« first day (2127 days earlier)      last day (3011 days later) » 

20:01
@TuxCopter also can you help me debug this code?
40 mins ago, by noɥʇʎPʎzɐɹC
unordered_map<int, pair<int, int>> seen_times;
seen_times[cur_cost] = make_pair(get<1>(cow), get<3>(cow));  // this runs, the debugger even steps into make_pair
// and yet my debugger still shows seen_times as having size 0
1. I set the value
2. The size is still 0
3. ????
idk how to C++
C++ is horrible. I need it less horrible.
Stick to good old C.
My situation is that I need to do stuff for a competition with a four-hour time limit and the contest is scored by program speed
I cannot do that in C
20:07
Use Rust.
@noɥʇʎPʎzɐɹC Wat?
@mınxomaτ They don't support it.
Or nim.
Wut competetion
A very weird one visibly
@mınxomaτ Official USACO - united states computing olympiad
Petition them! Do it for me!
20:08
@TuxCopter Not really; most competitions only do C C++ and Java
Since that's usually the mix of what you see at schools.
Occasionally Python
Use C to compile Rust and then use it.
@quartata British Informatics Olympiad supports every language as long as you send the marker an interpreter/compiler for it
Woo-hoo! I capped again for the first time in a while!
20:09
@noɥʇʎPʎzɐɹC What's the objective?
@quartata If they support java, they usually support Python, too.
@BlueEyedBeast That's unusual.
@KritixiLithos Solve the problem as fast as possible. (within two secs)
@mınxomaτ Huh?
Nim transpiles to C but it requires linked libs they don't have
I can't used boost
@mınxomaτ I'm starting to see Python more, true
@BlueEyedBeast as far as I'm aware - they let me do one question in Pyth
20:10
Used to be a lot less common
@BlueEyedBeast ahahaha
There was a challenge where Pyth was fast enough to calculate the answer? :P
@quartata It was a nice and simple 1 byte builtin as well
Ah, fair enough
You support Python which never wins.... and yet they never support Rust or Nim!?
yesterday, by quartata
Does anyone know where I can get the full xkcd Hoverboard comic as an SVG?
Still haven't been able to find one
20:12
@quartata Do you want a DDoS?
The xkcd Hoverboard comic?
Does anyone else have a yucata.de account?
@PhiNotPi watiz
^
That's it; I'm making my own darn language that transpiles to C++
Good luck
20:15
It's a website for "online board games."
@noɥʇʎPʎzɐɹC ???
Is this a threat or what? :P
@quartata the svg would be huge
@quartata Only $1.99 to buy a DDoS!
@noɥʇʎPʎzɐɹC wat
@noɥʇʎPʎzɐɹC Wouldn't be all that big.
@quartata Depends on how you build it
20:17
Yeah, I don't need it to scale.
I can do that myself obviously
A lot of it is straight lines so with just a regular path it probably wouldn't have a lot of points
I just spent five hours over 2 months trying to figure out how to use Bootstrap to do something which can be done in 3 lines of CSS 0_0
Hey I was wondering what your guys opinion is for leaving the output on the stack for stack-based languages. I saw this answer by Martin Ender meta.codegolf.stackexchange.com/a/8507/62670, but this feels like cheating in ><> (especially because characters are displayed as decimal instead of unicode on the stack, for the online interpreter at least). What are your guy's thoughts?
It's only for functions
A 'full program' must output the value on STDOUT or it closest equivalent
I think the general consensus is that functions can leave the output on the stack, but if it's a program, it needs to send the output to some form of STDOUT.
20:26
I like how HN mods silently edit the title of your post sometimes.
Okay, thanks. I guess ><> would be considered "functionless" then? Because to me, a function is just storing where I am on the stack, jumping to the desired function, then ultimately jumping back to where I was.
><> has a goto command?
Like that would be a function in asm, no? So I just find it a bit confusing I guess. Should I ask a question on meta?
20:32
IBM will give you 5k if you implement a thing: bountysource.com/issues/…
@noɥʇʎPʎzɐɹC that page is just white for me O:
GitHub said "yo I'll give 1k to y'all if you fix this" to Firefox
20:49
Linode issue resolved. 3 hours later. All backups gone.
@mınxomaτ so you lost everything? [dumb question]
All backups-
GitHub hasn't fixed the issue that allows automatic account creation. Many bogus accounts + Travis + AppVeyor = free computing power.
@mınxomaτ What have you been up to??
Building a botnet for free with GitHub and CI services.
@mınxomaτ and who's the victim?
20:55
No one, I'm just documenting this for a blog post. And crunching some numbers that needed crunching.
What's the botnet's infrastructure?
My PoC currently takes about 30s to spawn 5 VMs (that's registering, verifying, setting up a GitHub account and attaching 5 Travis tasks).
Do you have a botnet that constantly migrates from exploit to exploit as a standing army/threat insurance? :P
@mınxomaτ How do they report back?
@noɥʇʎPʎzɐɹC A master account contains the mapreduce code. New bot accounts just fork it.
@DJMcMayhem Huh, what’s the trick?
(To 8 bytes on reverse the input.)
20:59
If the task finished, the bots make a pull request on the master
I don't see how GH can prevent this kind of abuse
@mınxomaτ Supposedly the botnet will stay even after the bug is fixed...
@TuxCopter a CAPTCHA
@noɥʇʎPʎzɐɹC inb4 machine learning
though you can use services that will read CAPTCHAs for you
@mınxomaτ Hm, the code must be public
@noɥʇʎPʎzɐɹC I'll send the list of bots to GitHub in my report.
@noɥʇʎPʎzɐɹC That's ... the thing. You see, if GitHub's algorithm detects that you are a bot, all your repos become unlisted. But everything still works. That's actually helping.
21:04
Pattern of bot names?
I see you're doing this independently of the bugs discovered by the automated tools
I have an algorithm that creates a human-like profile. Complete with Bio, name, company etc.
@mınxomaτ Like Social Engineering, but to trick bots
Bot Engineering
@TuxCopter But yeah, unless GH implements some kind of challenge, creating bots will always be faster than detecting and removing them
@Lynn the filter command. !Grev<CR>ZZ
Pretty much the same idea
Classical CAPTCHAs can be solved automatically
21:08
@noɥʇʎPʎzɐɹC Even reCaptchas. Just opt for the "audio" version for disabled people and pipe that into the Google Speech API. Success rate is acceptable.
1
Q: How do We Define "Function"?

redstarcoderI got the idea for this question after reading an answer by Martin Ender here. It more-or-less says: Functions in stack-based languages can leave the output on the stack. This had me ask "what is a function?". I use ><> and to me, a function is just when I store my current location on the stack,...

@mınxomaτ Have you used Tor and gone to a CloudFlare protected site? The text is unreadable and the sound is alien
@noɥʇʎPʎzɐɹC Uh, SE is on CloudFlare. Works fine for me :-)
@mınxomaτ Go to one that's more sensitive to threats ; like strawpoll.com
Tested
@noɥʇʎPʎzɐɹC I'd call this misconfigured, not more sensitive.
21:11
@mınxomaτ Lots of sites get super mangled captchas, not just that one
@mınxomaτ What's this? github.com/turbo/Alkaline
TIL you can have multiple box-shadows on one element
@noɥʇʎPʎzɐɹC A sort of last resort against IoT botnets which I've not yet published.
@ETHproductions It actually make sense
@ETHproductions this is super useful when you want to shadow all sides
@mınxomaτ What does the code do to make it a last resort?
21:18
I guess
@noɥʇʎPʎzɐɹC Brick affected devices.
@TuxCopter without jQuery
@mınxomaτ where does this run? It still can't prevent Layer 2 attacks, if it's not on a firewall
@NathanMerrill Well, box-shadow is by default on all sides, so it's more useful when you only want the top and bottom to have shadows, since you can really only wrap four or one sides with a single shadow
Well, actually either brick or make them attack each other.
@TuxCopter Surprisingly accurate guess. :P
21:20
:)
@mınxomaτ Where does this run?
@noɥʇʎPʎzɐɹC It's P2P. It runs on the devices.
@ETHproductions right, but if you use offsets, then that messes everything up
So once run, you can't stop it.
I see....
Why can't the malware just delete Alkaline? (it can't, but why?)
21:21
@NathanMerrill Ah, yes, I see your point now
It also incorporates the H-Bomb code from one CloudFlare engineer that makes OpenDNS resolvers attack each other.
@noɥʇʎPʎzɐɹC That's a race condition basically.
git init antialkaline
git init antiantialkaline
git init meta-alkaline
Let's not have a race condition in chat, eh? :P
21:23
Bwhu?ty
git init (anti)*alkaline
We shouldn't use race-conditions. That often leads to threads.
4
Yogiisim for ppcg :P
TIL what race condition means
Just attach remote mines to every IoT device ever
21:25
IoT devices are remote mines.
2
Remote mines would be so useful in Age of Empires...
... note to self: previous remark not witty enough for stars.
4
I am sure ^ is an evil plan to get stars
^ best emoji ever
Someday I'll learn how to get stars like I get upvotes. Problem is, stars can't be earned by golfing...
21:33
Awww, I'm flattered.
^ is, what, a cheshire cat frown?
∠_∠
@ETHproductions Sr th cn.
Question: How to reload a file in vim?
:e! if that's what you mean
21:35
@HWalters In TNB, carets ate typically used to agree with or refer to the previous message. A double caret refers to the second message before, and so on.
Try not to use them frivolously though.
@El'endiaStarman Sometimes being overly literal is a form of humor
@El'endiaStarman wat I thought multiple carets was super agreement
21:37
They can also mean that, and also be a face, so they're really ambiguous
Oh neat vim automatically reload files
@noɥʇʎPʎzɐɹC Nah, bold them for that.
@trichoplax ^_^ problem solved
If you get to something like ^^^^^^^^^^ that's probably super agreement, but ^, ^^ and ^^^ refers to agreement with x messages above
The chat reply links are a great way to avoid ambiguity
21:37
^
@El'endiaStarman Or superscript. ^², read "caret squared."
@trichoplax Yes, definitely.
@Doorknob That's the same as "^^" though, no?
21:39
@Calvin'sHobbies Hmm, I would interpret two carets as referring to two messages up instead.
@ETHproductions ಠ̯ಠ
In any case, the ambiguity is all the more reason not to use them and just describe things in words instead. :P
@Doorknob But like x^3 = xxx. Or yeah we can just say "I super agree" :P
> how do you do that fancy thing where you send a blockquote message
> oh like that nvm
21:40
Ugh, I just now noticed that one of my previous messages has "ate" instead of "are". So annoying how my mobile keyboard likes to pick the former when I'm trying to type the latter.
> carets ate typically
That's the trouble with typing when you're hungry
@trichoplax I just are, though!
Described with words, like, this? "imagine two dots. Underneath the two dots is a straight line, with the righthand side forming a chord in a semi-ellipse, broken at symmetry along the shorter axis. Now rotate this image 90 degrees counterclockwise into standard smiley format"
(That was intentional. :P)
21:42
:P
When I'm typing on mobile, the keyboard autocorrects me wrongly. When I'm typing on a computer, my hands do that instead. :P
@all Which do you feel you are better at managing, time or space? Like, if you plan your week's schedule are you more likely to stick to it, or if you clean your room is it more likely to not get cluttered again?
I'm great with space, terrible with time :/
Space for me. I shouldn't be chatting right now :P
I'm bad at both
I'm not sure... I'm actually delaying cleaning up atm
21:44
@Calvin'sHobbies I don't know which is worse, but I'm terrible with both
@Calvin'sHobbies I'm great at managing my energy (AKA being lazy), which is somewhat related due to the whole relativity thing.
If I had all of time and space at my disposal, I'd probably waste all of both
I'm...not sure, actually. My spaces do gradually get more cluttered over time, but also, my schedules don't get cluttered that badly due to my introversion.
@ETHproductions But that would be bad time management, no? Like procrastination.
Exactly. I'm better at managing space
21:46
I'm certainly better at space management than time management.
> Which do you feel you are better at managing, time or space?
@ETHproductions right click > The Great Suspender > Suspend Tab
As in, I tend to use my space more effectively and efficiently.
@ETHproductions I can't read or remember
@Calvin'sHobbies I'm the same as you : I can manage space well, but I hardly ever have my time planned
21:49
@Calvin'sHobbies c) None of the above.
Any C++ macro recommendations?
Don't use them.
@quartata reccomend c++ macros
#define contains(x,y) (x.find(y) == x.end()) //works for maps
#define index(y,x) get<x>(y)  //tuples
@Dennis Context: programming competition with time limit
21:51
-1 not enough rickroll
@TuxCopter A waste of time and space
@noɥʇʎPʎzɐɹC I highly doubt typing speed would be a bottleneck.
@Doorknob Things to help reading code easier; like the macros I posted
@noɥʇʎPʎzɐɹC avoid them at all costs
@noɥʇʎPʎzɐɹC chaospp
21:53
@DJMcMayhem What about the contains and index functions?
@Calvin'sHobbies I'm terrible at both. :P
Probably slightly better at time
@noɥʇʎPʎzɐɹC If I was working with code that had those, the very first thing I do would be to remove them
why does this question exist
@DJMcMayhem It's not production code. It's for a competition. Once it's written, it's thrown away
@noɥʇʎPʎzɐɹC contains as you defined it is not useful
@Flp.Tkc Because FAQ Q/A
21:55
Just do x.count(y)
@HWalters renamed to mcontains
Only works on maps
x.count(y) is less characters even; no need to compare your iterator to end just to see if the map x contains something
@noɥʇʎPʎzɐɹC What difference does that make?
@HWalters ok I'll use that; is it fast?
It's roughly as fast as your map; wasteful if you need the iterator, but I thought you were optimizing for coding speed
The macro's equally wasteful in this regard
22:00
@HWalters ok
the two real things you need with a coding competition is 1. A smart brain and 2. familiarity with the language
@NathanMerrill I don't have 2.
Then why are you doing the competition???
does it have to be in C++?
@NathanMerrill C++ or C.
Has to be.
22:01
C++/Python/Java/Pascal IIRC
@TuxCopter Fun?
can you compile?
find a Python to C compiler :P
^ Bad Idea™
I'm sorry to be a bit late to the whole thing, but what's going on with this site's graduation from beta?
@NathanMerrill They require Statically Linked Libs™
And so does Nim
I opened an issue on GH
22:02
@Flp.Tkc We are graduated and waiting for a design
@Flp.Tkc It should graduate in 6-8 weeks (Read: Never)
@Flp.Tkc Any dates you read about graduation are probably jokes
@NathanMerrill nope; provide source; this plays
@noɥʇʎPʎzɐɹC who is They?
@NathanMerrill Python to C compilers
22:03
all of them?
Please, really do petition them - usaco.org
@NathanMerrill Yep; tested
They don't even support PyPy!
@NathanMerrill Nope
Well I've tried
Cannot get intermediate C source
I'm considering making my own lang
Haskell can compile to C IIRC
@noɥʇʎPʎzɐɹC that sounds like an even worse idea. if you aren't familiar with C then writing a compiler into it sounds like a host of bugs for you to find on competition day
22:05
@NathanMerrill I'm going to practice with it
then just use C :P
I'm practicing vanilla C++ right now
@NathanMerrill Basically it generates incomplete C++ code
Why don't you just use Java?
Why a sane person would use Java?
Also Cython like basically all such tools requires libpython
22:07
Why don't you just use Javascript?
@TuxCopter Well it would be the saner alternative compared to C++.
Right, but if speed is the goal of the competition Java is a Very Bad Idea™
@TuxCopter Because it's better than C++ if you don't know anything about either and he better get used to it now before being confronted with the same restrictions in college
They wouldn't have offered Java if it wasn't fast enough
^ ANN music, autogened; sounds great
@TuxCopter Why? If whatever competition he's doing is like any competition I've ever seen, the language speed has almost no impact at all compared to the speed of the algorithm.
Java has a slower runtime time compared to other languages
22:11
@Doorknob Yes it does.
@TuxCopter To what languages?
Yeah language speeds do not usually contribute orders of magnitude differences which is what they care about
C/C++/Rust/etc...
@TuxCopter Did you read my message?
@quartata The biggest testcase is sometimes unsolvable even with a fast language and the fastest algorithm
22:12
They design these problems so that if it takes longer than the time period it's because you're doing it wrong not because of the language
Java is also faster than you think these days
@TuxCopter I doubt rust is really faster than Java.
Rust is compiled to machine code, Java runs on a VM
So what?
Brainfuck is compiled too.
@quartata In an ideal world (evidence coming up)
> Note also that due to the slower speed of a Python program, it may not be possible to solve the largest test cases for some problems even with the inflated time limit given to Python submissions -- consider using a faster language for problems where execution time is critical.
direct from their website
Rust is also in beta and its compiler has not had 20+ years of work like gcc.
@noɥʇʎPʎzɐɹC Yeah that's Python. CPython is molasses
22:15
@quartata They don't support PyPy.
But we aren't talking about Python
We're talking about three compiled languages
@quartata I'm talking about:
> They design these problems so that if it takes longer than the time period it's because you're doing it wrong not because of the language
contest hosters said against so explicitly
@noɥʇʎPʎzɐɹC ♫ PyPy Miss American Py ♫
So it sounds like this particular contest is an exception
@trichoplax It's the official US one
@flawr ??
22:17
No they didn't. The problems are likely not designed for Python. I'm also pretty sure you said Python was not allowed anyeays
@flawr wrote my python with the keyboard but the runtime was slow
@quartata It is allowed.
Interesting inclusion of Pascal there
c/c++/pascal/python/java <- that's it
22:20
@noɥʇʎPʎzɐɹC What version of Pascal btw?
1.7.0_65 Java executables can't even be found!
> And Guido van was getting risky with tries
idk
If we're gonna sing about paɪ it should be pi
CMC: Find a Wikipedia page where the Talk page is not yet created
@TuxCopter done
Side Stage is a technology presented by Canonical Ltd. to be included in Ubuntu Tablet. It was first demonstrated on Ubuntu Tablet's YouTube announcement on Canonical's channel "celebrateubuntu". == Summary == Side Stage allows Ubuntu Phone apps and select Ubuntu Tablet apps to run in a smaller section of the screen, alongside the main app. It aims to "go even further" with the idea of multitasking, allowing screen space to be divided in this manner. Examples shown in the announcement video included a notes app being used alongside a web browser, and a user swiping from the right edge to bring...
22:31
1
Q: Stewie's sequence: + * - / + * - /

Stewie GriffinLet's use the four basic operations, addition +, multiplication *, subtraction - and division / (float, not integer). We'll number these operators from 1-4 in the order given above. Stewie's sequence is defined as follows: x = [x(1), x(2)] // Two initial numbers (one indexed) x(3) = x(1) + ...

@TuxCopter I found a missing (or extra?) parentheses on "from independent sources"
I wish there was a way to link to a find X in page kinda thing
@Calvin'sHobbies How is that relevant to my message... ?
Most annoying thing about wiki is for lack of better terms "redirect decay"... when someone decides to redirect X to an article Y that supposedly includes it, then Y deletes the content that included it
@TuxCopter Just making conversation :/
22:35
Yes but you replied to my message and your message is not relevant to the message you replied to
Well, I found that page while trying to complete the CMC
Technically, finding something completely unrelated to what you're looking for is at the heart of wikipedia
Starting at a page like Intel 8080 and ending on South Africa is common lel
@Calvin'sHobbies My browser is dying
22:41
Does anyone remember the challenge about finding the index of the highest running mean of length n in a binary list? If the input was: 1001111001` and 4 then the output should be 3, because the starting index of the 4-digit sequence with the highest mean starts at position 3 (zero-indexed).
I can't remember how it was phrased :/
There are 73 challenges tagged
67 of those are golf
@trichoplax Thanks!.
Don't know why I didn't think of searching for it using tags...
@trichoplax, at least now I can call this safe :)
Intel machine code library is huge.
@StewieGriffin Did you change your mind?
22:48
I actually came this close to adding popcnt to Jelly.
But it's still longer than the Jelly solution -.-
Crap, wrong challenge :(
There's also that, yes.
@trichoplax Yes... :(
The x86 instruction set refers to the set of instructions that x86-compatible microprocessors support. The instructions are usually part of an executable program, often stored as a computer file and executed on the processor. The x86 instruction set has been extended several times, introducing wider registers and datatypes as well as new functionality. == x86 integer instructions == This is the full 8086/8088 instruction set. Most if not all of these instructions are available in 32-bit mode; they just operate on 32-bit registers (eax, ebx, etc.) and values instead of their 16-bit (ax, bx, etc...

« first day (2127 days earlier)      last day (3011 days later) »