« first day (2052 days earlier)      last day (2805 days later) » 

12:05 AM
@quartata xD
 
12:28 AM
@Qwerp-Derp yes?
 
to all bot lovers: I've made a new room for Laffy called the Moonbase and he'll be in it full time once I get him transitioned to C9.
 
Can you name him something other than Laffy?
 
@quartata keep him with marky, its funnier when there are two bots saying non-sequiturs to each other
 
I will give a 50 rep bounty to anyone who can figure out what laff-o-tron and the moonbase is from in 24 hours
2
The room topic may help you
@Maltysen He'll stick around there too
 
@HelkaHomba I've got a challenge
 
12:35 AM
(if someone would star the bounty notice I'd appreciate it)
 
done
 
I will give out a few hints on request.
 
@ASCII-only I've got another bug:
var length_digit=00
circ successor(n)->[
    cond ((~n)<)->successor((~((~n)>))+0)/(~((~n)>)+1)
]
out successor(length_digit)
 
@quartata commodoreHUSTLE
 
This doesn't work
 
12:37 AM
@Maltysen Fuck
 
And returns a KeyError: circ
This is the full log:
 
All right I'll award it in a bit
 
@quartata the trick to googling it was replacing laff-o-tron with laugh-o-tron
 
Traceback (most recent call last):
  File "logicode.py", line 526, in <module>
    Run(code)
  File "logicode.py", line 483, in Run
    result = statement[0](scope)
  File "logicode.py", line 77, in <lambda>
    return lambda scope: scope[result[0]]
  File "logicode.py", line 217, in __getitem__
    return self.parent[key]
KeyError: 'circ'
 
@quartata nah its ok
i don't think that rep was really meant for this stuff
 
12:39 AM
@Maltysen That is strange considering laff is how it's supposed to be spelled, at least in canon
 
@quartata I'm not following this conversation, but is this just a version of chomskybot?
 
Not sure what that is. I've parsed chat's transcript into a BK tree of messages and their replies with Hamming distance as the metric
 
Does this mean we don't need to chat anymore?
 
The bot just takes a message, cleans it and looks up matches. The maximum threshold is number of words * 2
@trichoplax No no keep chatting. I need more data
 
The bot is hungry? Brings a whole new meaning to "chat feed"
 
12:43 AM
@HelkaHomba Can you make a primality checker in MC?
 
holy shit python is getting static return types for functions
 
@Qwerp-Derp MC can run a turing machine
 
Yeah
But isn't it hard to make a primality tester in MC?
 
@Qwerp-Derp so was the question "is it possible", or "do you know how?" -- I think I could figure out how; primality testing by divisibility is usually pretty simple even in screwy languages
regex golf has a primality testing challenge, and that was weird, but it's solvable there too
so I guess I don't know how, but I'm pretty confident
 
12:46 AM
Ah
It's definitely possible
I'm challenging him (and other people) to make one
 
oh. if it's an actual challenge.... hmm. i think clock circuits and comparators might be enough. On Helka's server, or in theory?
 
Actually in Minecraft, so I would go for Helka's server.
I would go on MC and try it myself, but I forgot my password
 
should the input be in unary or binary
 
Binary would make the /2 check easy
 
I think I can make a sieve of eratasthones.
if you line up blocks against the sieve output (lit redstone lamps), that solves it for unary
 
12:50 AM
So yeah
 
I think? I'm working through in my head what counts as solving this
 
So a positive signal indicates that the number is prime
And a negative signal indicates that its not
 
so a line of lamps with all of the prime number lamps lit up (not hardcoded) counts?
 
You can easily AND them all together to make one signal, so yeah that counts
 
Well, if you're feeling cheaty, you can always use command-block scoreboard arithmetic and convert to/from binary...
 
12:53 AM
@ΛεγίωνΜάμμαλϠΨΠʹ I can't use command blocks on Helka's server
 
oic this is on a server
 
Actually, I can't think of a way to encode divisibility without wiring the "2" lamp to 4, 6, 8, etc., and so on
so I don't know how. I might think about it later
 
There, just fixed my chat username, now it's LegionMammal978 instead of LegionMammal1680
 
@EricTressler I've wanted to try this but it is tricky
 
Try what?
 
12:58 AM
@HelkaHomba Yeah, the more I think about it, the worse it seems. But I'll think about it some more
 
@Qwerp-Derp I've built things that can simulate rule 110 so if you can write a prime checker in that (except it would probably require 100's of players to keep all the chunks needed loaded :P)
 
True
People have made Turing machines in MC
I wonder if it makes making a primality checker in MC any easier
 
oh, no
I'm pretty sure you wouldn't want to add another layer of indirection (use a turing machine to do it)
 
yeah
 
there's a chance that the prime checker could be smallish, and I just haven't figured out how yet
okay, helka: here's my idea that's too vague to use
We have positions for 0 - 100
0 initiates a signal that hits 2 - 100. based on timing, they repeat the pulse. 2 should interfere later with every multiple of 2
 
1:02 AM
@ΛεγίωνΜάμμαλϠΟΗʹ It's still showing the username from your parent profile (since it's not set to PPCG) just to let you know
 
if you can make that work for 2 and 3, I think finishing it would be easy enough.
 
@trichoplax ik, I set the new one on my productivity.SE account, blame caching
Of course, the PPCG username is still LegionMammal978
 
Oh well... just waiting then I guess
 
Note to self: On Ubuntu, Ctrl-Alt-F5 opens a new fullscreen tty that can't be exited except by restarting the system :/
(Just happened to me while trying to clear cache)
Note to self v1.0.1: ^\  can and will stop the init process
Note to self v2.0: Removed clutter
 
1:12 AM
For anyone with the expertise, we have a search space that needs roughly quantifying:
1
A: Sandbox for Proposed Challenges

Qwerp-DerpThere is only 1 Your task is, given a positive integer n, to generate an expression that equals to the number n. The catch is: you're only allowed the number 1 in the output. The operators at your disposal are: +, -, * and / sqrt (as s) ceil and floor (as c and f respectively) ! (factorial) ...

To decide how much time to allow to prevent brute forcing
 
Yey
 
I guess it would be easier to quantify starting from the optimal length expression for the most difficult input, but I'm not even sure that the most difficult input will be the largest number, so this might mean working out several...
 
Hmmm... so we get ceil, floor, sqrt, and factorial... all unary operators, but since the score is based on numbers of 1's they're all basically free?
 
If we're looking for an optimal score, then it will include the tie break based on number of operators too
 
100 is obviously way too small; the optimal expressions can probably be found.
Can factorial be called on a non-integer?
 
Strictly no... there are factorial like extensions... but you can ceil and/or floor anything and then apply it
 
@EᴀsᴛᴇʀʟʏIʀᴋ no, its gamma(x+1)
 
@feersum yes, if you use the definition n! = gamma(n + 1)
 
1:27 AM
@Maltysen oh, i thought that was for negatives
 
Then you can floor/ceiling
 
Needs to be defined in the challenge whether that's an error or an assumed floor @Qwerp-Derp
 
@ConorO'Brien I know that's the standard extension of factorial. This is a question for the challenge author.
 
there's also a nice special case with (n + 0.5)!
 
Which one?
 
1:28 AM
@feersum I evidently lack context, sorry
 
Also, you might need to specify division by 0
 
Ah
Division by 0 is an error
 
@ΛεγίωνΜάμμαλϠΟΗʹ otherwise 1/(1-1)-1/(1-1) :P
 
@Maltysen (1-1) / (1-1)
 
"(1-1)/(1-1)"& ← all solutions :p
ninja'd
 
1:29 AM
@ConorO'Brien Trying to decide how large an input and what time limit to prevent brute forcing this sandbox post
 
The scoring should sample from the entire range that the function is supposed to work in.
 
You are not allowed to do a division by 0
If you do, you are disqualified
 
Until the division by 0 is gone
 
Hmm... maybe pick some random numbers from (0..2^31) for scoring?
 
1:30 AM
@quartata from this?
 
Ummm
But that makes it tedious to score
Because I have to test everything
 
Maybe scoring should be sum of 1's and numbers of operators (ignoring the implicit concat one)?
 
@math-knowing-people: would it be correct to say that the atan(1) series for pi converges linearly since by alternating series bound the error would always be less than the last term which decreases linearly?
 
@Qwerp-Derp Not necessarily, just put a few numbers in the post to score on
 
Ummm, sure
So numbers like 11111111 and 123123123 and stuff like that, right?
 
1:32 AM
By putting a number like 1111111, you're begging for solutions to optimize for the test cases.
 
True
So just a random number generator, then?
 
That would be better.
 
@EᴀsᴛᴇʀʟʏIʀᴋ Maltysen got it in like 5 minutes, it's commodoreHUSTLE
 
Should I put it in the post now?
The random numbers?
 
1:34 AM
@Qwerp-Derp Also, please rename you challenge "There can only be one!" :3
 
There can be only one
 
@ConorO'Brien lol
 
highlander was a great movie
 
1986???
Wow that was a long time ago
 
oh my
 
1:36 AM
Have you actually watched it? It was a little before your time so I'd be surprised
 
The time from 1986 to my birth is longer than my birth to now
 
yes I have I like that movie
 
Yeesh
 
it's one of the few movies I watched and remember liking
 
Should I put the random numbers into the post now?
 
1:37 AM
@ConorO'Brien If you liked the original you should watch its 4 appalling sequels
 
oh my sweet goodness four of them??
 
please do not
 
how do you even make a sequel to that movie
 
(heavy sarcasm - because they were nothing like the original)
 
If there's only one then what comes next? Fractions?
 
1:38 AM
they are not gud
 
There can only be 1/2, there can only be 1/4....
 
Highlander Aleph_0: a curious tale of high landers
 
There was a series that wasn't bad at all... but the sequels were pretty disappointing
 
Awrite, I updated the post
 
@Qwerp-Derp Do you want to exclude negative square roots too?
 
1:53 AM
Yeah
 
Do you need the explicit exclusions since they won't evaluate to the correct number, which seems to make them invalid already?
 
Oh, yeah
 
I think it would be understood that giving an undefined output and saying "it could be anything, so it counts" is a loophole
I think the one that needs specifying is factorial, as people mentioned earlier. It can be defined for non integers but I'm guessing that wasn't your intention
 
@Qwerp-Derp Wait.. you're limiting the results to 2^31-1; that's obviously to equalize languages. But if that's the intent, shouldn't intermediate results also be limited?
 
2:00 AM
Good point. That will affect the optimal expression in some cases
 
Intermediate results?
???
 
Taking the square root of something bigger than 2^31-1
Can the minus sign also be used as negation, as in -1?
Maybe we should try to think up expressions you probably don't want, so you can add relevant rules...
 
@Doorknob Do you know if there is a formal grammar for Lojban floating around somewhere? EBNF would be nice
 
It might be possible to do everthing using two 1's with a gigantic chain of factorial and sqrt
Do you want that?
 
2:05 AM
c((s(-1)!*111)) = 56
@Qwerp-Derp an example with things you might not want to be possible ^
 
@trichoplax lel
 
Actually I chopped off the imaginary part, which I'm not sure ceiling will do...
 
Octave and Mathematica ceil the real and imaginary parts both separately.
 
ceil(sqrt(((1+1+1)*(1+1+1+1+1))!))=1143536; uses 8 1's (though there's a shorter way, ignore that), but 15! is a 37 bit number (example of intermediate result being bigger than 2^31-1)
 
True
OK...
 
2:17 AM
> Your score is then determined by your output:
> It is (number of 1's)*(number of operators)*(byte-count)
@Qwerp Meaning byte count of your code?
Or the expression?
 
@trichoplax Yeah, problems like that come when trying to order complex numbers...
 
I don't think ()'s should play into it at all
(Implied by "byte count of expression")
 
Byte-count of code
 
@trichoplax well it's called the least integer function
 
Oh, and I changed the scoring system again
 
2:18 AM
@Qwerp-Derp Ewwwwww
 
What why?
 
Would not participate.
 
Why???
:(
 
Then it's optimal to write a 10-byte program in a golfing language that only tries addition.
 
But I changed the system agian
 
2:20 AM
try for 5-byte program
 
Check the post
How do I remove the intermediate limit?
 
8 bytes in an okay golfing language
 
Hello?
 
I personally think the scoring system needs work but don't have specific suggestions... it's kind of difficult finding the right balance
 
@feersum I'm actually attempting to get 56 in this way.
 
2:26 AM
You mean the current scoring system?
I think it's alright
 
@ConorO'Brien 5: j\+m1
 
@trichoplax This paper suggests that it would be 56 - 17i
 
@El'endiaStarman Fun (but no longer encouraged by the challenge).
 
56=ceil(111/(1+1))
 
So far, I've determined that it would take more than 120,000 operations using a slightly-naive method.
Hmm. Maybe more like mostly-naive. :P
 
2:29 AM
@El'endiaStarman what method are you considering?
 
any Jelly heads about?
 
@El'endiaStarman i imagine that most of the operations would be sqrt since it grows much slower than !
 
Hello
 
Hello
 
@Maltysen Well, if one can manage to get 27!, then four square roots and a floor will arrive at 56. Thus, the challenge is to get 27, which can be done with 113! and then 7 square roots. To get 113, then, etc...
 
2:31 AM
can't get asleep a_a
 
@El'endiaStarman where did you get the 120,000 number from?
 
@El'endiaStarman 27 = ceil(sqrt(6!))
 
>>> def foo(k):
	print(k)
	m,n = 1,0
	x,y = k,k+1
	i = 1
	while 1:
		if i < x: m += 1; i *= m
		elif i > y: x *= x; y *= y; n += 1
		else: return m,n


>>> foo(56)
56
(27, 4)
>>> foo(27)
27
(113, 7)
>>> foo(113)
113
(264, 8)
>>> foo(264)
264
(539, 9)
>>> foo(539)
539
(189, 7)
>>> foo(189)
189
(512, 9)
>>> foo(512)
512
(3560, 12)
>>> foo(3560)
3560
(15492, 14)
>>> foo(15492)
15492
(118388, 17)
>>> foo(118388)
118388
 
@Qwerp-Derp ah, you are there.
 
@feersum nice
@feersum but it uses three 1's
 
2:33 AM
You need at least 3 ones to kickstart the process anyway, right?
 
No, because you can write 11.
 
link to what the chatter is about?
 
@feersum And 6 = f(s(11)), which means that the full expression is then 56 = f(s(s(s(s(a(c(s(a(f(s(11))))))))))), where a(n) is n!.
 
Oh, 11, didn't see that
 
@El'endiaStarman oic
 
2:35 AM
In[96]:= Floor@Sqrt@Sqrt@Sqrt@Sqrt@Factorial@Ceiling@Sqrt@Factorial@Factorial@Ceiling@Sqrt@Factorial@Floor@Sqr
t@11

Out[96]= 56
 
Ohh, I did a similar thing with fives and a more limited operator set before.
 
2
A: Sandbox for Proposed Challenges

Qwerp-DerpThere can be only 1! Your task is, given a positive integer n, to generate an expression that equals to the number n. The catch is: you're only allowed the number 1 in the output. The operators at your disposal are: +, -, * and / sqrt (as s) ceil and floor (as c and f respectively) ! (factor...

 
The discourse has since moved into its own room:
Sep 3 at 19:41, by trichoplax

 Numbers from 5

trying to bruteforce numbers from 1 to 100 using only the numb...
 
I did it with 4s, with my friends a few years ago:
20
Q: Using floor, ceiling, square root, and factorial functions to get integers

El'endia StarmanSo, several days ago, I was introduced to the "Four game" again. The object of the game is to use four 4's to produce as many integers as you can. You are allowed to use addition, subtraction, multiplication, division, square root, and factorial functions in addition to concatenation (44), use of...

 
2:38 AM
@Qwerp-Derp To put my discomfort in the scoring system into words... I'm imagining what an "interesting" solution looks like, and what a "lowest scoring" solution looks like, and am not quite sure the latter fits the former. That's pretty fuzzy, but that's the balance act.
 
@quartata I have never used Java from Octave, but you can get the screen size with pure Octave (or Matlab) code:
get(0, 'ScreenSize')
 
@El'endiaStarman That should actually be 56 = f(s(s(s(s(a(c(s(a(a(f(s(11)))))))))))). I forgot a factorial. :P
 
We should try a new "Numbers from 5" ruleset.
 
@El'endiaStarman @flawr, have you seen this question I linked to?
 
here's a question idea
remember my integer calculator question
now try to generate the shortest representation of a number in that form
like 1023 would be like ++**********-
and 4 would be ++*
and 3 would be +++
 
2:49 AM
can someone explain to me what is wrong in this?:
brainflak/parser/parser.h:25:69: error: expected ')'
    unordered_map<string, parser::rule*> *toLanguage(parser::rule[] rules);
                                                                    ^
brainflak/parser/parser.h:25:53: note: to match this '('
    unordered_map<string, parser::rule*> *toLanguage(parser::rule[] rules);
 
it expected a )
 
@Downgoat Maybe you forgot a semicolon?
 
@EᴀsᴛᴇʀʟʏIʀᴋ ... i know that
 
@Downgoat Don't use array types in function parameters.
 
@Downgoat exactly
 
2:49 AM
@feersum why not?
@El'endiaStarman nope :/
 
@DmitryKudriavtsev Good idea!
 
Thanks, posting it now
 
If you do, it actually means a pointer.
And it's a stupid way to write a pointer.
 
@DmitryKudriavtsev Wait wait, Sandbox it first!
 
no, I didn't actually sandbox the calculator challenge
 
2:51 AM
@feersum Though AFAIK array length is compile-time meaning it's lost when it's passed to function. so then how do I keep length?
oh I use a vector...
 
Is there a Ruby equivalent of Python 2's translate()? All I'm getting in Google is questions about Google Translate and translating Ruby.
 
@Downgoat Wait what is this
 
@DmitryKudriavtsev Try googling "tr" instead of translate.
 
@DmitryKudriavtsev The point of Sandboxing is to get feedback before posting. You have more than a dozen comments on your calculator challenge. Mostly stuff that could've been worked out in the Sandbox.
 
2:54 AM
ok
 
@ASCII-only Im writing parser engine in C++
 
can someone link me to sandbOX?
 
@Downgoat Wait what kind of parser engiine
 
200
Q: Sandbox for Proposed Challenges

SandboxWhat is the Sandbox? This "Sandbox" is a place where Programming Puzzles & Code Golf (PPCG) users can get feedback on prospective challenges they wish to post to the main page. This is useful because writing a clear and fully specified challenge on the first try can be difficult. There is a much...

 
2:55 AM
@ASCII-only Brain-Flak.
 
I'm playing Andor's Trail now. :3
 
thanks @El'endiaStarman
 
@Dennis It's larger than it seems.
 
3:00 AM
@LuisMendo Figured it out, thanks though
@Doorknob awsum thx
 
The initial baton is sold for a whopping price of 47 gp.
I bought myself an iron axe and a round shield.
And a pat on the head as a gift. :P
 
For eating should I make au gratin potatoes, fried potatoes, or potato fries? Hmmm
 
au gratin
 
n.b. that for those I have less milk than is called for so they might turn out weird (and they take like 2 hours)
 
I was just picking which one is objectively the tastiest from the list. I didn't realize you were going to make and eat it right now I guess
Potato fries then?
 
3:09 AM
Well, maybe I am, if I have the energy. Either way talking about potatoes is fun
 
I hope this is a side and you aren't just making a giant plate of fries for dinner
 
y so judgmental
 
@feersum I mean, I'm not really judging. But I just hope that if he has the means he'll make a salad too
 
potatoes are vegetables. a plate of fried potatoes is a fine salad
 
3:15 AM
@Dennis - any suggestions on these to long pieces?
1
A: Counting Goats to Sleep

Jonathan AllanJelly, 62 60 bytes ⁶ẋ6;“\ ___/o>-(___)" '' '' ”s8 ¢“-,o-'`”yЀ 1£ẋ€¢ż@Y N~Ç Test it at TryItOnline

0
A: Esolang-Comment-Template-Generator

Jonathan AllanJelly, 41 bytes This looks like it has far to many increments, and probably too many links... ṫø⁹‘‘µFL‘⁶ẋ ‘ị ḣFL⁶ẋ$;ç@;1ŀ J’ç@€ ṣ”|Ç;€Y Test it at TryItOnline How? ṫø⁹‘‘µFL‘⁶ẋ - link 1: get the spaces for after the code, dyadic(split string, index) ø - next chain as a nilad ⁹‘...

 
i... guess technically that's correct?
 
(two)
 
i just had an idea for a goat challenge. i need to jot this down
 
I meant more a salad of the non fried leafy green variety though
 
goat challenge made me smile
 
3:17 AM
@HelkaHomba Ping me for new AA videos, I'm going to publish these in TRB, with transcripts of mini challenges :3
Oh, and also, I'll invite you to the Remaining Bytes.
 
There's a cool thing called a subscribe button
 
I already subbed.
But messages none.
 
So you have a grid of goats. They faint when they're startled and scream when they wake up. any goat that's within distance 2 and awake when a goat screams immediately faints. Someone startles one of the goats. Find the state of the goats the next day
 
I don't read email (except when really needed (i.e. email verification)), so...
 
3:19 AM
@Downgoat is this what happened
 
@zyabin101 Just subscribe. I'm not gonna remember (or bother) to ping you
 
what song was it
 
I love a bit of goat-core
 
Goatstep
 
3:20 AM
yeah, on the side
good for raves on cliff sides
 
skrillex heh
 
Drums and Goat
Goatronic
 
that song ^^^^ is amazing
 
oh god both playing at once :/
 
needs more cowbell
 
@JonathanAllan I'll take a look tomorrow. I have been working on TIO all day, and I'm exhausted.
 
yeah no worries, you dont have to do anything!
more girls, and more Bruce Lee is required...
 
3:29 AM
@HelkaHomba Well, I did sub, but messages none.
Except very regular email that I don't read.
 
@JonathanAllan This thumbnail is quite inappropriate.
 
It is finished
k.o
 
:/
no age restriction on yt
...and Bruce Lee playing table tennis with nunchucks, it can't be wrong
 
hello and good morning to some
 
3:34 AM
some people I see in this chat regardless of time :)
 
FWIW, I tried to invalidate that flag, but I was too late. Certainly the thumbnail could be considered offensive, but the poster doesn't choose the thumbnail.
 
good way to start the day by answering a question, I've been doing only edits of my posts in the last 3 days
 
I didn't find the video full of redeemable qualities, but I think one would have to be a true prude to actually find it offensive.
 

« first day (2052 days earlier)      last day (2805 days later) »