« first day (1896 days earlier)      last day (2947 days later) » 

9:00 AM
@KennyLau @MarsUltor Thank you both for your answers
 
@zyabin101 program it as A lipogram! \o/
It will be a liprogram
 
@Katenkyo Uhh, I'll create a Stack Exchange chat room. :/
 
@zyabin101 Yes, but won't you put a bot in there that will alert when someone used the banned letter?
 
@Sherlock9 If you want to calculate entropy that way, it's going to be really slow
 
I think my friend was just curious
 
9:05 AM
Maybe the number of moves using a well-defined method would be better, and a lot faster
 
And I'm aware that the algorithm is humongously expensive
 
@Katenkyo Nope. I'll let the community alert. I'll pin an alarm message (for example, *BUZZ*) and the community uses it whenever someone uses the banned letter.
 
@Sherlock9 It's not even really an algorithm, more like brute force
 
Brute force counts as an algorithm :D
2
It's a sequence of steps. It's just a really inefficient sequence of steps
 
@zyabin101 we can't use it if the link to it contains a bad letter D:
 
9:08 AM
@Katenkyo Uhh, then only parsed messages count?
I mean, the content of the message you see in chat.
 
@zyabin101 I think everybody would do it this way ^^
 
@_@
Also, I can't serve a bot all day.
 
@zyabin101 Why not?
 
@MarsUltor I have to sleep, as does my computer.
 
Your computer sleeps? o.O
 
9:13 AM
@MarsUltor Nope, I just shut it down.
Uhh, I have to go to school.
 
Hmm
No spare computer?
 
Nope.
 
@zyabin101 no spare server?
 
@Katenkyo nope
 
@zyabin101 maybe you could afford something like that it should be enough to run a chat bot :D
 
9:22 AM
Only 1PB ram and 3 million cores? psshh
 
@MarsUltor it's 1024 TB RAM :D
 
Anonymous
I'd hate to see the price tag for one of those suckers
 
@Mego I think the price for one of it's processor ( a 12 core) should be something like 1500-2000$... even only with processor, it becomes enormous
 
My guess is each Xeon processor is around 500
(as a generous guess) That means $130m just for the processors.
 
I found a page where cpu are compared with equivalents -> cpu-world.com/CPUs/Xeon/Intel-Xeon%20E5-2692%20v2.html
(scroll down)
Would expect it to be around $1500 ^^
 
Anonymous
9:35 AM
So what I'm getting out of this is that it costs a small nation's GDP to build one, and then another's to run it monthly
 
@Mego Yeah, some billions for a supercomputer... Wow
 
the*
 
maybe we could use this to test the golfscript/CJam etc answedr that are supposed to run until the dawn of our solar system
 
@Katenkyo -4.6 billion years? That would be handy.
 
@MarsUltor never mix up your words, never... -_-
 
 
1 hour later…
10:48 AM
1 year later...
We're in 2017. Oops.
 
Mathematica:
Count[Normal@SeriesData[x,0,Reverse@RealDigits[EulerGamma,10,#][[1]],0,#,1]~NRoots~x,_Real,{1,2}]&
 
any mathematica user want to post an answer to codegolf.stackexchange.com/questions/77051/… ? It's one line :)
I sadly don't have mathematica
anyone have Maple here?
 
11:18 AM
@Lembik Looking at it now; trying to find the correct vector.
 
Why is no one here?
 
The Americans are mostly asleep
 
ATTENTION PYTHONISTAS
Python 3.6 is now interesting for golfing!
Because of s="strings"; f"Format {s}!"
 
2
A: Tips for golfing in Python

LynnUse f-strings Python 3.6 introduces a new string literal that is vastly more byte-efficient at variable interpolation than using % or .format() in non-trivial cases. For example, you can write: l='Python';b=40;print(f'{l}, {b} bytes') instead of l='Python';b=43;print('%s, %d bytes'%(l,b))

Having said that, silly Windows won't let me compile it yet :/
 
@Sp3000 should be no problem with msys2
 
11:27 AM
Hmm will check it out, thanks
 
11:52 AM
I will get 3.6 when it's out for full release and some modules have been updated for it, since I am a n00b and I tend to break things to begin with
 
To all followers of Zalgo:
in RegEx - Regular Expressions on Stack Overflow Chat, 8 mins ago, by nhahtdh
Black magic here .split("(?=-)(?<=[^-]*-[^-]*-[^-]*)-"). The regex just checks that there are at least 2 dashes before the current dash and split at that dash. The look-ahead is to avoid costly look-behind if the next character is not dash. I have checked the compiled structure and the regex is compiled correctly. However, the code may break in the future when they fix the bug in Java, so I strongly recommend against this method. Just split the string and rejoin them as appropriate. — nhahtdh 1 min ago
cc @MartinBüttner
 
12:39 PM
@MartinBüttner We should have a Retina course, Unary arithmetic in Retina.
 
@zyabin101 I think I'd be up for writing that... I wanted to add a tutorial to the Retina wiki anyway at some point (although that might have to wait until the language is a bit more feature-complete)
 
@LegionMammal978 the vector is all 0s and of length n
 
How far does the tips scope reach? Would "tips for golfing unary arithmetic" be on topic?
 
Tips for writing tips questions.
 
12:57 PM
We found boiled crawfish yesterday for $4/lb and it included potatoes (but not corn). Also I got engaged.
 
@QPaysTaxes At least one person speaking directly above me was going to compile it themselves, so definitely not everyone
 
@Rainbolt Congrats on the crawfish!
3
 
Thanks
I had three lbs of spicy and it was amazing
 
@Rainbolt Wow, really? Nice!
 
1:05 PM
@QPaysTaxes IDK, mostly that colloquial English is sometimes weird
 
s/colloquial//
 
It's more lazy than weird. Or at least, even non-colloquial English is weird.
 
English is weird.
@Geobits ninja'ed.
 
English is weird.
 
@QPaysTaxes I am rather literal. I say “most"
 
Thinking of my lips as little flaps of skin grossed me out for a moment
 
And we set standard meanings on some sounds, other people throw other meanings at them. Other languages, other dialects, colloquialisms, etc
 
@QPaysTaxes No weirder than moving tubes of skin to manipulate machines to communicate for you.
 
Spotted my obvious mistake
Ignore me...
 
1:08 PM
@QPaysTaxes I know. I am :P
 
I love puns, though, and playing around with languages makes you a good programmer and a good golfer
 
English is a lot weirder (well, has more Exceptions) than almost all languages
 
@Lembik However, Mathematica's definition doesn't take the negation of the transpose of the vector.
Also, apparently, the matrix must have an imaginary part.
 
Trying to call crush.date() and getting SlapInterrupt due to poorly formatted inputs
 
@QPaysTaxes like parent or similar?
 
1:10 PM
@Rainbolt congrats on the potatos.
 
Thanks
 
@QPaysTaxes Usually parent is used for the enclosing scope, child is used for enclosed scopes
Hmm
 
But isn't the only other free symbol _?
 
Something up-related? ^ maybe
 
1:12 PM
JS console has $x, $0, $_, $$. Maybe $p and $c?
so then ^ and v?
 
Just go with mom or dad :D
 
@LegionMammal978 you need i P/pi for the matrix
 
Is this a golfing, eso, or general-purpose language?
 
But won't ^ be used for xor or power?
 
@Geobits s/so/soteric/
 
If it doesn't need to be golfy, why not go with something clear?
2
@MarsUltor s/soteric/so/
 
maybe reserve $ for system names?
$p/$c
 
hi @TonHospel
 
or $^ and $v
 
1:15 PM
Nothing really matches $ though. Not a single character anyway.
 
@Lembik: Hi
 
Go Unicode. Then you can use other currency symbols. ‎£ is a good one for parent >_>
 
and ¢ for child
 
@TonHospel how things?
 
1:18 PM
@QPaysTaxes I know. I just mean you can make all scope-symbols currencies. Choose your favorites ;)
 
@QPaysTaxes But that's British currency
 
@Lembik: Trying to improve the error estimate
 
@Geobits parent = USD, child = ???
 
@TonHospel ooh.. that sounds exciting
 
I have to say
PPCG makes me feel 👌
 
1:19 PM
boxy?
 
@TonHospel I found a paper which is too complicated for me to understand. But if your math is strong maybe you can grok it?arxiv.org/pdf/1005.2632v1.pdf
the math is beyond me :)
 
@Geobits install a proper font
 
@Lembik: Frustrating. There are matrices where the current bound stops too soon (though your generator doesn't seem to generate them)
 
but it might solve exactly my challenge
 
@orlp Use a proper symbol :P
 
1:20 PM
@Lembik what are you working on this time?
 
@Lembik as always?
 
@orlp true.. arguably these are a little bit different
 
@Lembik oh wait this is still the same problem
@QPaysTaxes 👌
 
@orlp oh ok.. I wasn't sure how far back we were going
 
1:21 PM
@EᴀsᴛᴇʀʟʏIʀᴋ :D
 
@orlp there is some literature on this topic it seems but I can't understand it as my math is too weak
 
@Lembik: That does not seem applicable. It has i in the exponent, so it's the oscilating part of the exponential
 
@TonHospel but ... when we use the SeigelTheta function in mathematica for example, we just pass i P / pi . As in mathematica.stackexchange.com/users/2305/martin-b%C3%BCttner
@TonHospel so the sum becomes real
 
@Sp3000 actually, MSYS2 is not supported by Python
but I had no problems compiling with VS2015
 
@orlp how is your math?
 
1:24 PM
@lembik: You can't do that for the formulas in the paper since it says the polynomial has integer coefficients, so no place to smuggle back a i
 
@Lembik I dno?
discrete? :P
 
@Lembik: The proper relevant paper is arxiv.org/abs/nlin/0206009
@Lembik: I was implementing their error estimate, but it's way too conservative
Though they have a good way to remain in the ellipsoid instead of using a box like I do now which I'll definitely implement. It will likely completely compensate for their pessimistic error bound
(as far as I know all implementations in these math packages are based on this paper)
 
1:46 PM
0
Q: Sure, it's a CRDT!

Filip HaglundCo̶ń̶͞͡҉f̛̀l͜į̷͏c̷̶̡t̡̛͡-̛͢͢͞͠f͟͏̀҉r̴̛e͏͘͞͠e͢͞҉ Replicated DataTypes are the biggest hype in distributed computing since sliced bread. CRDT's are great for eventual consistency, since they converge to the same state no matter how you apply new elements, as long as all elements are applied at le...

 
@NewMainPosts What's with the Zalgo? Anyone else seeing that?
 
Yep
 
Yeah.
I just commented.
Go upvote it if you don't like the zalgo.
 
The first word of the question is Zalgofied.
 

« first day (1896 days earlier)      last day (2947 days later) »