« first day (2511 days earlier)      last day (708 days later) » 
00:00 - 20:0020:00 - 00:00

8:00 PM
Amen, Jenny_mathy has a dope hat
 
Confirmed. Jenny_mathy's hat is secret
@Mr.Xcoder Amen to that :P
Wait, Howard has a hat?
 
8:17 PM
Hm I haven't gotten any hats :(
 
being active helps, that's why the hats exist
winter is generally the time when the site has the least activity, so hats are here to help it a bit
 
I got two secret hats the other day, and now they're gone.
 
what
that can't happen, refresh perhaps?
 
@KSmarts stealth hats
 
Schrödinger cats hats.
5
Stupid strikethrough makes it look like Schrödinger eats hats.
 
8:30 PM
Schrödinger hat einen Hut
 
@cairdcoinheringaahing I donmt think Jenny Mathy’s hat is secret. It seems to be in the profile picture itself, so I don’t think it is a hat at all (I can see it from mobile)
 
@Mr.Xcoder Yeah, realised that when their profile didn't have '1 hat' in it :/
 
it is a dope hat, not a winterbash hat
 
WTH is a dope hat?
 
A hat that is dope. Keep up. ;-)
 
8:32 PM
@cairdcoinheringaahing (Posts Sandbox link) ...
(30 minutes later... nothing)
@Mr.Xcoder Alternatively, it is a heroin hat.
 
Well, depends on the definition of dope
 
Dope means weed where I'm from :P
 
Dope can mean Marijuana, Heroin or nice
 
dope means nice
much like how gay used to mean jolly
thanks for ruining english~
 
8:40 PM
at least the word english is not used as negatively as french
 
As in "Pardon my French?"
 
haha:)
So what do you think about making a tag for ? I feel we've had a quite a bunch of challenges where it would fit.
 
how would you define the tag?
 
just to clear up possible misconception: coding theory != coding (= programming)
Coding theory is about communication, encoding, compressing etc
 
@flawr Could you give an example of where you think it would fit?
 
8:47 PM
e.g. anything e.g. about the hamming code or gray code, let me gather some links
 
@Unihedron Can I ask where your profile pic is from?
 
@DJMcMayhem challenges that would fit ^
 
dope also means (noun) fool or (verb) enhance (probably why a drug is called dope)
 
0
A: Sandbox for Proposed Challenges

SteadyboxMirrored Digital Clock Many digital clocks display the time using simplified digits comprised of only seven different lights that are either on or off: When mirrored horizontally, the digits 018 don't change, because they are symmetrical. Also, the digits 2 and 5 change places, 2 becoming 5 a...

 
9:01 PM
@DJMcMayhem any conclusion? :)
 
@flawr I like the idea. I haven't looked that closely at the challenges you linked but I think it works. Only thought is that their might be a better name
(information theory? Encoding-theory?) Idk
 
encoding theory is something else
information theory is more general, but includes coding theory
 
2
Q: Binary Sequences

J843136028Given a binary number A as input with d > 1 digits, output a binary number B with d digits according to the following rules for finding the nth digit of B: The first digit of B is zero if the first and second digits of A are equal; otherwise, it is one. If 1 < n < d, then if the (n-1)th, nth an...

 
do you mean because people might misunderstand what it is about?
 
Yeah.
Like you said "Coding" can have multiple meanings
@flawr What is encoding theory?
 
9:08 PM
@DJMcMayhem it is also about communication, but in sociology
 
We have , which is admittedly pretty broad but seems to fit well here. Maybe information-theory fits better.
 
(don't know a lot about it, just knew the name because I know someone working in that field:)
 
@AdmBorkBork Number theory is waaay broader. Coding theory is about the most efficient/reliable ways to transit data, whereas number theory is stuff like distribution of primes
 
@AdmBorkBork I think number-theory does not include coding theory,
and I think information-theory might be just too broad
then again we also have ...
but I mean just adding another very broad tag is not very beneficial
 
@flawr this?
@flawr Coding theory has my 👍. If you want more opinions, you could always take it to meta.
 
9:11 PM
Yes I think this is what I had in mind, but again, I'm not really familiar with that:)
@DJMcMayhem good:) I just wanted to see what people think before I start adding meta-clutter
 
Right, not saying that number-theory includes coding-theory, just that I wonder if coding-theory is too narrow, and so information-theory may be a better tag in the same reason that we have number-theory.
 
I see, well it would really have the advantage of a better name, or at least one that would probably be less misunderstood
 
That's also true
 
@AdmBorkBork Can you think of any challenges that would fit that don't fit ?
 
btw that color profile bug in chrome 63 is annoying...
 
9:19 PM
@DJMcMayhem Actually it seems many courses at universities bear the title "information- and code theory" (or something along the lines thereof)
 
BTW, speaking of information/coding theory, I remember a really interesting problem I ran into like 3-4 years ago
If you have a huffman encoding, (mapping arbitrary bytes to a sequence of bits of variable length), what is the most compressed possible way to write down the key?
For example, what's the shortest binary way to represent the following:
a: 0
b: 10
c: 11
 
@DJMcMayhem Sure, one of mine -- Patch the Paragraph as it involves natural language processing and pattern recognition but without any form of coding.
 
@DJMcMayhem that depends on what your representation could represent
 
What do you mean?
 
for example if it can only represent that table, then 0 bits is the shortest it could get
 
9:22 PM
@AdmBorkBork I'm not sure how relevant would be for this
 
@EriktheOutgolfer Well obviously. But I mean for any table in a format similar to this
Where you have an unknown number of {byte: bit-sequence} mappings
 
No, I agree, it's not a perfect fit, but it is something off the top of my head. Alternatively, every single KC challenge is an information-theory challenge, hehe.
 
But importantly, The bit-sequences form a prefix code. I'm not sure how that could be used to shorten it
 
well, for one, I'd say 11001011
11 = max 2 bits
0 = separator
0 = leading 0 == 00
10 = 10
11 = 11
but others may have a different opinion
that's what I mean
 
it also depends on the nature of the tables you would like to encode
i.e. should we optimize for the size of the table or the length of each element in the table?
 
9:26 PM
@EriktheOutgolfer I'm not sure that could be uniquely decoded. But more importantly, that doesn't map each sequence to it's corresponding byte, it just tells you what ever bit sequence is
@LeakyNun The size of the whole table
 
@DJMcMayhem oh you need the byte too?
 
Yes
 
@DJMcMayhem i.e. your tables would typically have a lot of elements?
 
Because you don't know what the set of bytes used will be
 
@DJMcMayhem then
 
9:27 PM
00 = separator
10 = 0
 
11001100001001100010100110001111 (110 = 2 bits max, 01100001 = a, 0 = leading 0 == 00, 01100010 = b, 10 = 10, 01100011 = c, 11 = 11)
 
11 = 1
10001110001111
I win
 
@LeakyNun and where are the bytes?
 
@EriktheOutgolfer what does that mean?
 
a, b and c are bytes in DJ's example, you should include them in too
 
9:30 PM
oh...
 
that's what I mean with this
8 mins ago, by Erik the Outgolfer
@DJMcMayhem that depends on what your representation could represent
 
I'm out
 
and also how it's understood
 
@EriktheOutgolfer Hmh, Interesting. But I don't understand how you know when the key is over. Obviously, you can tell if the mapping is the max length, but what if it's shorter?
So what's the significance of the leading 0 in your example?
If you're curious, here's what I ended up doing:
01100001(a) 10 01
01100010(b) 11 10 01
01100011(c) 11 11 00
So 10 == 0, 11 == 1
And 01 == end of key, 00 == end of map
But I'm sure that's not 100% optimal
 
10:25 PM
@DJMcMayhem @EriktheOutgolfer @AdmBorkBork on meta: codegolf.meta.stackexchange.com/questions/14400/…
 
Any feedback on this challenge idea? codegolf.meta.stackexchange.com/a/14169/56433
 
@Laikoni would you also allow nested lists, for instance [4,3,2,1] -> [[2,1],[4,3]] -> [[[1],[2]],[[3],[4]]] etc?
 
Or list-like structures in languages where lists are distinct from, say, arrays / tuples?
 
2
Q: Tag Proposal: Information- and Coding Theory

flawrAfter some discussion in the chat, I propose to create a new tag about Information- and Coding theory. We've already had a number of challenges that would fit this tag very well and I think there are many more to come. It would add context to those challenges that is not possible to express with ...

 
@NewMetaPosts Am I supposed to add each a Yes and a No answer such that people could vote?
 
10:34 PM
@flawr Yes, I'd say this is a reasonable representation, even though it's evil from a Haskell point of view.
 
@flawr I'd say no, don't post an answer you don't agree with
 
@DJMcMayhem so should I add an answer at all? I mean I already voiced my opinion in the post
 
And posting a "yes we should add this" doesn't add a whole lot, because it's already obvious that you agree with the idea (since it's yours)
 
@Οurous Yes, an array or tuples would also be fine.
 
@DJMcMayhem so I just wait until someone answers or comments, right?
@Laikoni Is there a particular reason you chose Merge sort? (I think there are other algorithms that have a similar way of recursively spliting up the list, right?)
 
10:40 PM
@Laikoni would you want to look at invalidating any merge-sort builtins? I know they wouldn't do the challenge for you, but I can think of a few ways to hook into the merge-sort implementation in Clean to have it return the sorted list and the required output. Clean isn't really the most golf-y language, so I imagine there are ways to do that in others as well.
 
@flawr Quick sort also recursively sorts two sublists, but there you need to pick a pivot element and compare the other elements to it. So the nice thing about merge sort is that it works no matter how you split the lists, which might be interesting for golfing in some cases.
 
@flawr Yeah pretty much
 
@Laikoni Ah I see, that makes sense:)
 
@Οurous I'm not a fan of disallowing built-ins, as this usually creates some ambiguity. I also do not think that any language will solve this challenge just by a built-in itself. So if you can use some merge sort built-in creatively to solve the challenge, I'm happy to see the answer.
 
I think built-ins comes with implementation details which are specific to the tools you use and requires a certain level of mastery and awareness of your language to make use of. Such as perl counting lines in $., knowing these features aren't cheating but rather using the tool the way it's designed for.
They don't introduce unfairness in the sense that they're already there to be studied.
 
10:57 PM
@Οurous Btw the other day I had a look at Clean, probably after seeing one of your answers, and thought: "The syntax looks quite similar to Haskell". Only now Wikipedia told me that Clean is actually older than Haskell and indeed one of its influencers.
 
@Laikoni Yeah, as strange as it sounds, Haskell is like real-world Clean. Clean is both blessed and cursed by it's development team being a handful of computer science academics.
 
anyone want to duel for 50 rep in newline?
@Pavel you want to?
 
?!?!
do I smell rep
 
@Christopher2EZ4RTZ Too busy playing rainbow six
 
11:03 PM
@Pavel priorities (you got them right)
 
Which I'm streaming by the way: twitch.tv/cat_dev_random
 
yay
@Unihedron you game?
@Pavel how much language should I expect? (no headphones)
 
I think I'll lose, so nah ^^
stakes2high4me
 
@Christopher2EZ4RTZ I won't be swearing if that's what you mean
 
@Pavel yeah I just don't want my sister to overhear some rage rant
@Unihedron :_(
 
11:06 PM
I'll try to avoid anything like that
 
I generally don't yell at my teammates
 
i see you don't play much dota then
@DJMcMayhem you want to try?
 
Well I yell at my teammates in Dota :P
 
@Pavel good child I say to the person who is ~1 year younger
 
11:08 PM
Yay found a match
 
ROUND ONE FIGHT
 
Problem: Teammates are swearing
 
rip
Internet is slow so can't watch :(
@Unihedron caps
 
4
Q: Visualize Merge Sort

LaikoniMerge sort is a sorting algorithm which works by splitting a given list in half, recursively sorting both smaller lists, and merging them back together to one sorted list. The base case of the recursion is arriving at a singleton list, which cannot be split further but is per definition already s...

 
11:16 PM
where is the regex god when you need him :(
 
@Christopher2EZ4RTZ who’s that? :P
 
martin ender
 
do you need help?
 
i need the right side of numbers from this: oeis.org/A002852/b002852.txt
and he has retina the language that will do that but I don't know the language
 
well I do
 
11:19 PM
would you do that for me?
 
I mean, even pyth can do it
but sure
 
the code is:
\d+ (\d+)
$1
paste your input to the "input"
 
thankkkks
 
@LeakyNun exactly how I would have done it
 
11:20 PM
XD
 
@micsthepick hmm... need to golf it then :P
 
@LeakyNun dammmn
 
m`^\d+
(with space and newline afterward)
@micsthepick here, golfed it
change \d to . for more golfing
 
:P
. would match the whole line
 
but the space afterward would tell it to stop
 
11:22 PM
true
 
Ok golf challenge: now take the input and replace every newline with a comma
 
@LeakyNun have you tried that?
 
m!`\d+$
7 bytes
 
how to get people to do my work for me
 
@micsthepick yes
 
11:23 PM
(for One OEIS after another)
 
2 mins ago, by Leaky Nun
(with space and newline afterward)
@Christopher2EZ4RTZ after or before removing the indices?
 
what?
ohh after you run the first program you run the comma replacement one
@LeakyNun
 
m!M`\d+$
¶
,
(all-in-one)
just take the last two lines if you've already replaced the input
12 bytes
 
new golf challenge: replace a 1 or 2 digit integer with the next integer.
 
11:28 PM
@Christopher2EZ4RTZ Playing DOTA? nah, I'm good
 
@micsthepick don't get it
 
@DJMcMayhem at dota? :P
 
oh
you mean by retina?
 
@Christopher2EZ4RTZ Ha!
No way
 
I've done that to death @micsthepick
 
11:29 PM
@LeakyNun if you dare!
 
Most definitely not good at DotA
 
i am ok at it
I only have 400 hours
 
I might play some GOIWBF right now
 
hahah goooooood luck
 
@LeakyNun looks good.
 
@user202729 Although I'm not sure how much this interests you; Funky has now actually got return statements, although their implementation does not break your OEIS solution.
 
@cairdcoinheringaahing how was my answer invalid?
annnd my answer is ded
 
@Christopher2EZ4RTZ It was decided a few answers ago, that while being technically valid, people should only hardcore if the sequence is finite. The first answer that hardcodes (the Rust one) is at about +10/-10 as an example
 
00:00 - 20:0020:00 - 00:00

« first day (2511 days earlier)      last day (708 days later) »