« first day (2850 days earlier)      last day (2291 days later) » 

00:45
bad code challenge: crash somebody's computer with web browser javascript
can plausibly be done with rowhammer but plausibly doable doesn't help here
01:08
@Joshua entire computer? very hard, most have already been found by security experts
flash... it might be (a lot) easier with flash but that's not js
@Neil 66
Somebody build a computer with the smallest amount of memory to browse the internet. Then just start opening tabs, possibly voilà
also @Neil i'm assuming for(9) Print([3]); is shorter than oblong/Times([3],9)? (haven't tested them yet)
@Zacharý only on a very, very bad browser. which falls under security holes probably (well not really)
@ngn because ASCII is a complete acronym by itself :P
hmmm wait i messed up
@Neil 65?
01:35
01:48
@feersum doesn't hilbert curve use 2x2
@ASCII-only Didn't I mention that in my message?
oh
so that's what you mean
sadly this is hilbert not peano
 
2 hours later…
03:56
@Downgoat you online?
04:33
> Literate programs should not be totally different modes of the parser/interpreter (contrast rhoScript or "big Pyth"). Literate Paradoc and golfed Paradoc should just be styles of programming used by the programmer that can coexist in one program, and it should be possible to manually translate between them.
> Debugging, at least printf-style debugging if not more, should be natural and easy.
hmmmm
oh also. futuristic???
@ASCII-only yessir
How do most people pronounce 05AB1E? I pronounce it 'osabie' like 'wasabi'
@ASCII-only :O you are number two
hmm. i'm thinking of making charcoal more general. idk how tho
@Downgoat only because most people don't add tags. also, just look at all the SE people there O_o
@ASCII-only we should work on remake with extensible parsing
like multi-file
but that requires multi-stage resolution
@Downgoat :O good idea
we could compile to LLVM :D
04:42
@Downgoat but extensible means what???
@ASCII-only meaning that it can handle multiple 'statements'
@Downgoat example?
@Downgoat good idea but takes time, so... later i guess
@ASCII-only hmm idk example
@ASCII-only actuly very easy
wdym by multiple statements
if we write in C-compatible language like vSL or rust
or python
or go
Has anyone here tried to make a large-scale project in a fucntional lang?
04:47
hmmm.
i probably need at least 1. non-canvas output and 2. better function support
oh yeah do you know if it's possible to use browserify to bundle third-party modules by themselves
05:10
the institute for social ecology account that spends all day searching twitter for “bookchin” did not seem to comprehend my sarcastic tweet and accidentally liked it
i feel powerful
 
2 hours later…
06:42
0
Q: Irrational arrays without strings

guest271314Write a function that takes input as an integer, decimal or irrational number and outputs an array containing the integer or if a decimal number includes the last integer and decimal separator portion of input set as a single element with remaining integer and decimal portions of input set as adj...

 
3 hours later…
09:38
@NewMainPosts oh god
 
2 hours later…
11:44
-1
Q: C++, JavaScript and Python polyglot

PolyGlotPlease help! I have a JavaScript source file, a Python6 source file and a C++ source file that I want to merge into a single polyglot source file. I can polyglot JavaScript_Python6, Python6_C++ and also C++_JavaScript, but can't get a single file to polyglot all three. Any help gratefully appre...

@NewMainPosts "Python6"?
@EriktheOutgolfer :o
12:23
@lirtosiast hm, so it looks like you've now requested access over JHT, right?
 
2 hours later…
14:18
0
Q: Area 51 Proposal: Math Challenges (bit like PPCG)

TheSimpliFireOn September 1st, I started a proposal on Area 51, for users to share, discuss and answer mathematical challenges that they have. I did this because I felt that these types of problems have no place in any of the current, relevant sites about mathematics/puzzles. If you want to envision the prop...

14:39
@EriktheOutgolfer Python 2 * Python 3.
what is (language * language)? ;-)
that is, it runs in both. i have seen it used before although not really like that
you can google "python6" with a few direct results, but the term is actually pretty shady...
14:52
So, basically, print always uses (). Always put list around range-like classes in 3, and make sure to use //. Most of 2-3 incompatibility, I think.
Or it could be Python 1 * Python 2 * Python 3.
15:04
I am thinking about setting a longest palindrome subsequence challenge and trying to make test cases.....
what is the answer for localareaman ? I have aaaa but I worry there might be a longer one of length 5
@Anush Not necessarily contiguous, right?
I can think of a simple O(n³) solution and a simple O(2^n*n) solution.
@Anush for reference, there's aaraa too
you can make a reference implementation and make test cases out of that
make sure it's straightforward and easily verifiable
@EriktheOutgolfer oh right!
aaraa and aaeaa look like the longest ones to me
thanks
I just wrote some python code to give the length.. I need code to output the actual sequences too
wow.. it would surely have been easier to write in python :)
you can see that's a trivial challenge for Jelly
Well, Jelly is very easy to type in LInux International keyboard layout.
yes.. although I am guessing that code won't work so well with an input of length 100?
15:15
Of course, Jelly is not designed to be fast...
Thinking a bit more, n² would be easy too.
Given that if this problem can be solved efficiently then longest common subsequence can be solved efficiently too, it may be worth reading LCS algorithms.
but it needs a little more code to actually output the subsequences
@Anush That one is O(2^n) in worst case too...
@user202729 well I need to add memoization :)
Then it would take O(n³), and with some optimizations O(n²).
I think O(n^2) with memoization
why do you say n^3?
15:23
@Anush String slice is O(n), unless you store a stringview (first:last pair of indices)
oh I see!
good spot
it seems a worthwhile codegolf challenge with restricted complexity
Mini challenge: prove that the time complexity of longest common substring ≤ time complexity of this problem.
where you have to output the full subsequence, not just the length
(any algorithm for this problem can be used to solve LCS problem with linear explosion in input size)
Unfortunately, it seems that the fastest theorical algorithm known for LCS is O(n²/log n).
true
although there are various solutions parameterized by different things
e.g. the length of the LCS
@Adám you can click on the network profile of both and see that no, they're not.
(at least I'd assume that's how it works)
How does one ping one of two users with identical names?
@Adám If both users have been in a room, no idea. (maybe a mods superping?) If only one has, only that one will be pingable
@dzaima Right. When I start typing @p I get one name with two icons…
@Adám oh heh
that should ping both IIRC
16:31
@Adám Dennis can't see the IP address of both...
@EriktheOutgolfer AFAIK, mods can see the hidden network profiles.
you need three letters to ping
But because the info is only available to mods...
@LeakyNun I am using all the letters, but the two have identical usernames.
@Adám If I was able to find out, I wouldn't be allowed to tell you.
16:35
@Dennis Oh well, sorry to disturb. WIBNI there was a way to ping a specific user?
@Adám That's a mod feature. :/
I think reply-to-message should work.
That requires a message though.
@user202729 This challenge is just lcss with the reverse. QED
16:41
@37160 ← this should work.
@PostLeftGarfHunter :/ it actually works
@Adám Should not work. That is a mod feature right?
With @@, yes.
@user202729 I don't see why that would need to be restricted.
@user202729 What actually works?
@Dennis But that's a super-ping even though the user was never here, right?
16:42
Yes.
So any reason why just a single-@ with a userid should not work?
@Adám SE doesn't like working on chat
Initially I thought about having the first half in the string and the second half in its reverse, then it's necessary to check whether the sum of two endpoints are small enough, etc.
which is much more complex.
@dzaima Probably because chat.SO isn't very popular.
Oh yeah. No need. You can guarentee the end point are the same by not splitting the string
16:45
However, my mini-challenge requires converting a lcs problem instance into a longest palindromic subsequence instance.
(the opposite direction)
Actually I think I only ended up proving that lcss >= subpalindrome
(right)
Hm the other way is not immediately apparent to me and I'm about to die
(assume the two strings has equal length. If not, I have no idea)
You can pad strings to make them equal length
Just choose a character from your alphabet not in either string
or add a character to your alphabet if none such exists
16:48
But nm ≤ max(n,m)², so it's worse.
hm
Why do you need equal length?
Solving lcs(n,m) takes about n×m. There is no way to transform it into a lps instance with size √(nm) right?
(is there? I can only convert lcs(n,m) into lps(O(n+m)) )
(assume lcs(n,m) takes n×m and lps(n) takes n²)
I'm not sure I get the question
I was thinking of putting another bonus on codegolf.stackexchange.com/questions/175470/… but is it better to put a "permanent bonus" on in some way instead?
16:51
Find a way to transform a lcs instance of two strings with len n to a lps instance with len O(n).
it seems sad to have only one answer in Haskell for a fastest-code question
s/bonus/bounty/
I don't really want to solve that... (Anush's challenge)
@user202729 Why do you need them to be the same length?
16:53
@PostLeftGarfHunter Because I still can't find a way to convert lcs(n,m) to lps(√nm), and converting it to lps(max(n,m)) would be worse.
I think this is best if we talk about order notations in terms of a single variable.
@PostLeftGarfHunter That's another reason right?
@user202729 :( But why not? :)
No instead of talking about the length of separate strings we talk about the length of th total input.
I'm not sure if you can turn lcs into calls of lps
I think the easiest way to prove this would be to find an algo that does lps in better time than lcs
17:32
@user202729 if there's a link to the network profile, the profile isn't hidden
@Adám a username can perfectly consist of only digits...
take, for example, 12431234123412341234123
@EriktheOutgolfer OK, then @:123456 or some other syntax
@Adám which is exactly what @@ is
@EriktheOutgolfer OK, but that's a superping. We need a normal ping-to-userid rather than a ping-to-username
why not just make @@ not superping for regular users
@dzaima Doesn't superping go to a SE userid?
17:38
@Adám no idea ¯\_(ツ)_/¯
@dzaima because ew
I would never want to get pinged in a chat room I've never visited
@EriktheOutgolfer I think he means that the syntax is available, but doesn't have superpower for normal users
@EriktheOutgolfer I specifically said "not superping"..
18:01
@ASCII-only It's the same length as Print(Times([3],9)); and shorter than Oblong(3, 9); Move(9, :Down);
18:34
@EriktheOutgolfer Yes
 
1 hour later…
19:48
1
Q: Number of transformations until repeat

mromanGiven a sequence of integers or to be more specific a permutation of 0..N transform this sequence as following: output[x] = reverse(input[input[x]]) repeat For example: [2,1,0] becomes [0,1,2] and reversed is [2,1,0]. [0,2,1] becomes [0,1,2] and reversed [2,1,0]. Example 1 In: 0 1 2 S#1:...

@NewMainPosts uh... what?
> output[x] = reverse(input[input[x]])
how do you reverse a number...
0
A: Sandbox for Proposed Challenges

lirtosiastDisplay a rational tangle code-golf math arithmetic ascii-art In a quest to classify mathematical knots, J. H. Conway discovered that certain simpler structures called rational tangles can be uniquely represented by rational numbers. A tangle is an arrangement of two strands of rope such that ...

so, it looks like it should actually say "output = reverse(input[x] for x in input)"
20:15
0
A: Sandbox for Proposed Challenges

AdámSplit and recombine a number This code-golf challenge has two related parts. Your task is to write two functions/programs as per the below specifications. You may share code across your submissions, the submissions may call one another, and you may even submit a single submission which handles b...

20:40
@lirtosiast you can do it in 8 bytes :P
coming closer... just one small thing
btw, there's a bug
_ isn't included in the submission itself
@EriktheOutgolfer Thank you for the friendly heckling :D
heh, that's not too uncommon a thing, especially since it was supposed to be my solution ;-)
(Jonathan Allan first did that to me...)
 
2 hours later…
22:23
@ASCII-only elaborate?
To those who enjoy long challenges: Should I award bounties? Would 500 points motivate you to spend 5 or 10 hours on a problem you wouldn't otherwise?
Anonymous
@lirtosiast First I'd have to find 5-10 hours to spend on PPCG challenges :P
Doesn't have to be consecutive.
@lirtosiast It depends. There are some challenges i find interesting but might not be motivated enough to put in the time without a bounty. Then there are challenges I just don't find interesting and wouldn't do either way.
22:37
Yeah these would be long term bounties that I promise in the question or bounties list, but don't actually put on the question until I'm ready to award
Of the form "First seriously golfed solution gets 500 rep, any improvement of k% gets k*10 rep with a one-time bonus of 200 if it's in a different language from the previous best, further bounties awarded to quality answers on my discretion" or something
23:14
@Downgoat like let's say i want to bundle up brotli so i can just include it as module browser-side
@lirtosiast doesn't work like that
you must at least double previous bounty, max is 500 rep
@ASCII-only yeah totally
we do something like that for Axtell
@ASCII-only Oh right
Probably just 500 for first answer and another 500-1000 for whatever needs further encouragement then
@Downgoat but just sincle modules?
@lirtosiast nah 500 is max. you cannot post another bounty after a 251+ one iirc
Pretty sure I've done just that before
oh, huh. maybe you can then
23:20
@ASCII-only yeah like we don't want to include slug modules or Ace or markdown modules until needed so we compile into seperate files that webpack automatically then gets with a dynamic import()
:O nice
@Downgoat any chance you'd be able to compile brotli and lzma into modules?
@Zacharý no way, 1 is too different
trying the nandgame.com but i can't drag any components in >:(
@DestructibleLemon you broke it
tbh it is pretty bad
23:28
no i didn't
@DestructibleLemon any adblocker/noscript stuff?
ok seriously though how do i drag the stuff
@Zacharý also. most? nowhere near. one of the most glaring issues is differences in unicode handling
@flawr it might be that
@DestructibleLemon :|
23:29
have you tried a different browser?
:| :| :|
or one in safe mode?
or incognito
how would incognito help?
anti-adblock killer apparently was doing it >:|
23:31
Ok, I needed a minute to process this tripple negative
@DestructibleLemon wat?
@DestructibleLemon uninstall then :|
@DestructibleLemon what browser are you using?
oh i turned it off for now
its working now
@ASCII-only well that would be part of build process with webpack
yay
Anonymous
23:32
@ASCII-only Ads are bad. Adblock blocks ads. Ad distributors don't want their ads blocked, so they started adding nagware to "encourage" users to disable adblock (read: hold site content hostage if users are using adblock). Now there's a back-and-forth war.
hmm
clearly we just need machine learning to solve this once and for all. on clientside ofc
sometimes i just click inspect element and delete stuff
epic hax
there is also ublock origin
@DestructibleLemon you know you can right click an element and add it to your personal ublock filters right
23:34
See also: radar detectors which detect and/or avoid being detected by radar detector detectors
that's so rad
@lirtosiast like when I was a kid and my mom suspected that I knew where she hid the cookies and secretly made some cookies she knew I didn't like and I continued to get some of them just so that I can make the argument that it couldn't have been me that stole the cookies because I don't even like them.
@flawr :o
What a kid you were
turns out my ruse didn't work as she found crumbs on my bed
but in my imagination it was the perfect crime
Anonymous
@flawr You could've claimed you were crashing cookie airplanes together
23:41
@Mego shhh, don't let her find out about the airplanes
Anyway, now I'm a mature adult and I can eat as many of the damn cookies as I please.
CMS: What were your childhood crimes?
My elementary school had [mascot] dollars, which were slips of paper you were awarded for good behavior and could turn in for small rewards / use as raffle tickets
In kindergarten I tried to steal them
Anonymous
23:58
@lirtosiast Similar story: kids would get 1 reward token for turning in homework, and an additional 1 if they got an A. I bribed kids into giving me one of the two in exchange for me doing their homework for them.
this chat is turning dark

« first day (2850 days earlier)      last day (2291 days later) »