parallelization, I think you could reach n = 26 within the 5 minute limit.
 
 
2 hours later…
2:51 PM
Just as a reminder guys, we have a PPCG team for Halite III
 
3:20 PM
@lirtosiast I agree. Validating the EBNF is a good challenge by itself and it seems like we didn't have such a challenge before (not sure about that, though -- I haven't looked very closely). The fully modular criteria are more laborious.
 
4:07 PM
Really hoping my teacher's tests don't go up to Integer.MAX_VALUE because my code depends on a % b == ((a + 1) % b -1) % b
Maybe I should make them longs
 
4:20 PM
> A QA Engineer walks into a bar. Orders 1 beer. Orders 13 beers. Orders 0 beers. Orders -1 beer. Orders INT_MAX beers. Orders jdsheuwiod beers.
12
> First real customer comes into the bar. Asks where the restroom is at. The bar explodes and everyone dies.
 
4:33 PM
0
Q: Digital Sumorial

OliverGiven an input n, write a program or function that outputs/returns the sum of the digital sums of n for all the bases 1 to n. $$n + \sum_{b=2}^n \sum_{i=0}^\infty \left\lfloor \frac{n}{b^i} \right\rfloor \bmod b$$ Example: n = 5 Create the range [1...n]: [1,2,3,4,5] For each element x, g...

 
 
1 hour later…
5:45 PM
0
A: Sandbox for Proposed Challenges

MatthewProgressive APEG - ASCII Pictures Expert Group Sometimes, you get stuck on a poor connection, and even ASCII art seems to take forever to load. When you just want to see if it's worth loading the full picture quickly, wouldn't it be nice to have a progressive ASCII art format, which gives you a ...

 
6:03 PM
0
Q: Simple pseudorandom permutation for bytes

user84345I'm looking for an implementation for a function f that: takes as input a byte and returns as output another byte is a permutation of the 256 possible values 0, 1, ..., 255 of a byte makes use of few simple operators like: addition, multiplication, bitwise and, or, xor, bit shift; and few memo...

 
 
1 hour later…
7:23 PM
19
Q: Fastest Mini-Flak Quine

Post Left Garf Hunter 1000000000 and I are currently offering a indefinite bounty of up to 1100 reputation (details here) for a solution to this challenge that beats the current best Mini-Flak is a subset of the Brain-Flak language, where the <>, <...> and [] operations are disallowed. Strictly speaking it must ...

 
 
1 hour later…
8:37 PM
0
Q: If we find a programming language in the Library of Babel, can we use it?

PyRulezThe Library of Babel is a huge library with many books. For example, the book titled ",fkwpival.vbfz" defines code golf on page 29 code golf is a type of recreational computer programming competition in which participants strive to achieve the shortest possible source code that implements...

 
8:49 PM
0
Q: Golf a number bigger than Loader's number

PyRulezAs a follow up to Shortest terminating program whose output size exceeds Graham's number and Golf a number bigger than TREE(3), I present a new challenge. Loader's number is a very large number, that is kind of hard to explain (since it was itself the result of a code golfing exercise with a fle...

 
9:00 PM
@JoKing have you seen this? code-golf.io/leap-years#perl6
 
Oh nice
though I think the definition is wrong...
 
Only the description. The output is right
 
J seems to have an advantage for "Fast" holes.
 
It's good at most of the mathematical constants aswell
 
Hm, score is by number of UTF-8 characters. Let's get APL onto that!
 
9:08 PM
> These longer years occur in years which are multiples of 4 but not multiples of 400.
Uhmmm.... No? It's multiples of 4 but not multiples of 100 (except for multiples of 400)
 
@DJMcMayhem Right.
 
@Adám NARS, to be specific.
 
@Zacharý Except trains!
 
@Dennis Still one of my all time favorites
 
9:13 PM
@Zacharý How about my extended Dyalog?
 
It already has its extended codepage, no?
Or are the other challenges UTF-8 bytes
 
@Zacharý code-golf.io uses UTF-8 characters to score.
 
There are no UTF-8 characters. UTF-8 is an encoding.
5
 
Thus, just unicode then, no?
 
some languages are defined in terms of the characters that make up the program, some in terms of the bytes that make up the program
that can lead to the header being formatted differently in the two cases
if the language is defined in terms of characters you have to encode it into bytes to be able to score it…
 
9:15 PM
> How Are Solutions Scored? The score of your solution is the sum of the UTF-8 characters in your source code.
 
…so TIO finds an appropriate encoding and lets you know how many bytes the program would be in that encoding
@Adám that doesn't make sense, how do you add characters? add their codepoints?
 
why did you make a "temporary account" ais?
 
@ais523'stemporaryaccount I also don't know what it means, but I guess you can't use characters that UTF-8 cannot represent. ⍨
 
@Adám They probably mean Unicode characters.
 
9:17 PM
@Adám I wonder if you could get a score of 0 by writing the program entirely using byte sequences that are invalid UTF-8
some languages are Turing-complete even with that restriction, such as Incident (obviously), and possibly Jelly
 
@Dennis I'm sure they do. And by sum, they probably mean count.
@ais523'stemporaryaccount Incident?
 
No clue. Where does it say that?
 
Someone should open an issue.
 
9:20 PM
@Zacharý I added prime number functions today, together with an (⍺+)i×⍵ function.
 
I guess what they probably mean is "Your submission must be valid UTF-8. We will decode it into Unicode and count the resulting number of characters."
 
@ais523'stemporaryaccount That's also meaningless. You can't decode it into Unicode. Unicode is a computing standard. The Universal Character Set is a list of characters, not an encoding. UTF-8 is one of many Unicode encodings.
 
@Adám you can decode a sequence of bytes (UTF-8) into a sequence of characters (Unicode)
"characters in the Universal Character Set" would be a more precise way to explain it, I guess, but I don't believe the original is meaningless, it's a plausible abbreviation
the point is that UTF-8 is a concrete representation, whereas Unicode is an abstract representation, but they're both representations…
 
@Dennis I've made a PR.
 
@Adám s/in UTF-8/when encoded as UTF-8 bytes/
maybe
 
9:29 PM
the existing language gives an example that clarifies that each character only counts as 1, even if encoded as multiple bytes
we know from experience that that's a bad idea, but…
 
Are there any generic-language applicable guides to concurrency using POSIX system calls? (Coming from a windows-centric developer viewpoint so I have a good understanding of multithreading but not really of POSIX)
 
there's a man page that lists guidelines as to what thread-safety possibilities are available per-language, give me a moment
 
@Adám Let me guess, prime uses pi.
 
@Οurous see the glibc manual starting here; they've put in the work to understand thread-safety rules already
 
@Zacharý No, as I didn't want to kidnap Greek letters. I use as "not-divisible". Also, it should look like a (semi-)scalar function. No Greek-letter functions are scalar, whereas | is scalar and ~ is semi-scalar.
 
9:34 PM
"libc" is a bit of a misleading name, really, given that pretty much every language (not just C) uses it to interface with the kernel via POSIX APIs
 
Maybe use a double struck pi? :P
 
although the examples will normally be given in terms of C
 
@Zacharý Pi is used for the prime counting function, but this one is so much more, and the monadic for is factors anyway.
 
Wow.
 
@ais523'stemporaryaccount thankyou very much
 
9:41 PM
I should disclaim that GNU are documenting the current behaviour of the functions in question on GNU-based systems (most Linux distributions are GNU-based); if you want to know the guarantees that POSIX give, they're listed here
 
10:31 PM
Hmm, if you had to glue together all the iconic challenges together, what would it look like?
cat 99 Bottles of Fizzbuzzes on the World!
Wait, cat 3.14 bottles of fizzbuzz. Can't forget the pi challenges
 
you could easily combine fizzbuzz and 99 bottles of beer
fizz bottles of beer on the wall, fizz bottles of beer; take one down and pass it around, 98 bottles of beer on the wall
I'm not convinced it would be an interesting challenge though
 
I'm 95% certain 99 fizzbuzzes does exist as a challenge already
 
oh, changing the "beer", not the number
 
Net effect is the same
 
perhaps you could ask for a CHIQRSX9+ interpreter, that'd be at least a fairly standardised way to combine all the standard tasks
allowing for any definition of X that makes the language Turing-complete
 
10:36 PM
@ais523'stemporaryaccount pretty sure it exists
 
I'm not trying to make a language, just hash out what the all encompassing output would look like
 
no questions about implementing it, just a few answers in it and a few questions banning it
@Veskah write, I'm thinking in terms of a challenge
"write a program that shows off how your language does all these simple tasks" expressed concretely in terms of a particular task that requires you to do all of them
although I isn't particularly simple
 
Ah
Making it quine related feels almost necessary but wonkey
 
implementing Q doesn't require writing a quine
it just requires remembering the program you're running and being able to output it again
 
10:52 PM
Hello, Ð!
 
11:27 PM
0 10925
1 10930
2 10830
3 11048
4 11033
5 10762
that doesn't look uniform, does it
 
Given no context, not really
 

« first day (2865 days earlier)      last day (1994 days later) »