« first day (2546 days earlier)      last day (2300 days later) » 
00:00 - 16:0016:00 - 23:00

12:44 AM
@Mr.Xcoder also most haskellers like to use div, mod, elem as infix functions because it's easier to tell which argument is which (5 `div` 2 is more obvious than div 5 2)
 
TIL this is even a site
 
1:35 AM
I did extrmely well on the ACT, but my GPA is kinda crap, so now I'm getting all this mail from univirsities I don't actually have a chance of getting into inviting me to apply :(
 
How well?
I got a 33? Or a 32? I don't remember exactly
 
@DJMcMayhem 36 in science, 35 on the rest.
 
1:55 AM
35 on the math, crappy everything else. I will keep retaking until I get that 36
 
Oof. That sucks
 
I mean crappy by my standards, they were all low-to-mid 30s
So, I've heard about the Broccoli freak-out
 
@Zacharý ¯\_(ツ)_/¯
 
Yep
 
@Zacharý ?
 
2:03 AM
 
Finding the original creator, and the overall atmosphere about it
 
Wait but what's so magical about Broccoli
 
@MDXF IDK, they mentioned it on the podcast and it's a PPCG meme now.
 
If only every dead language got that love
 
@Pavel Ah, fun
 
2:18 AM
@Pavel GPA really doesn't matter, I mean your grades should be generally B or above but if you have exceptional ability and passion I don't think you should hesitate to apply to top universities
 
@Downgoat Obviously I'll still apply everywhere, but :/
 
Somehow, my weighted GPA's still above 4. I have no clue how
 
:O thank you for reminding me that weighted GPA is a thing
I suddenly feel better about myself.
 
:)
Let me guess, everything's about .5 to 1 point higher?
 
@Zacharý Every class my school offers is either honors or AP.
 
2:22 AM
@Pavel ._.
 
...
 
What kind of school is that, eh?
 
@Zacharý Over here we have certain schools that you can apply to and get in by lottery called choice schools. They're generally much better than the regular public schools.
 
Ahhh, that makes sense as to how everything's honors or AP. But I would expect that they would offer classes that are normal, for people who SUCK at something (like my writing)
 
@Zacharý They used to, but only 1 or 2 students would apply every year, so there was no purpose keeping them around.
 
2:27 AM
Oh, 1 or 2 applied to the lower-level courses?
(Hopefully not the school)
 
@Zacharý Yeah
 
Ah
 
And obviously everyone knows who that is, since it's a small school, so there's a lot of pressure not to.
 
The fact that a group of people would pressure for something like that, seems very foreign to me. Where I live, ..., let's just leave it as that sounds foreign
 
Anonymous
2:44 AM
@Pavel Sounds like magnet schools in the US
 
@Mego This is WA. I imagine it has different names in different places.
 
Anonymous
Oh, yeah
 
Anonymous
In Texas they're called magnet schools
 
in MO I've never heard of them .-.
 
@Mego Must just be southerners?
 
2:55 AM
Any good C++ golfers in the room?
 
I can golf other's C++ code decently
 
I've already golfed the methods down, I'm just looking for various source improvements. gist.github.com/aaronryank/8aa359b14864dd999184a63313f185c8
You know, #define F for and such
 
Ooooh, macros. I love doing those
... until I realized how big it was
 
Yep, currently 1600-ish bytes
:)
 
aren't #defines shorter than typedefs?
oh wait they're the same size
 
2:57 AM
They're different, #define doesn't require a semicolon
 
... just noticed I have two typedefs
 
yep lol
 
Wait, no, #define requires a newline
Same size
 
But ... I 've been ninja'd
 
3:00 AM
157 bytes of array declarations ... _ O[]={0,262143,259263,74943,74898},A[][8]={{0,1,2,3,0,1,2,3},{4,7,6,5,4,5,6,7},{0‌​,9,4,8,0,3,5,4},{2,10,6,11,2,1,7,6},{3,11,7,9,3,2,6,5},{1,8,5,10,1,0,4,7}};
 
do you need to add the _ at the beginning? doesn't gcc assume int by default?
 
O[p]&(1<<m) => O[p]&1<<m. And everything like that, operator precidence
 
@NickClifford not in C++
 
((s[e]>7)?2:(s[e]&1)): 0/10, too many parens
:p
 
3:02 AM
Hmm how could I golf that down without pulling up five guides about C++ operator precedence
 
The latter one is just common sense
 
Can I safely remove parens from <<(2*e)?
 
Especially not needing parens around s[e] > 7
@MDXF Yes]
2*(m%3) => m%3*2
I actually love doing this, for some reason.
 
It's rather fun but I'm out of practice, I've been golfing in (my) esolangs lately
 
I mean I love golfing order of operations and macros on other people's code a bit too much.
 
3:06 AM
Yay under 1600 bytes
Why can't the parenthesis from (i&3)==3?i-3:i+1 be removed?
 
Because & is evaluated after ==, for some reason. The way it does work helps for golfs as well, when chaining equalities, & can be used instead of &&
 
Odd but thanks
Why doesn't this compile? It's fine in C
 
Because unicode. (I have no clue)
 
3:21 AM
@MDXF C++ is generally stricter than C. You can't do things like printf without #include <stdio.h> in C++ either.
 
@Mr.Xcoder Who's the line that surpasses Chuck Norris Jon Skeet?
 
I'm really curious about ^ too
 
Answer: Jon Skeet
/s
 
[] + [] anyone?
 
3:47 AM
I got two points towards epic without even hitting 200 rep once.
wth
 
@HeebyJeebyMan Do you have Mortarboard?
 
No, I already had it
 
I do, and I've never hit 200 rep, ever.
 
I mean yesterday I had 39 points twoards epic and today I have 41, but I only earned 185 rep yesterday.
 
._.
Good for you I guess
 
3:51 AM
I guess, I feel kindof like I'm being cheater
 
4:06 AM
1
Q: Must askers be able to answer their own question?

user202729 It might be useful to keep the question undeleted. That way it serves as a way to more easily find the question it is a dupe of. - Heeby Jeeby Man The question is all in the title: Must askers be able to answer their own question? EDIT: As PeterTaylor pointed out below, this question ...

(with some edits this time) Anyone can take a look?
 
...
 
Hey, SO has new notification! (or not-so-new)
 
not SO new
 
@MDXF By negating the conditional (using <8 instead of >7), you can swap the order of the branches, and avoid the parenthesis. That gives you (s[e]<8?s[e]&1:2)
 
4:22 AM
(can I edit duplicate list?)
 
Anonymous
@user202729 Only mods and users with the gold badge
 
Anonymous
And since nobody has a gold badge for discussion, that leaves just mods
 
These two answers 1 2 are basically the same, with the authors having each copied the other post with a trivial golf.
 
 
1 hour later…
5:35 AM
0
A: Sandbox for Proposed Challenges

SteveFestAverage Alphabet code-golf Introduction We can average a number using this formula: (1st term + 2nd term + ... + nth term) / no. of term But unfortunately, this does not apply to alphabets Challenge Your program should take an input as a string, which is one-line & alphabet only. Other ...

 
5:48 AM
@Mego Well, that's as of now/
 
How's PPCG going chat?
Still no design :/
 
6:41 AM
Is there a shorter way to write this program? Try it online!
(except removing whitespace and use other function equivalent to print)
 
7:26 AM
To Whichever Mod It May Concern: Please unfreeze the tinylisp room. Thanks!
 
It was an old question and in the oldest form it doesn't really conform to the current rules. (e.g. 10 seconds on which machine?) Anyway, reopened.
 
Yeah, I've removed the parts that no longer work with our rules (such as the time restriction)
 
Apparently I can edit comments on deleted posts...
(can anyone have enough rep on SO to view deleted posts check this?)
I wonder why SO have the [ ](stackoverflow.com/questions/tagged/pi) tag...
^ why doesn't the formatting work?
 
Anonymous
7:57 AM
@user202729 The formatting only works for plaintext. is a link already.
 
Anonymous
8:14 AM
Chat just died for a few minutes. I almost started rioting and looting.
 
@LeakyNun Probably on "bra", but stress really should be irrelevant
 
Anonymous
@Fatalize That is a strange message without context
 
Context: how to pronounce "brachylog".
 
Anonymous
I understood the context, but not until I finished reading it for the first time :P
 
@Mego ( ͡° ͜ʖ ͡°)
 
 
3 hours later…
11:23 AM
0
A: Sandbox for Proposed Challenges

CharlieRegrowing trees code-golf tree-traversal Background Just another work-inspired challenge. Imagine you have a service that returns the list of nodes in a tree structure (much like the files and folders in a file system), but in no particular order. For every node you get a tuple with its name a...

 
 
2 hours later…
1:11 PM
CMC : the string {"lengthMenu":[[10,25,50,-1],[10,25,50,"All"]]}
 
@Adám SOGLs compression, 35 bytes
 
@Adám it would be 2 bytes shorter in Python if string were written using double-quotes using __repr__ but unfortunately I need to quote it because otherwise it won't be right :(
and also :[space] and ,[space]
I was typing my history essay yesterday and I finally get now why people like mechanical keyboards so much :P
 
They are amazing!
I have a blue mech keyboard and if my siblings use it they have to wash their hands. No grease on my keyboard!
 
Anonymous
@Adám Two Actually solutions for 50 bytes each: "{'lengthMenu':[[10,25,50,-1],[10,25,50,'All']]}"± (trivial) and ε' [-1,"All"]⌠[10,25,50]q⌡M$"{'lengthMenu':%s}"%±Æ (slightly less trivial)
 
@HyperNeutrino can you hop on discord?
 
Anonymous
1:26 PM
@Christopher2EZ4RTZ I've met some people that are more protective of their mechanical keyboards than their children :P
4
 
Haha took me forever to star that b/c mobile lol
When when I poke kids they complain but the mechanical keyboard just clicked all nicely
 
@HyperNeutrino Yup. Model M FTW!
 
@Christopher2EZ4RTZ No sorry I'm on school wifi right now and discord is blocked
 
CMC: Shortest string which has Kolmogorov complexity > its length in all programming languages (that we know of, so this calls for teamwork!)
 
1:34 PM
I don't actually have access to a mechanical keyboard for typing using my computer because the mechanical keyboard I have right next to me when I use my computer has the wrong type of plug and can only be plugged into the really old (like probably 10 year old) computer that still runs Windows Vista
 
@HyperNeutrino PS/2 connector?
 
what is that :p
 
How is OS related?
 
it's not
 
1:35 PM
@Emigna 35 bytes
 
it's just to give an idea of how old it is
windows vista
@Adám sorry can't see images
@Adám "aaaaaaaaaaa"?
or something like that
?
 
@HyperNeutrino No, in PHP that would just be that, no?
 
maybe. the string might have to be longer to make it golfable in languages
 
good luck with unary :p
 
Anonymous
@Adám That's only possible if we exclude cat-like languages
 
1:38 PM
@Mego They are not really programming languages, are they?
 
Anonymous
@HyperNeutrino Converters exist for PS2 -> USB
 
PS2 is a weird circle thing?
 
@HyperNeutrino Yeah, something like (::÷)
 
hm ok I think that's the one my keyboard has. I have a USB -> PS2 converter for whatever reason which is rather annoying :P
 
Anonymous
@Adám By our usual definition? No. However, some people think that definition should be ignored.
 
1:39 PM
@Mego So, what's the issue with my CMC?
@dzaima Why? Unary certainly requires > length for most strings, no?
 
@Adám oh, got the equality the wrong way around :p
 
I'm thinking something like \'"< against JavaScript, APL, many langauages, PHP.
Hm, actually, most languages require at least two bytes more than the payload. We only need to "defeat" those that do not.
 
there has to be a unicode non-ASCII character that is in Charcoals codepage, otherwise Charcoal gets equal score
 
@dzaima OK, that's a beginning. And we also need a < to confuse PHP, right?
 
Anonymous
@Adám Such a string doesn't exist if you use the current site rules, rather than the (IMO sane) rule of only allowing programming languages. If you restrict it further to TC languages, it would be very interesting :)
 
1:46 PM
wait so I think I finally understand the inequality, so essentially if the string has length n then it has to take strictly more than n bytes to output it in any language?
 
Anonymous
@HyperNeutrino Exactly
 
oh ok
 
@Mego I do restrict it to proper programming languages.
 
Also, I could make 256 languages and make the CMC impossible :p
 
so then aaaaaaa doesn't work
i was thinking that it had to be golfable in all languages :P
so essentially we're making the ultimate evil challenge
 
1:48 PM
@dzaima Unicode. You'd need a few more, but existing-languages' loophole.
 
So I am looking for one more speaker for the bi-weekly ppcg podcast. Anyone interested?
 
what's so special with unicode? do you count the target string in characters but program in bytes?
 
@Christopher2EZ4RTZ I'd be interested; depends on when it is though because I can only VC at certain times.
 
So far it is on Thursdays from 9 to 10 pm EST (not intentionally ripping off Mego but that is the one time ik I can be on lol)
And not on Thursdays that Mego is podcasting
 
I could probably join for one
 
1:54 PM
Sweet
 
this is a different podcast from Mego's monthly one right
 
ok
I think currently I can be on on Mondays from 7 to 9 and Fridays from 3 to 5
those are the only times I know for sure
 
hello the 19th byte
 
1:56 PM
hello unihedron
 
@Unihedron hello
 
I might be able to do other times
 
Anonymous
@Unihedron Hello new friend!
 
if i can get my parents' permission
 
I'm trying to compose a Quine in Seed, it's been hours and I've explored 1304000000 seeds to no avail
 
1:56 PM
Yeah even if you can only drop it on the time that's fine
 
any hints? how do you write your Seed programs - should I attack the mersenne twister?
 
@Christopher2EZ4RTZ what do you mean by drop it?
 
That is the only way you can
 
drop in?
 
Drop in lol
 
1:57 PM
For Seed you might want to ask Feersum :P
 
... I've tried reading mersenne twister, and...
 
Are there any examples? I felt a strong flame burning in me the moment I saw feersum's Seed codes and I really want to make my own
 
only the initializer (seed) is non-linear, the other parts are all linear.
Example for what? There are only a few written Seed program.
 
@HeebyJeebyMan hop on discord?
 
a way to crack the mersenne twister, I meant
 
Anonymous
1:58 PM
@Unihedron Attacking MT19937 (the specific implementation of Mersenne twister that is used) would be much more effective.
 
Anonymous
Difficult, but effective
 
@Mego Ah, I found a github repository that had an implementation for mt19937, but I was not aware that's what the code meant
 
Last time I tried to extract the relevant part to C++ code...
 
If you do that you have infinite rep
 
(actually C code) from CPython.
 
1:59 PM
@Christopher2EZ4RTZ I care about honor, not rep, fellow golfer! It's for the challenge!
 
Well that makes one of us
 
https://tio.run/##tZntb1tHesU/R3/FYBdIRFuS533mRpYKb3fRLrZWDCduC7iKQUlXNi2KFEgqtrLrvz39nbmURDlp0nxYBqHIy5l53s45zzPJ6dXV7tvT059@@uNkdjq9PuvN08l8uVr048vDrftnP/Snq/li88npeLnsF6vDra0nT56Yo/mqN6t34xVvk6U5n0x7w19Wzy@vpj0/bp/N@6WZzVfmdD5bjSezJUt7s7y5PJlPJ6fmdMqBI057ZF70i8n8zFyNF@NLti6WZnfXfPeuX/ZmvODf6dRcjt9OTveM@fN89hUnvhvP3vZ75tGTrT@e9eeTWW@OTPbx7ttzE7py/@3Zdy//@t9vnhn7setsPbFn568ML0zj23I1nq3MELAZb5756sWLv7x88/zZt39jZ7XD61Xbd0nOzHLydjY5n5xq/4fdhTmZrJab@//jm/@631/Oh9ewf9qPPzvgfvsW5bg@XZmX49nZ/PKbk/e4Zv6@JY8nLJzMzvqP@@3rNd@Df8NBq/Gqf310vL/1ab@ldDKbrCbj6eRHanC54hfzYbJ6R3jLvj@TkR/mk7O26s3bfrbA0vYDc4@
 
@user202729 Thank. I will research this.
 
I refuse to use jelly and other golfing languages to protest the unfair voting culture
 
I refuse to use jelly and other golfing languages because I value my time.
my computer right now:
 
2:01 PM
(says while attempting seed quine :P /s)
 
Program length is 8 characters.
95^8=7213895789838336 possibilities.
Estimated time: 834941 days, 11 hours
Brute-forcing seed...
1311000000
1312000000
1313000000
1314000000
1315000000
1316000000
1317000000
1318000000
1319000000
1320000000
1321000000
1322000000
 
Have you looked at what's the shortest Befunge quine?
 
Try pain-flak if you want a challenge!
 
Essentially, Seed quine = Befunge quine + MT19937 decoder
 
it would be massively different though, since we gotta print numbers (the print num operator also adds the space for us for free)
 
Anonymous
@Unihedron So cracking MT19937 isn't actually all that hard. If you have 624 consecutive 32-bit values generated by the PRNG, you can determine all previous and future values. By also knowing the number of random values produced prior to those 624, you can determine the initial seed.
 
(Tokenized Jelly, anyone?)
 
@user202729 well, kind of, except it can't be a Befunge quine because it needs to print the seed program itself, not the twisted Befunge program itself
 
@Mego Yes, that part (determine original state) is easy, but revert it to the seed is hard.
 
2:04 PM
yeah
 
@Mego This is valuable information and it is consistent with my ~10 minutes of googling, I found a github repo that implements this but I was unsure of / stuck on how to get said set of random values, or how to scale my program into a list of random values into the right places
 
@Unihedron Forgot to say, the input format: first number = number of bytes, other numbers: base-(2^32) digits of the seed. That corresponds to the Seed Hello World program.
 
@Christopher2EZ4RTZ sure
 
@user202729 is it not the case that first number could > actual number of bytes, as long as our seed program ends with ;?
 
Anonymous
To crack Seed, though, it's a bit more complicated: you'd have to look at the actual CPython implementation of randrange (which randint uses, which is how Seed chooses the characters) and determine how to get the necessary information.
 
2:06 PM
... Yes. I don't know Befunge so I'm not sure. I assume ; is program termination.
 
with this we can cut on the amount of "rejected" quines significantly, since we can print a lot of more valid numbers for the first output in the befunge
I believe so.
 
Thanks for reminding me about Seed, BTW.
 
I mean this -> seed "X Y" -> program vvvvvv -> seed "X+5 Y" -> program vvvvvv????? <- ignored
 
@HyperNeutrino proxy?
 
@dzaima No, but even with 256 langs, they'd need a way to specify strings. Some Unicode char, I presume.
 
Anonymous
2:08 PM
So because random() doesn't use the entire state, you'll need more than 624 values to get enough information
 
@user202729 You're welcome. Seed looks awesome. I'm so fired up right now.
 
Anonymous
So the moral of the story is to use a long quine to hack Seed
 
I doubt it but I don't know. Also I don't know of any unblocked proxy sites anymore because the school board is blocking them more quickly than they're being created
 
Write one yourself.
 
lol yeah I might do that sometime
but then i'd need to run it on my computer at home or something like that
 
2:09 PM
The more important thing is not to write one, but to not get blocked.
 
Find a free server.
 
That isn't blocked.
 
@Unihedron For that... find a nice name.
(I have a free openshift server)
 
hm
maybe
 
@Mego ... inventing information is easy.
 
2:11 PM
github.com/BishopFox/untwister and github.com/GeorgeArgyros/mt_derand seem to be excellent proof of concepts on calculating the source based on the expected output
 
I didn't really Google things online, just read the source.
(for the record, I tried to use Mathematica symbolic processing ability to solve this, but just expanding a multiplication takes forever)
... we don't need future output.
 
@HeebyJeebyMan what server are you on? Can't find you
 
@Adám ok, 65536 langs, if first 2 bytes are "", output the rest of the code plus the language ID decoded to 2 bytes, otherwise interpret as brainfuck. :P
 
@Christopher2EZ4RTZ I was just on the podcast one
 
@dzaima BF is TC, that won't work.
 
Anonymous
2:14 PM
Also, another challenge to writing a quine in Seed: the PRNG is seeded with a single 64-bit value, so the seed is stretched to fill the initial state. This means that the initial states that can be obtained is a very small subset of all possible initial states. It's entirely possible that none of those initial states can produce satisfactory sequences of values.
 
@user202729 the CMC needs TC langs to be valid
 
@Mego ... No, the seed is Python big integer.
 
Anonymous
Oh wait you're seeding from Python, duh
 
Anonymous
So yeah, bigint
 
Anonymous
2:16 PM
So it's possible to achieve every initial state of the PRNG
 
With brute force (or symbolic manipulation) it may be possible to achieve Seed Hello World in 40 bytes.
 
Anonymous
Still, that just makes it theoretically possible to have a quine. Finding a quine will be very, very, very difficult.
 
which would also explain why no one has done it yet
but think about how spectacular it would be though, a language that runs on and generates code in another language that prints its own source code in the source language
soon: Seed^2
 
Anonymous
Because of the loss of information (32 bits to 8 bits) in character generation, you need a lot more information to determine the state, so the Befunge code will have to be very long.
 
Also, considering the fact that the seed will unlikely to have very few digits, the befunge code is going to be very long
to generate that exact number
the efficient methodology would be to run all the code we generate to compare, but I don't have the power to do that, so I'd rather employ tricks like quickfail for code (e.g. that doesn't print anything) and manually sort through candidates
 
Anonymous
2:23 PM
Trying to brute-force a quine would take a very, very, very long time
 
@Pavel two weeks is up for the call for categories
unless you wanna let it run for the rest of this week
 
3:02 PM
It just occured to me that quine in seed is as dynamic / difficult, of not more, than narcissus
 
3:46 PM
@Poke CC @Dennis @MartinEnder please make nominations post
@Mego iirc the shortest hexagony quine was discovered by brute force
 
@Pavel But that one is length 6.
and there are 3 fixed characters
(; for print, @ or / for terminate, ? for read, if I recalled correctly)
So there are only (at most) 95 ^ 3 * 6! candidates
which is just 500 millions.
(the actual number is lower)
 
@Pavel I won't be able to until tonight. Could you do it youself?
 
@user202729 that exists?
 
I remembered the wrong program.
The length 6 is the truth machine. (which was brute forced)
And... no, the hexagony quine is VERY long and full of numbers. Can't be brute forced.
 
@Pavel if you know a shorter Hexagony quine than this one I'm all ears :D
 
3:56 PM
If that was found by brute force, I'd very much like to buy the computer.
 
59
Q: What would happen if an alien race gave us a computer capable of 2^256 calculations per second?

SnowmanThe security of public key cryptography relies on computers not being able to generate anywhere near 2256 guesses per any reasonable time length. The obvious implications of a computer this powerful would be that Bitcoin and all other cryptocurrencies would be hacked immediately. But what other l...

 
I hope they also give us an equally powerful cooling device.
 
00:00 - 16:0016:00 - 23:00

« first day (2546 days earlier)      last day (2300 days later) »