« first day (1968 days earlier)      last day (1270 days later) » 

12:01 AM
@DrGreenEggsandIronMan Second Tuesday of the month is, actually. The time is 2-something in the morning, I think.
 
It's 3:32 AM, and it's the third Tuesday of the month.
Dang it, ninja'd.
Haha, is it inappropriate to star that?
 
12:22 AM
It's time to invoke the hive mind with JS.
 
...welp.
 
12:41 AM
Ȩ̶̡͡x̵̡͡ç͘͠e̴l͏̵̷l͜͝҉͏é̛͜ǹ̨͘t͝҉҉̨̀
 
1:12 AM
^^
just why
@AlexA. are you going to the naked bike parade?
( ͡° ͜ʖ ͡°)
 
@EᴀsᴛᴇʀʟʏIʀᴋ That reminds me of this:
 
1:34 AM
New challenge in one minute!
0
Q: All Armstrong numbers

Leaky NunAn Armstrong number (AKA Plus Perfect number, or narcissistic number) is a number which is equal to its sum of n-th power of the digits, where n is the number of digits of the number. For example, 153 has 3 digits, and 153 = 1^3 + 5^3 + 3^3, so 153 is an Armstrong number. For example, 8208 has ...

 
@LeakyNun Why can't the separator be a line break?
 
@DrGreenEggsandIronMan clarified.
 
Oh, I get it. May (not be) rather than (may not) be.
 
@El'endiaStarman hahaha
@DrGreenEggsandIronMan yeah, seattle's parade is tomorrow iirc
 
1
Q: All Armstrong numbers

Leaky NunAn Armstrong number (AKA Plus Perfect number, or narcissistic number) is a number which is equal to its sum of n-th power of the digits, where n is the number of digits of the number. For example, 153 has 3 digits, and 153 = 1^3 + 5^3 + 3^3, so 153 is an Armstrong number. For example, 8208 has ...

 
1:43 AM
@DrGreenEggsandIronMan Yeah, I missed that particular interpretation as well.
 
@DrGreenEggsandIronMan You know, overloaded English expressions.
 
@LeakyNun It's scary how quickly you gained a ton of rep. You're going to pass me pretty soon here.
 
@DrGreenEggsandIronMan By having no life.
jk, programming is love, programming is life.
 
You mean to imply anyone here has one?
 
2:10 AM
@LeakyNun Oh hey, thanks!
Grr. I wanna do it in C, but the last number would need 128-bit ints.
 
@DrGreenEggsandIronMan >_>
@LeakyNun fuck no
dont bring that shit here
 
@EᴀsᴛᴇʀʟʏIʀᴋ calm down
 
I now have pet dogs \o/ (in mc that is)
 
@LeakyNun no
@HelkaHomba yay!
 
@EᴀsᴛᴇʀʟʏIʀᴋ don't calm down then.
 
2:24 AM
@HelkaHomba are you doing the video now or already did?
 
@EᴀsᴛᴇʀʟʏIʀᴋ Did it and should be online tomorrow
 
@HelkaHomba -1 for not having pet goat
goat pet is best pet
@Quill Is your computer still stable or very buggy?
 
@DrGreenEggsandIronMan I was gonna add "then if you buy a lottery every year, then every time you win the lottery, you pick 5 card; then when you have a full house, you haven't passed a hundred millionth of the time"
 
@Quill ;-; node y u do dis
 
@Upgoat it's actually pretty buggy on macOS Sierra
but this is an early beta
 
@LeakyNun The new rules are kinda dissapointing because it's now purely a compression challenge, and those particular number have no real pattern to them.
 
2:48 AM
@DrGreenEggsandIronMan I've raised it to an hour
 
Calculating them is out (they're far too large)
 
@DrGreenEggsandIronMan If it is really the case, how were they calculated in the first place?
 
I also don't see any patterns to exploit other than numbers ending in 0 can also end in 1
 
@Quill yeah :/ idk if i should update at all because theres nothing cool except siri but who talks to their computer
 
Because they didn't have a time limit when they first calculated them.
 
2:49 AM
@DrGreenEggsandIronMan so they basically used 10^21 years to calculate them?
Don't loop through the numbers, loop through the combinations
 
I'm sure there's plenty of techniques you can use to optimize the calculation.
 
anyone know how to drop all commits before a specific commit (e.g. delete all commits before HEAD~12)
 
The largest number there is 39 digits long
 
Sure, but any code that goes to the trouble could certainly be beaten bubblegum.
 
2:52 AM
and 7*9^39 is already 39 digits long
so there can be at most seven 9s for 39-digit
don't check if the number is equal to the sum of power digit
check if the sum of power digit is an ordering of the number
 
@LeakyNun user for 5 months and 6,169 rep? Dang you are earning rep fast ._.
 
@Upgoat coz i ain't got no life
jk, programming is love, programming is life
 
1 hour ago, by Leaky Nun
jk, programming is love, programming is life.
is that your motto? :P
 
1 hour ago, by Dr Green Eggs and Iron Man
@LeakyNun It's scary how quickly you gained a ton of rep. You're going to pass me pretty soon here.
does everyone like to ask the same question?
(ok, it wasn't a question)
 
^^^
@CᴏɴᴏʀO'Bʀɪᴇɴ halp why does your usernacript add #NaN to the ahedr
 
3:29 AM
majik
 
@EᴀsᴛᴇʀʟʏIʀᴋ btw got to 70% coverage on Cheddar :D :D :D :D
 
3:52 AM
new challenge in 10 minutes
 
@Quill anyway to make mocha not die a fiery death on a single error?
 
I added the error code
 
wat that mean
halp, how fix shitty commit history. I tri git rebase -i HEAD~10 but eror
 
2 nights in a row methinks. this dragon be trying to sleep
 
4:09 AM
@CᴏɴᴏʀO'Bʀɪᴇɴ >_> sorry
 
oh 2 javascripters in the chat rn
here we go
idk if I'm missing any operators
 
@MamaFunRoll ಠ_ಠ zalgo abuse eleven
 
hey it's totally worth it
 
@Quill are you be good at git
 
0
Q: Hexdump generator

Leaky NunSome answers to our challenges include unprintable characters, i.e. characters in the range of U+0000 to U+001F and U+0080 to U+00FF. PPCG filters out those characters, except U+0009, U+000A, and U+000D. Therefore, we include the hexdump in our answers in those situations. Your task is to gene...

 
4:16 AM
@Upgoat for some values of good... what did you do?
 
@Quill its not what I did. It's what I did in the past. So I have really borked commit history. How do I delete all commits before commit ea57570e9ded91a86d0d8d50a95cb1d46a507bf5. pls halp i tri fo three dai but no werk
 
posted on June 19, 2016 by Leaky Nun

Some answers to our challenges include unprintable characters, i.e. characters in the range of U+0000 to U+001F and U+0080 to U+00FF. PPCG filters out those characters, except U+0009, U+000A, and U+000D. Therefore, we include the hexdump in our answers in those situations. Your task is to generate the hexdump, with demonstration below: 0000000: 2e 22 20 79 10

 
@NewMainPosts y u post deleted question?
 
you spelt knee wrong
 
newmainpost ninja'd
 
4:18 AM
@Upgoat checkout the current commit into a new branch, delete master and then make master again from that new branch
 
@LeakyNun You shouldn't of deleted it and rather just of marked it as duplicate. Duplicate posts, serve as markers for the actual post. Making it harder for users to make the dupe again
 
@Upgoat no avocad juic?
 
@Quill can u pls come into chedr c9 and do magik. i tri what i think you said but it didnt work
@MamaFunRoll ಠ_ಠ avocad meme is old
 
@Upgoat Done
 
@Upgoat wait WAT RLY
finally
 
4:21 AM
@MamaFunRoll mods really went into full fun-police mode and killed memes with this post. so no more memes :'(
 
s/no more memes/no more overused old memes/
 
no it's no more memes at all
If I want to ping mods should I ping all the mods or just one?
 
@Upgoat aww
rip avocad
 
@Upgoat depends what the ping is about
 
if the lock on memes post will ever come off
 
4:25 AM
quick open gl question: with glClearColor, what are the values? I mean I know that they are r, g, b, a, but normally those would range from 255, 255, 255, and 1
 
57
A: Delete Github commit history

matthewdanielJust pick the hash you want to go back to and in your clone do: git reset --hard hash# git push -f origin branch where branch is the name of the branch you want to push. Voilà. Be carefully with the force push. You may want to copy your working directory until you are familiar with it.

 
that question is talking the 10 most recent commits
 
like what is the name for that color system?
 
@Upgoat ping one of them, preferably whoever is online or who put the lock on the post
 
@AshwinGupta I'm not sure for gl but sometimes a can range from 0-255 too (or rgb can be from 0-1)
 
4:27 AM
oh
so wait
 
@Dennis Is the lock on Many Memes permanent? Or just until things with memes settle down?
 
so for rgb 0, 241, 137, I could do 0.0 ,0.95, 0.54?
 
I think I was out of it all during the week this meme crackdown happened, which is why I feel deprived of my dank PPCG memes :P
 
@MamaFunRoll what dank PPCG memes? :P
 
@AshwinGupta Yeah. I'm just saying that I've seen it both ways, though I'm not certain for your particular case.
 
4:30 AM
yeah just tried it
I think it worked
=/ looks about the same
 
@AshwinGupta This page says that all values are clamped to [0, 1], so yes, you have to divide by 255.
 
@Upgoat sometimes, I just have the urge to eat avocad and shooting ಠ_ಠ to chat stars y'know
 
O_o okay.....
 
pretty tied after spending the past 18 hours troubleshooting these computer issues! gonna relax with a nice bath 😊 https://t.co/oNVrI7hjdW
 
4:47 AM
@MamaFunRoll I am somewhat mad at you now.
 
@El'endiaStarman Now you know how I feel.
 
Reminds me of Vi Hart's video on lists.
 
I got to the end, and thought the implication was that he'd eaten them all or something
 
@Upgoat It's unlikely that we'll unlock the post in the future. We just don't see anything good coming from it.
 
yeah, it doesn't really get updated anywhere else
 
4:54 AM
holy sh*t
this low level crap is making my brain hurt
like a LOT
 
@Babylonian as an android user this just made me come
wtf
 
5:10 AM
Here's a doggo realizing you can stand in a pool. 13/10 enlightened af (vid by Tina Conrad) https://t.co/7wE9LTEXC4
10/10
 
Anyone know what IDE this guy is using?:
 
@Justin in the first picture? visual studio it appears
 
I can tell he's using Windows 7, has Spotify open, uses Google Chrome
He has the command prompt open at least
I can't see the visual studio logo in the taskbar
What's the logo that looks like a cube with a sub-cube pulled out of it?
 
@Justin idk intellij?
 
No that's not the intellij logo
I thought maybe virtualbox, but it's not either
 
5:21 AM
@Justin VS Code
 
@Quill Nice
 
the green one looks like Line
 
Oh wondered what that was (on the far right)
I know I've seen that blocky logo before
 
@LeakyNun how long does your armstrong take to complete?
 
5:36 AM
@xnor still timing
 
@LeakyNun i'm really doubtful it will finish in an hour
i estimate it takes about a week
 
@Quill 10/10 hardest to goolge name
What functions should exist on the String namespace?
 
@Justin Maybe it's regedit?
 
Hmm I don't think it'll take a week, but half a day seems more likely than an hour
(just based on a 10^5 ~ 4 sec estimate)
Hmm I'm off by a zero there aren't I... yeah maybe a week
 
@xnor i estimated 7.8 days, so I'll delete it
 
5:48 AM
it looks like an optimization is needed beyond testing counts of digits
 
Well, it is plausible that they generated the list using 7.8 days
but sure
well, the longest number is only 39 digits, so using base 10 for count of digit (which allows 100 digits) seems to be a bit un-optimized
Maybe we need to consult @Optimizer for this task.
 
Hello
 
the number of partitions of 39 into 10 parts is 2e9
 
@zʏᴀʙiɴ101 Hello, almost done with cheddar functions >_>
just gotta do optimizations
 
@Upgoat But not finished?
 
5:52 AM
almost >_>
 
Okay, ping me when you're fully done with functions.
 
@xnor one zero less!
@xnor Also, any idea what the upper bound would be, if we didn't know that the largest number is 39 digits long?
 
maybe it's worth it to hardcode the partitions and then generate the numbers from them?
 
I'm having a difficult time constructing an upper limit based on x*9^x - 10^x because gives me 34 instead of >39
 
@LeakyNun is there a known a priori digit bound?
 
5:55 AM
@xnor nope
Actually, x*9^x - 10^(x-1) gave me the correct upper limit: 61
 
@LeakyNun 61 digits gives 6.5e10
 
;-;
 
@xnor how did you calculate that?
so, 6.5e10 should take less than a month
which is completely plausible
 
the number of partitions of n into k ordered parts is choose(n+k-1,k-1)
 
oh, nice
 
5:59 AM
Is this positive parts or nonnegative parts?
 
non-negative
 
65033528560
definitely << 10**39
where << is waaaaaay smaller
@xnor any further optimization?
 
maybe there is short-circuiting you can do, such as not having so many 9's that it exceeds the digit count
or so few that the rest of the digits <=8 can't give you the right number of digits
 
@xnor for example?
 
@Upgoat Quill and an user called "went offline"?
 
6:02 AM
ಠ_ಠ definetly
 
@zʏᴀʙiɴ101 wat
 
@xnor I thought 61 already accounted for that
 
(Oh, stars and bars. Took me too long to realise why that's true...)
 
@LeakyNun for example, with 39 digits, having 8's or less only takes you up to 37 digits, so there must be at least one 9
 
... but what about partitions into ordered parts with maximal part 7?
 
6:04 AM
@Sp3000 is that from knowing the answer, or is there a reason for it?
 
Are we trying to do this from knowing the answer or not from knowing the answer? (can't tell atm)
 
It's incorrect anyway (there is a 9 somewhere, go search for it)
 
@Sp3000 either way, just curious
 
@LeakyNun [9 4 4 4 3 3 2 2 2 1] oh, missed it amongst the noise
 
@Sp3000 nice
 
6:05 AM
Hmm maybe we can at least bound, say, the number of 9s?
 
@Sp3000 yep, that's what we're doing
the upper bound is 61
Actually, 60
 
No I mean, for each number of digits n, there's only a small range the number of 9s can be, say
 
which gives us 56672074888 (5.6e10) instead
@Sp3000 well, we're checking every number of digit anyway
but I'll compute that
 
I'm just trying to see how much xnor's short circuiting comment actually helps :/
 
6:11 AM
you need at least 6 copies of 9
 
For 1 digits, all being 8 gives 1 digits, and 9^1 has 1 digits.
For 2 digits, all being 8 gives 3 digits, and 9^2 has 2 digits.
For 3 digits, all being 8 gives 4 digits, and 9^3 has 3 digits.
For 4 digits, all being 8 gives 5 digits, and 9^4 has 4 digits.
For 5 digits, all being 8 gives 6 digits, and 9^5 has 5 digits.
For 6 digits, all being 8 gives 7 digits, and 9^6 has 6 digits.
For 7 digits, all being 8 gives 8 digits, and 9^7 has 7 digits.
For 8 digits, all being 8 gives 9 digits, and 9^8 has 8 digits.
@xnor Not sure what you would do from this data
 
with exactly 6 copies of 9, that leaves ~2e7 choices for the other digit counts, which is more tractable
 
@xnor how do you know?
@xnor "this data" referring to the thing I just posted above
 
>>> k=5;float(9**39 * k + 8**39 * (39-k) )
8.776523532338918e+37
with only 5 9's, you only get 38 digits
 
Actually, the upper bound is 60 instead of 61
@xnor
For 1 digits, you would need 0 copies of 9, making the search space 1.
For 2 digits, you would need 0 copies of 9, making the search space 10.
For 3 digits, you would need 0 copies of 9, making the search space 55.
For 4 digits, you would need 0 copies of 9, making the search space 220.
For 5 digits, you would need 0 copies of 9, making the search space 715.
For 6 digits, you would need 0 copies of 9, making the search space 2002.
For 7 digits, you would need 0 copies of 9, making the search space 5005.
> 1 copies
 
6:19 AM
interesting, it looks like the search space gets narrower as you get more and more digits
the number of non-9 digits never goes above 24
 
@xnor added
Edit:
For 1 digits, you would need 0 copies of 9, making the search space 9.
For 2 digits, you would need 0 copies of 9, making the search space 45.
For 3 digits, you would need 0 copies of 9, making the search space 165.
For 4 digits, you would need 0 copies of 9, making the search space 495.
For 5 digits, you would need 0 copies of 9, making the search space 1287.
For 6 digits, you would need 0 copies of 9, making the search space 3003.
For 7 digits, you would need 0 copies of 9, making the search space 6435.
Total search space: 1,760,831,806 (1.7e9)
In this case, the real upper bound is 39 digits, making the search space 721,266,811 (7.2e8)
@xnor
 
These "you would need x copies of 9" might make an interesting sequence.
 
@xnor can you help me check?
 
But what was the problem first?
 
@zʏᴀʙiɴ101 my challenge
 
6:32 AM
Link? :3
 
3
Q: All Armstrong numbers

Leaky NunAn Armstrong number (AKA Plus Perfect number, or narcissistic number) is a number which is equal to its sum of n-th power of the digits, where n is the number of digits of the number. For example, 153 has 3 digits, and 153 = 1^3 + 5^3 + 3^3, so 153 is an Armstrong number. For example, 8208 has ...

 
6:48 AM
Ooh, Fith can do this one! It's gonna be long..
 
@bkul Fith?
 
@zʏᴀʙiɴ101 My new pet language. Repo if you're interested
 
7:30 AM
Question which I can't find the answer to on Meta: when writing code in C and friends, do we take integer overflow into account? I would think we would, but using GMP is so.. verbose.
 
7:53 AM
I think in general it seems to be no, but don't abuse it
(similarly, some Python 2 answers have a problem for large inputs because of the extra L at the end of longs)
 
8:11 AM
There's not a great consensus on it (at all), but do you mean something like this?
 
@Bigtoes Yes, thanks.
 
What else is Big? @Bigtoes ?
 
Your neck
 
I know your lies are big.. now your toes.
@Bigtoes its long. not big.
 
Long is a type of big o_O
Also, it's roughly as wide as your head, so big that way too :P
 
8:15 AM
better than having just big toes.
 
I never said just.
>_>
 
yeah ? :P
NotJustBigToes
 
The T doesn't need to be capitalized, but other than that it's fine.
 
You see, that's why I didn't want to clarify the time limit
A better time limit (and thus an actual, strict time limit) would help this challenge a lot. Raising the limit from an hour could allow some non-hardcoding submissions, which IMO would make this a much more interesting challenge. As it is, with the "soft" time limit, I'm tempted to VTC as unclear or primarily opinion-based. — Mego 5 mins ago
 
if you don't clarify it we can post answers that take thousands of years
 
8:21 AM
sure
what would be a good time limit?
 
there is no good time limit
 
what?
 
if it's too short all answers are harcoding, if it's too long all answers are computing it, and if it's in between all answers are a mix of the two
 
then what should I write
 
and imo hardcoding is usually uninteresting
 
8:23 AM
@Fatalize if it's too long all answers are brute-forcing it
 
keep 1h I'd say, to not screw Dennis over for example
 
I just raised it to a day
 
it's too late anyway
 
we just discussed how to optimize it to 721,266,811 (7.2e8) searches
so a day is good
 
 
1 hour later…
9:26 AM
\
 
9:37 AM
@Optimizer ?
 
@zʏᴀʙiɴ101 Optimizer is trying to escape :P
 
@trichoplax escape :P or was it a joke to escape?
 
10:31 AM
@zʏᴀʙiɴ101 It was just a very bad joke... :)
 
11:09 AM
someone spray painted the mute sign over donald trump's star 👏🏽👏🏽👏🏽 https://t.co/JY4bLADZcy
 
^ use this for the cracked soil challenge
 
Blue origin webcast today: blueorigin.com
@Quill why the heck is this guy even on the walk??
 
cmc: ackermann
 
11:35 AM
This is so hillarious: vimeo.com/126177413
Another one from the same guy
 
12:28 PM
in Sandbox, 22 mins ago, by qwertyuiop
                                                                            r9hH####G.
                                                                           i#@@@@@@@@M
 #@:                                                                      s@@@@###@@@@h
 M@@@s                                                                  :@@@@#MMBB#@@@@
  @@@@@&                                                              .A@@@MHBBBMMM#@@@;
   @@@@@@@r                                                          r@@@#MBHAAMMHHB#@@s
(click in Sandbox, 22 mins ago or visit the Sandbox to see the cat in full.)
@LeakyNun ?
 
Just testing why it can't be expanded here
 
12:56 PM
Good morning TNB!
 
Good morning part of TNB!
 
@LeakyNun I think it would be better if you didn't say "a reasonable amount of time (say, less than a day)". You should just say "it must finish within a day." Some people are claiming "he said 'say a day', meaning the time limit is just an example."
 
@DrGreenEggsandIronMan people say everything to themselves
 
Still, it could help remove ambiguity.
 
Primes are complicated...
 
1:07 PM
@LegionMammal978 O_O @_@
 
A more readable(?) version:
 
(stupid question here)
is it faster to reverse an array by in-place reversion or by recursive divide-and-conquer reversion?
actually, it's a bitwise reversion for a number
so should I build lookup tables for 4-bit numbers and then use divide-and-conquer (kind of combine the two algorithms)
def highestbit(n):
	shift = 1
	while (n&-~n) > 0:
		n |= n>>shift
		shift <<= 1
	return -~n
def bit_reverse(n):
	mark = highestbit(n)
	lo = 1, hi = mark
	while lo<hi:
		if ((lo&n)>0) != ((hi&n)>0):
			n ^= lo|hi
		lo <<= 1
		hi >>= 1
 
1:25 PM
What happens inside an Inform 6 "Hello, World!" script:
     4 classes (maximum  64)             4 objects (maximum 639)
     0 global vars (maximum 233)       480 variable/array space (maximum 10000)
     0 verbs (maximum 200)               0 dictionary entries (maximum 2000)
     0 grammar lines (version 1)         0 grammar tokens (unlimited)
     0 actions (maximum 200)             0 attributes (maximum 48)
     2 common props (maximum 62)         8 individual props (unlimited)
   118 characters used in text         104 bytes compressed (rate 0.881)
And a story file size of exactly 2.5 KiB.
For those interested, the script I meant is:
[ Main;
  print "Hello, World!^";
];
 

« first day (1968 days earlier)      last day (1270 days later) »