/dpt/ - Daily Programming Thread

Erlang is the future
https://www.youtube.com/watch?v=rRbY3TMUcgQ

>>56932624
erlang has been the future for decades

>>56932624
>what are you working on
>Erlang is the future
okay.

>>56932624
>Had to learn erlang for my distributed systems class
>Literally skipped that entire part of it and just bothered with the calculus
lol

Is erlang actually fun to use?

>>56932637
i've never used it but everything i've used written in it crashes

I'm thinking about making a 4chan clone (and release it both as a forum website and its software), but adding some nice features (higher, configurable expiry limit, with pages showing 100-300 posts each), and replacing other ones (external pics/videos, to make it waste less bandwidth... use pic hosting websites and their APIs)

do you think it'd be worth and would attract some users it if it was more user-friendly than 4chan, or the network effect is simply too hard to defeat?

>>56932637
learn Go instead

>>56932661
No

>>56932672
Why would I learn fucking Go nigga?
>Immature
>Unproven
>Shit
????

anyone here know Android?
I'm trying to get the screen resolution. using

Display display = getWindowManager().getDefaultDisplay();
Point size = new Point();
display.getSize(size);
int width = size.x;
int height = size.y;

gives me the correct height value but the incorrect width value, it seems to be showing me (screen width - navigation bar width) even though I've set it to hide the nav bar.

anyone know how to get the full screen resolution?

>>56932661
No
Instead make a decentralised 4chan

>>56932661
>or the network effect is simply too hard to defeat?
This.

Making any kind of social network site is impossible.

>>56932661
Make a chan using Blockchain and call it blockchan

>>56932735
so when someone posts child porn or something copyrighted or whatever every single node gets taken down permanently?

sounds like a fun game to play for about 10 minutes.

File: 14519707_727387330733859_8829537541648926442_n.png (119KB, 400x497px) Image search: [Google]

119KB, 400x497px

XD

>>56932742
You know, that's a good point... What DOES happen if you post illegal material to a blockchain?

>>56932752
int people[3] = { 100, 200, 300 };

>>56932752
a,b,c = 100,200,300

>>56932702
>>Immature
>>Unproven
>>Shit
look at this:
>>56932162
>>56932404
> https://www.youtube.com/playlist?list=PLE7tQUdRKcybLShxegjn0xyTTDJeYwEkI
>VLC
>text editors
and rust keeps breaking its own std lib
*this* is what I'd call immature

meanwhile, Go software made 3+ years ago will work just fine, and LOTS of companies run stable Go software for MONTHS and even YEARS.
I myself have a server written in Go that has been running for months at a time, only stopped to add minor stuff and to keep the compiled code updated (to avoid security bugs, etc)

Go is awesome. try it: https://tour.golang.org

>>56932755
anyone hosting it gets taken down for serving illegal content. god help you if you're letting people store arbitrary data on a blockchain.

your best case scenario is to take down your instance of it and try to delete it (the way you would try to rewrite history in a git repo if someone pushed child porn to it). so this isn't an unprecedented problem, by any means. it's just clearly not a good idea to host a website on bitbucket or github by asking people to push their own shit to it (partly for this reason).

File: launcher.jpg (913KB, 3320x1920px) Image search: [Google]

913KB, 3320x1920px

Just an idea I had.
Yay or nay?
>This is a functional concept, design is not important at the moment

>>56932811
>god help you if you're letting people store arbitrary data on a blockchain.
https://www.ethereum.org/

They're one of the biggest blockchains and they allow this.

File: df0c19d025f308b739a43882f951401d.png (128KB, 957x782px)

128KB, 957x782px

OS Anon here, I just made a github repo if anyone wants to check it out. i'll try and stop shitting up threads until I get something major done like the filesystem.
https://github.com/Penot1/FutureOS

File: 8BEA8D90-B181-48F6-830D-3E5F50B488A5.jpg (227KB, 1365x2048px) Image search: [Google]

227KB, 1365x2048px

>>56932599
what's her favorite language?

>>56932842
Risp.

>>56932828
do you need me to post links to all the stories of attacks on the ethereum network?

>>56932813
lookg good imo, but I haven't had a >5in phone, so can't comment on how it would function. for smaller phones... it'd be difficult to use, I think

>>56932842
https://github.com/danielwaterworth/plastic

>>56932813
>He, thanks Pajeet for the idea

>>56932696
>>56932726
oh, well...

>>56932717
>>56932735
no idea how that would work

>>56932852
>do you need me to post links to all the stories of attacks on the ethereum network?
I've heard of some of them, but they were unrelated to what we're talking about. If you know of any cases of people using it to host illegal content then yes I would very much like those articles.

>>56932599
I am trying to make a Xamarin.Forms (C#) app that opens up a page when an element of a ListView is selected. A page opens when the ListView element is selected, but it's just a blank instead of the page linked to in the ListView. Where is my mistake? Is it in the DetailPage class?

Search.cs

    public class pageList
    {
        public pageList(string pageTitle, Type fileName)
        {
            this.pageTitle = pageTitle;
            this.fileName = fileName;
        }

        public string pageTitle { private set; get; }

        public Type fileName { private set; get; }

    };

  class DetailPage : ContentPage
    {
        public DetailPage(pageList page_list)
        {
            this.page_list = page_list;
        }       

        public pageList page_list { private set; get; }

    }

  ObservableCollection<pageList> pageListings = new ObservableCollection<pageList>
{
new pageList("Anaphylaxis", typeof(Anaphylaxis)),
new pageList("Alcoholism", typeof(Alcoholism)),
new pageList("Anorexia Nervosa", typeof(AnorexiaNervosa)),
};

           listView.ItemTapped += async (sender, args) =>
             {

                 var itemSelected = args.Item as pageList;
                 if (itemSelected != null)
                 {
                     DetailPage detailPage = new DetailPage(itemSelected);
                     await Navigation.PushAsync(detailPage, true);                                     

                 }
                                  
             };

i passed with a c on a java programming methods class in spring, did nothing over the summer, and im now doing java data structures in the fall. Idk what the hell a node is and i feel stupid and i dont know if i will pass the class. what do i do?

Why are libraries in high level languages usually low quality?

>>56932884
Kill yourself.
Not that a node has anything to do with your java programming methods class - it's basic English, Pajeet.

File: screenshot1_small.png (127KB, 360x640px) Image search: [Google]

127KB, 360x640px

>>56932856
Shouldn't be that difficult, I mean it's a simple DrawerLayout menu, similar to many existing apps.

>>56932884

Consider another subject, not computer science.

>>56932871
the point of the critique is that arbitrary permission to add data to the blockchain is a serious risk. the ethereum attacks essentially hinge on the ability to put arbitrary content on the blockchain (transaction spam is the textbook example).

ethereum's solution to the earlier attack (the replay attack) was to have a hard fork. this is no small feat; there needs to be a vote among all the nodes (read: everyone), and anyone who disagrees is fundamentally disagreeing with the continued operation of the majority. it's hard to think of an analogy, but it might be right to say that this would be like a forum splitting *every time* the mods decide to delete a bad user's posts. maybe the split won't be so bad - maybe everyone will agree that a user's posts should be purged and the blockchain rewritten - but this is so far outside of the design of blockchain that you really, really, really should just use a conventional RDBMS and maybe give people the ability to export the DB if you really want.

the ethos of blockchain is okay (i say "okay" very carefully, because i'm a few steps away from calling it dangerous and/or toxic). the implementation of blockchain on very social systems (like user forums, for instance) where any number of outcomes can end up documented on the ledger is absolutely nutso.

>>56932901
im not indian, im some puerto rican american latin hybrid, and i havent been focusing on my programming lately. maybe because im lazy or i just dont have a drive, ive been learning alot of music by myself than learning cis in class

also my teacher sucks, all the cis teachers in my school are loud, incoherent indians with at least master degrees. the one teaching me now somehow had a job with google for 10 years until he retired

>>56932891

People who can't program to save their own lives generally tend to prefer easier to use, higher level languages. They then go on to write libraries for those languages.

File: 2016-10-05-044803_406x24_scrot.png (5KB, 406x24px) Image search: [Google]

5KB, 406x24px

wtf is pic related
I've seen about 5 high paying jobs asking for it

also, I've just seen a entry level c programmer job that pays 13$ an hour but is in an area 3 hours away

what r chances they're tyring to slave me??

I've spent an entire day trying to get openCV installed on my computer. It finally works. I want to die.

>>56933085
Let me guess, windows?

>>56933061

>$13 an hour
Assuming you worked 40 hours a week, 52 weeks a year, you'd make a little over $27k / year. Those are buttfuck wages. Don't take it.

>>56933061
I made more interning as an InfoSec Analyst

>>56933173
but I have no experience at all
im literally some fag on his third semester freshmen year in CC

>>56933061
Wew, internships are paying more than that

I bought an e-book about programming for school without paying attention.
How do I view DRM protected epub files without using windows?

>>56932798
It fucking drives me up the wall that people still defend this bullshit

>>56933220
There's a GPL reader called Calibre available for Linux.

File: twirl-v8.webm (179KB, 640x640px) Image search: [Google]

179KB, 640x640px

I'm working on converting 154 "ass" instragram videos and I've noticed the libvpx-v8 is much better at transcoding mp4 then libvpx-v9.

time ffmpeg -y -i BKSyn9kA4gJ.mp4 -c:v libvpx-vp9 -cpu-used 2 -threads 8 -an twirl-v9.webm
real    3m20.130s
user    5m45.536s
sys    0m0.300s

time ffmpeg -y -i BKSyn9kA4gJ.mp4 -c:v libvpx -cpu-used 2 -threads 8 -an twirl-v8.webm
real    0m37.513s
user    1m1.920s
sys    0m2.272s

Although the file size on vp9 is smaller its takes almost 5x longer. Anyone has any idea why?

File: 12446248_1519540071682775_517013983_n.jpg (38KB, 750x750px) Image search: [Google]

38KB, 750x750px

>>56932599
>who is your waifu and what has she programmed
Taeyeon. Tistory scraper. Scraping pictures of herself.

File: cute anime pic 0514.png (86KB, 379x290px) Image search: [Google]

86KB, 379x290px

Guys
The program is the proof!

>>56932798
The videos make it clear who the demographic of rust is. Immature programmers, bonus points if they liked to use Python our Ruby before rust.

File: Calibre_logo_2.png (394KB, 554x499px) Image search: [Google]

394KB, 554x499px

>>56933265
When I try to view my book using calibre I get the message "This book is DRMed".
Are you talking about using a calibre plug-in to remove the drm protection?

>>56933352
Idk then. Calibre is what I use. I have never used a DRM ebook. I just get the paper textbook so they can't boss me around.

>>56932599
>who is your waifu and what has she programmed
my wife but she can't program

>>56933305
x : ⊥

where's your god now

File: twirl-v8.webm (711KB, 640x640px) Image search: [Google]

711KB, 640x640px

>>56933289
That quality was bothering me

>>56933338
I just can't imagine how they think it's going to work. This isn't Scratch, writing Rust (or anything at systems level) is fucking hard even with muh safety. They'll wash out fast.

why is integration testing such a pain in the ass
have to test some email stuff, which means i need to setup an smtp server and client for my test enviroment for like 30 test cases

>>56933460
Are you by any chance using an impure language with a limited type system?

>>56933460
>Testing
lmoa
this is why you don't become a wagecuck

>>56933508
im not in the mood for this shit

>>56933510
dont you have fizzbuzz to write?

>>56933520
>im not in the mood for this shit
No, you're too busy integration testing.

Anyone else feet get cold at night? Trying to code but my feet are so cold.

>>56933520
>muh testing
l
m
o
a

>>56933556
Maybe that is why traps are better at programming. Their brightly colored socks keep them warm at all times through the day so they can focus.Perhaps you should invest in a pair.

>>56933556
get some thick wool socks

reminder to follow the advice of your haskell bretheren
https://www.youtube.com/watch?v=ojZbFIQSdl8

>>56933556
Fetti was geht?

>>56933579
>CppCon
>C Preprocessor Convention
Wat

>>56933586
plus plus

>>56933579
>using types
>c++
pure cancer

File: memes.jpg (46KB, 695x431px) Image search: [Google]

46KB, 695x431px

>>56933597

File: 2016-10-05-231235_1170x926_scrot.png (165KB, 1170x926px)

165KB, 1170x926px

>>56933592
I'm 100% sure "cpp" means C Preprocessor.

c++ is cancer in the most literal sense. it's growing all kinds of unnatural stuff.

>>56933621
In this case it means C++

>>56933633
But right now they're on the way to adding good features

>>56933644
>In this case it means C++
You can't just redefine phrases willy-nilly.
Why the hell didn't they call it C++Con, or actually make it about the C Preprocessor?

>>56933657
>why didn't they call it C++ Con
the character '+' is not portable

>>56933670
Why are you even using that shitty language then?

>>56933676
Are you a fucking idiot?
Why the fuck don't you see people with in their username on most sites?

Seeing if a larger sample size might help out my Markov chain
>currently scanning: All posts on 4chan
this should produce something a little more robust

>>56933700
>it didn't even show the character
My point fucking proven.
Perfect.

>tfw you do functional programming again after a long break
feels good man
I wish there was a compiled functional language with the speed of C, there would be literally no reason to use anything else.

>>56933586
>>56933621
>>56933657
>>56933676
>>56933710
Pure, unfettered autism

>>56933714
C and the rest of the original imperative crew fucked everything up

>>56933700
>>56933670
CXX is already a standard way of referring to C++.
Why the hell would they be retarded and redefine it as the C preprocessor.

Fuck off Go shills

>>56932882
This is the full file in case anyone cares to read it.

https://gist.github.com/EricGrahamMacEachern/57009becb2081e17ef45a10f4304401e

>>56933714
I think some of C's speed depends on its mutability

>>56933752
Haskell has mutability in the IO and ST monads

File: save-the-princess-in-8-programming-languages.jpg (1MB, 800x2798px) Image search: [Google]

1MB, 800x2798px

/dpt/-chan, daisuki~

>>56933714
https://en.wikipedia.org/wiki/ATS_(programming_language)

>>56933619
Python has types.

>>56933556
Where are you from ?

>>56932884
Node is a powerful and very popular implementation/distribution of Javascript for backend development.

>>56932831
Thank you for using GPLv3

>>56932752
First or Third

>>56932661
>do you think it'd be worth and would attract some users
Lainchan may be interested. see https://lainchan.org/%CE%BB/res/18076.html

>>56932615
As a code artisan, i disagree. A code artisan is a programmer who has crafted his own development tools.

>>56932599
Please use an anime image next time.

File: Fridgebike.jpg (166KB, 1280x529px) Image search: [Google]

166KB, 1280x529px

>>56933782

I liked that picture. Have this one, if you don't already have it.

>>56933577
If that's what it takes..

>>56933556
I like to program from the comfort of my own bed. If my feet get cold, I clearly should be keeping them under the covers.

>>56933752
>>56933768
The question to ask here is: is idiomatic Haskell fast?

>>56933831
java and python are 10000x better and run compiled faster than c

File: arc-text.png (10KB, 1001x465px) Image search: [Google]

10KB, 1001x465px

>>56932599

Hey guys I'm attempting to produce a php image by merging text onto a pot ribbon, the text is custom fed through the $_get['name'] bit in the url.

I've been using the GD library to do all this but I don't know how to skew the text towards the top of the ribbon instead of just going around a arc.

Image and code attached.

http://codepad.org/9YgoDNah

>>56933782
>You give up and go to stackoverflow to have jon skeet rescue the princess for you
okay i fucking laughed

>>56933831
i dont get it, how can you love C but dislike C++
doesn't make any sense

>>56933944
Are you a liberal?

>>56933893
Exactly some of the time.

Just ignore your memory usage.

I'm trying to learn Node and it doesn't make any goddamn sense. Why did they keep Javascript callback syntax if they're all asynchronous and run on an event queue anyway?

>>56933893
Haskell can be quite fast
http://benchmarksgame.alioth.debian.org/u64q/compare.php?lang=ghc&lang2=gcc

>>56933977
wtf am I reading

>>56933905
>Java
In most circumstances, C outperforms Java. Java is only able to perform better on a few tight loops that are able to be optimized by the JIT using recompilation.

>Python
Not even close.

>>56933944

I don't hate C++. I actually rather like it. But you have to admit, it is a clusterfuck.

>>56934011
It's basically a bunch of memey science problems ("calculate the spectral norm of an infinite matrix" etc.) and then people can submit the most optimised solution to that problem for their language. I think it's a bit biased towards number-crunching but it's still kind of interesting.

>have a large programming project one month after half the class started programming for the first time
why
its like im paying tuition to teach others how to code

>>56933944
>how can you love C but dislike C++
C is actually designed reasonably well. It's a language that has a purpose.
C++ is just a mishmash of random shit.

>>56933831
What lang is the fastest around Nürburgring?

File: anal bead.png (17KB, 655x251px) Image search: [Google]

17KB, 655x251px

>>56933782
T-thanks Jon Skeet

>>56934055
German

Is it worth getting into Functional Reactive Programming or "just" ReactiveX?

>>56934071
it's a meme isn't it?

>>56934046
Some professors literally do not understand that they potentially have tons of first-time programmers in their class, particularly cross-major students who have the option of taking one or two programming classes.

On the bright side, teaching someone how to do something is the easiest way to fully understand it yourself.

>>56933972
>Node.js
I found your problem.
Are you using promises? It makes things slightly more bearable.

>>56934114
good post

>>56934092
I dunno, that's why I'm asking. The Netflix devs seem to like it. Of course, the Netflix people have different problems than the rest of the world

>>56934142
You said it yourself; they have different problems.

Using things that Netflix, Google, and Facebook use is not necessarily a good idea, because they have environments of a scale that basically none of us will ever touch in our lives.

>>56934174
That does not mean that you cannot benefit from learning another approach

>>56934114
Thanks anon, that might help

I still can't understand Node at all though, it feels like two opposite ideas jammed together

Getting tired of all these new high level languages. Do I really have to write my own incremental improvement of C?

>>56934208
Haskell is the next step up from C

>>56934189
No one said that.

File: internet.jpg (415KB, 1400x981px) Image search: [Google]

415KB, 1400x981px

>>56934208
No

http://c2lang.org/
http://ark-lang.org/

>>56934046
I was in a similar situation in an OOP class.

We did the design for a project in larger groups, and then split into smaller groups for the implementation.
I lost patience with my group and ended up splitting off, doing the whole implementation myself. I was still the first one to get done in the whole class, and the only implementation to get accepted on the first submission.

This... probably doesn't speak very well for our design.

>>56934285
>http://c2lang.org/
>For array types, C2 introduces a new operator, namely elemsof().
"no"
>http://book.ark-lang.org/
"no"

I tried to do merge sort based on half remembered functional programming about a year ago, but I'm pretty sure I didn't make merge sort... I think you're supposed to perform it like a tree structure, merging each 1 into lists of 2, then 2 into 4, to do it in log(n) time, but I'm thinking this is polynomial time.

mergesort([A | [B | R]]) ->
    if A > B ->
            sort([B, A], mergesort(R));
        true ->
            sort([A, B], mergesort(R))
    end;
mergesort([]) ->
    [].

sort([A | R1], [B | R2]) ->
    if A > B ->
            [B | sort([A | R1], R2)];
        true ->
            [A | sort(R1, [B | R2])]
    end;
sort([], B) ->
    B;
sort(A, []) ->
    A.

>>56933579
>tfw want to watch this
>https://www.youtube.com/watch?v=fMvO0Mcq894
>tfw the guy speaks as if he has downs and I don't understand anything

>>56932705
You gotta hardcode that shit. Google navbar widths on different versions.

>>56934357
Jesus christ

>>56934357
>You gotta hardcode that shit.
I don't even know what you're talking about, but i'm pretty sure you're doin' it wrong.

>>56934342
He's a pretty good presenter desu fampai.

File: spede aesborgor.jpg (20KB, 600x660px)

20KB, 600x660px

>>56933556
i wear socks 99,99% of the time except when summer, swimming, and sauna is this normal

File: GiXfONL.jpg (664KB, 2592x1936px) Image search: [Google]

664KB, 2592x1936px

Anon! Why aren't you programming a bot for reddit ?

https://www.reddit.com/r/OnePiece/comments/55yr6y/one_piece_chapter_842/d8eue1c

>>56934331
Ok, I think this is it:

mergesort([A | [B | R]]) ->
    sort(sort([A], [B]), mergesort(R));
mergesort([A]) ->
    [A];

>>56934655
Same.

I'm rarely without socks, and shoes (if you consider house-slippers shoes).

>>56934739
>Why aren't you making something useless and annoying?

Also, what country are you from?

Only communists put a space before the punctuation at the end of a sentence.

>>56932599
Hi /dpt/! Writing a tile-game in python,
how would I go about cloning an instance of a class overwriting an existing class; removing values that are not in the new instance? Any pointers would be much appreciated

>>56934777
>Also, what country are you from?
France.

>>56934820
>being french
lol

>>56934757
Ok FUCK
I learned about the split and length functions.

>clone a github repo
>configure and make
>gcc spits out errors, undeclared identifiers, etc.
>decide to fix it
>open the h file where the stuff is defined
>it's empty, 0 bytes
>check the file on github's website
>it's 12 kilobytes

I had to re-clone the repo THREE times before every file contained something and weren't just empty. Shouldn't it, you know, NOTICE these kinds of errors instead of just continuing like nothing happened? It is a version control tool, so producing empty files is unacceptable.

Did I hit a Github bug, did git derp or what the hell went wrong? git's already arduous to use and then it doesn't even clone a repo correctly. I miss Mercurial, but everyone uses/expects a git repo these days and won't even look at anything else.

>>56935175
Download the zip and create a new repo, like a normal person.

>>56932849
i raffed

Is there any way to do this sort of split without the auxiliary function?

mysplitA(0, L, F) ->
    {F, L};
mysplitA(N, [H | T], F) ->
    mysplitA(N-1, T, [H|F]).
mysplit(N, L) ->
    mysplitA(N, L, []).

>>56935256
Full mergesort(for realz)

mylength([]) ->
    0;
mylength([_ | T]) ->
    1 + mylength(T).

mysplitA(0, L, F) ->
    {F, L};
mysplitA(N, [H | T], F) ->
    mysplitA(N-1, T, [H|F]).
mysplit(N, L) ->
    mysplitA(N, L, []).

mergesort([]) ->
    [];
mergesort([A]) ->
    [A];
mergesort(L) ->
    {FirstHalf, SecondHalf} = mysplit(mylength(L) div 2, L),
    sort(mergesort(FirstHalf), mergesort(SecondHalf)).

sort([A | R1], [B | R2]) ->
    if A > B ->
            [B | sort([A | R1], R2)];
        true ->
            [A | sort(R1, [B | R2])]
    end;
sort([], B) ->
    B;
sort(A, []) ->
    A.

>>56935190
I could have done that. Next time I might.

But the issue is git's inability to clone a repo. No error messages, no warnings, nothing. Just empty files. This is THE version control tool? Isn't it supposed to be reliable?

>>56935318
This is the reality of open sores, anon. Enjoy :^).

>>56934655
>sauna
I wish I could experience this.

I'm in need of a numpy array that contains 30x17 roughly same-sized rectangular areas that are numbered. For example:

[[  0,   0,   0,  17,  17,  17,  34,  34,  34, ... , 476, 476, 476, 493, 493, 493],
 [  0,   0,   0,  17,  17,  17,  34,  34,  34, ... , 476, 476, 476, 493, 493, 493],
 [  0,   0,   0,  17,  17,  17,  34,  34,  34, ... , 476, 476, 476, 493, 493, 493],
 [  1,   1,   1,  18,  18,  18,  35,  35,  35, ... , 477, 477, 477, 494, 494, 494],
 [  1,   1,   1,  18,  18,  18,  35,  35,  35, ... , 477, 477, 477, 494, 494, 494],
 [  1,   1,   1,  18,  18,  18,  35,  35,  35, ... , 477, 477, 477, 494, 494, 494],
 [  2,   2,   2,  19,  19,  19,  36,  36,  36, ... , 478, 478, 478, 495, 495, 495],
 [  2,   2,   2,  19,  19,  19,  36,  36,  36, ... , 478, 478, 478, 495, 495, 495],
 [  2,   2,   2,  19,  19,  19,  36,  36,  36, ... , 478, 478, 478, 495, 495, 495],
 
                                               ...
 
 [ 15,  15,  15,  32,  32,  32,  49,  49,  49, ... , 491, 491, 491, 508, 508, 508],
 [ 15,  15,  15,  32,  32,  32,  49,  49,  49, ... , 491, 491, 491, 508, 508, 508],
 [ 15,  15,  15,  32,  32,  32,  49,  49,  49, ... , 491, 491, 491, 508, 508, 508],
 [ 16,  16,  16,  33,  33,  33,  50,  50,  50, ... , 492, 492, 492, 509, 509, 509],
 [ 16,  16,  16,  33,  33,  33,  50,  50,  50, ... , 492, 492, 492, 509, 509, 509],
 [ 16,  16,  16,  33,  33,  33,  50,  50,  50, ... , 492, 492, 492, 509, 509, 509]]

Is there a way to create this that's easier than looping over all elements and calculating their values? The dimensions of the array can vary, but the number of patches (30x17) always stays the same. Right now I'm doing this:

segs = np.zeros((height, width))
for x in range(width):
    for y in range(height):
        segs[y, x] = np.uint32((17*y)/height + 17*((30*x)/width))

It works, but it's slow as fuck.

>>56935516
yes, write it in c

>>56935541
And if I'm stuck with Python because the project manager says so?

>>56935516
Maybe python has some memcopy functionality? Hopefully a lower level API within python would do better than whatever makes you complain about the speed in this absolutely trivial operation. How many cycles per integer is this taking?

>>56935574
le python of interpreted language
In order to write fast python you need to be a master of one liners.

>>56935614
Did you not read what I wrote?

>>56935516
maybe this
https://docs.scipy.org/doc/numpy/reference/generated/numpy.fromfunction.html#numpy.fromfunction

>>56935614
I know, I was hoping someone knew one.
I guess I could define an array where each value exists exactly once, scale that with cv2.resize and round to get my integers back.

>>56935631
That's cool, might come in handy for some of the more complex arrays.

File: 1379594532738.jpg (123KB, 706x674px) Image search: [Google]

123KB, 706x674px

>>56935628
nice autism
suck my cock dude

How many of you are students of cs? Or have any degree in cs?

I am on my way to get CS degree, on faculty of science and mathematics, but is it worth it?
Studies were 3 years long, if i want masters i need 2 more years.

>>56935654
Not sure what you're trying to say here. Were you shitposting or are you saying I am? Either way. If python doesn't JIT or does anything to get rid of interpretation during compilation it'd still help. Since every row is repeated for the number of columns you'd just copy the first row columns number of time.
Certainly faster than constructing it every time. A memcopy would certainly not be implemented through pure python functions. Kinda loses its point then.

>>56935671
No
I have a degree in CS, from a world top 100 uni, not great, but not complete shit either. I'm now an unemployed, social retard, who shitposts on /g/ and /s4s/ while occasionally working on small programming projects.

Just stop already before it's too late :(

>>56932831
What language is this?

>>56935671
I was gonna do CS but I realised I was more interested in developing software. So I quit and started developing software. Took a fair few loose courses to help that. Not sure if that was an efficient use of time or not.

>>56935722
>Since every row is repeated for the number of columns you'd just copy the first row columns number of time.
Thing is, the number of columns that are the same might vary depending on whether width is divisible by 30 or not (hence me saying roughly the same size).

>>56935722
>during compilation
Meant at run time obviously.

>>56935671
http://www.businessinsider.com/google-hiring-non-graduates-2013-6?IR=T

http://www.wsj.com/articles/why-im-not-looking-to-hire-computer-science-majors-1440804753

>>56935671
I did MIS at a business school and am now primarily a developer.

Many companies know they can teach someone Python, C#, Julia/R/whatever statistics language in a few months.

If you do CS purely to be a programmer, you're doing it wrong. Anyone can be a programmer.

>>56935758
I'm not fully sure what this array looks like still but surely you could handle the off-cases the slow way?

I'm just speculating around the usefulness of a memcopy function either way. I don't know python.

>>56935732
x86 assembly

>>56935516
Try reserving the entire array before you do your insertions.

I'm about to finish my degree but I still suck at programming.

>>56935834
Was it aimed at teaching you programming?

>>56935834
CS degrees really need to come with a disclaimer.

>>56935787
Maybe if I tell you what it's for you'll understand better.
I have an image that might be one of:
> 1920 x 1080 (full HD)
> 960 x 540 (full HD picture scaled to half dimensions)
> 960-b x 540-b (full HD picture with b pixels wide rim removed and downscaled)
What I need is a mask that numbers segments of the image; I want 510 segments (hence 17 x 30 since that's close to 16:9 aspect ratio).
While 1920 and 960 are divisible by 30, 960-b isn't unless b = k*30. Usually it's 3.
Checking for those odd cases would be relatively slow as well; slightly less so than calculating each segment separately of course.

>>56935814
What do you think

segs = np.zeros((height, width))

does?

>>56935808
>x86 assembly
My nigger. Why though?

>>56935879
why not just use >>56935631?
I bet it would be shit loads faster.

File: cs.jpg (1MB, 1904x4706px) Image search: [Google]

1MB, 1904x4706px

>>56935671
When you see that someone like Ruby has a degree in CS but doesn't even know what passing by reference does mean, it really makes you think.

>>56935671
I wish I had my degree in CS. Or somewhere anywhere in STEM for that matter. Really hard to even network or find someone to give you some personalized pointers if your degree doesn't jive, and I'm not keen on these crappy PR/"communications"/admin assistant jobs being thrown at me on LinkedIn. Trying to fix that, though, now that I have some kind of stable income.

>>56935909
>pic
Haskell is making programming great again

>>56935952
It does represent a major paradigm shift into thinking about programs algebraically. That said, it's also a hot topic for nu-males who want to "prove" that they're super hipster coders because they can use IO and do notation to do imperative programming like usual but not REALLY guise.

>>56935725
>Just stop already before it's too late :(
lol, i have 2 more exams left till degree, that is not option
>>56935742


Problem is if you want to get into better company they all want to see your degree. At least that how it works in my country.


>>56935784
Well, 1st year we have 4 mathematic and 4 programming classes
2nd year we have 2 mathematic, 4 programming, computers architecture and computer systems classes.
3rd year we have Computer networks(working with cisco), software engineering, web dev(php/mysql) advanced english and some shitty classes

>>56935909
We start with c, then move to java and c++. Later do some c# and that is all.

>>56935943
what i got out of school that was helpful for programming was knowing people who know people at companies
i've never taken a cs class but i heard they were pretty bad
anyway, sorry, that's not very helpful. do you live in a big city? there are ways of at least meeting people that aren't totally sleezy

>>56936021
apparently i didn't attend spelling class either

File: 1469491065420.jpg (34KB, 710x720px) Image search: [Google]

34KB, 710x720px

>tfw everyone is shitting on the course you regret not doing
pls

>>56935894
Yeah, that's better.
God I love one-liners.

segs = np.fromfunction(lambda y, x: 17*y/height + 17*(30*x/width), (height, width), dtype = np.uint16)

File: 1467212015047.jpg (40KB, 722x349px) Image search: [Google]

40KB, 722x349px

>>56932599
>water you working on?
Nothing. Started with league and weed again.
How do I realize I need to get my shit together?

>>56936067
>wojackposter
>stoner
>vfag
>literally blaming reality

It's all over

>>56936067
Join your local programming community group and contribute to an open source project. That or ask your doc for some Adderall

I don't have a CS degree. What do?

File: 12-most-influential-programmers.jpg (146KB, 1024x768px) Image search: [Google]

146KB, 1024x768px

>>56936103
https://github.com/jwasham/google-interview-university

have fun

>>56936126
when is he gonna make it

>>56936089
>wojackposter
wat?

>literally blaming reality
I'm not blaming it on reality lol.

>>56936094
Thanks. Can't get Adderall here (eu fgt).
I think I'm too shit for contributing to a project tho. Only learned python for a bit and did html/css/js course on codecademy.
Also, does /g/ have IRC channel?

>>56936126
>N-ary (K-ary, M-ary)

File: Screenshot from 2016-10-05 09-54-55.png (531KB, 1920x1080px) Image search: [Google]

531KB, 1920x1080px

Can someone help tell me what I'm doing wrong? I'm new to GTK and I'm just making a test program.

>>56935471
It's just sitting around naked and sweating in a very hot and humid room.

>>56935879
This is an extremely weird problem if I'm getting it right. So I have a few questions.:
1. Did you understand the idea I had behind memory copying? You have two independent terms. Assigning a large number of static elements is far faster than assigning a large number of dynamic (calcilated) elements to an array. If you just concern yourself with the first '(17*y)/height' term. Calculate that for every y (making an X-wide vector of it and inserting it for every y row) Then calculate the 17*(30*x)/width for every x and adding a y-high vector to every column (add, not replace/assign). It should allow for much wider operations.
2. What's the point in numbering the segments of the image based on the indices? You're not really getting any additional information out of it as far as I can see.
3. Why are you using a math library for image processing? (even if it's not image processing it seems applicable here). Is that just a python thing?
4. Have you tried swapping the way you iterate the array to be width first? I'm not sure what the underlying representation for python arrays are but that's a general thing you do to avoid cache misses. (you can notice it even on 8int wide arrays in C, it's nuts).
>>56936052
Is that good enough?
Man I'd hate to have wasted a bunch of time.

Javascript question:

Below is a function that filters duplicate entries out of an array. If the object 'seen' has the property we're currently iterating over (i.e. it has been iterated over or 'seen' before by the filter loop) we return false. Cool, it doesn't get added to the array returned by the filter method.

What I don't understand is the final operand in the ternary operator. how does (seen[item] = true) resolve to true within the filter? Does this expression coerce to true - i.e. the assignment is truthy?

function uniq(a) {
    var seen = {};
    return a.filter(function(item) {
        return seen.hasOwnProperty(item) ? false : (seen[item] = true);
    });
}

>>56936154
You're not supplying enough arguments to the indicated function.

>>56936154
Does the function only take two arguments?

>>56936183
I don't know about javascript but in C derivatives
(x = e)
returns e, e.g.

y = (x = e)

>>56936126
>only 3 respectable programmers in picture
Which 3?

>>56936204
it's the same here
javascript is basically c

>>56936183
Is it that (seen[item] = true) resolves as an expression to just seen[item], which is a truthy statement?

>>56936184
Oh I thought I was. What else do I need to add?

>>56936214
>>56936204
Cool, that's what I suspected. Thanks for the help dudes

>>56936214
>javascript is basically C
Hilarious anon. /wdg/ probably knows more JS than dpt though.

>>56936021
>what i got out of school that was helpful for programming was knowing people who know people at companies
I never got to meet anyone like that, unfortunately. Closest person I know is the best friend of a friend I know from online video games who works at Microsoft and that's such a long shot. One guy I know became a webdev but that stuff doesn't hold my attention.

>i've never taken a cs class but i heard they were pretty bad
I've taken two CS classes, one at my community college and one at my university. I liked them both a lot. I graduated, got some odd but ok fresh-out full-time job, and I'm taking what's called CS1 (CSC 120 - someone's bound to recognize the course code) at that same community college right now and the professor's great but the students are so bad. Some are bad scholastically and others have bad attitudes, mostly "heh i already know everything about java, why am i even here" kind of people.

>do you live in a big city? there are ways of at least meeting people that aren't totally sleezy
Yeah, I live in THE big city (NYC) so I've been trying to find something worthwhile. I went on some website called "meetup.com" but that's got all sorts of dead groups, webdev, and weird "girls only" groups.

>>56936154
>Atom
Found the problem.

>>56936231
>What else do I need to add?
How about you read the fucking documentation?

>>56936265
>getting this triggered over someone else's programming

wew lad

>>56936143
>Also, does /g/ have IRC channel?
No, but there a lot of programming related channels on freenode.

>>56936277
>no /g/ irc channel
Whatever you say anon.

>>56936170
1. Yes, and it's very common in other places in my program as well. I try to do matrix / vector operations when possible, rather than iterating over single elements. It didn't occur to me that I could define columns and then copy them.
2. The idea is to segment the image, calculate characteristic values for each segment (in my case median, average and standard deviation for the pixels in HSV color space) and then train a support vector machine with groundtruth data so it can recognize e.g. what part of the image is sky and what isn't. This rectangular segmentation is obviously trivial; I use it as a reference point. The actual program uses more sophisticated segmentation algorithms that I want to compare.
3. I'm comfortable with numpy and creating this segmentation mask is a simple matrix operation. The shape of the segments in this case is totally independent of the source image so it's pointless to use an image processing library.
4. I always assumed numpy arrays are stored in vertical order first but I might be completely wrong, now that you mention it.

1 5
2 6
3 7
4 8

> Is that good enough?
Yes, the fromfunction thing is perfect. Thank you.
I'll post some screenshots in a sec.

>>56936143
/wdg/ has a discord

>>56935784
>Julia
You can never teach my waifu
https://www.youtube.com/embed/ftQfpAeyxPo

>>56936265
This.

Use a goddamn IDE with auto-complete/IntelliSense if you're not going to RTFM.

It's a basic bitch error, and anyone should see that they're simply calling the function wrong.

>>56936265
Sorry

>>56936208
Personally, Bjarne and the middle four are good in my book.

>>56936341
Atom does have autocomplete though.

>>56936289
Well what is it? And is there specifically a /dpt/ irc channel?

>>56936143
#haskell on freenode

>>56936358
Then that anon needs to install the necessary add-ons/plugins.

Honestly, an IDE is basically text editor + fucktons of addons and plugins, so the lines are blurred between them anyway.

>>56936373
he is probably referring to g/scip which is mostly populated by BOTs and BNCs and has nothing to do with /dpt/ at all.

IRC = Internet Cancer

>>56936346
>bjarne
Satan himself. KYS

>>56936126
"no"

File: Screenshot from 2016-10-05 17-14-47.png (1MB, 974x1133px) Image search: [Google]

1MB, 974x1133px

1/4
Top image is one of the screenshots, bottom is the rectangular segmentation mask. The colors are the median values of the pixel values of each segment in HSV colorspace (interpreted as RGB, hence the false color appearance). Greenish segments correspond to desaturated areas (i.e. there are mostly grey / white / black pixels in the segment), orange / magenta corresponds to saturated light green with some red (i.e. grass).
The crosses mark pixels that I annotated by hand (red = background, cyan = player, white = field).

File: Screenshot from 2016-10-05 17-16-15.png (661KB, 975x1127px) Image search: [Google]

661KB, 975x1127px

2/4
This shows a screenshot from a video that wasn't used for training, and the result of applying the trained SVM to classify the segments. In the top picture, segments that are part of the background (or players) are shaded. In the bottom picture, those segments are black while field segments are white.

Should I hide/private/delete my old repositories that nowadays make me cringe?

File: Screenshot from 2016-10-05 17-27-29.png (1MB, 974x1125px) Image search: [Google]

1MB, 974x1125px

3/4
Here's the same image as in post 1/4, except the segmentation algorithm is different (skimage.segmentation.slic). Note how small details are more noticeable (players and field markings in particular).

I'm still not sure why red players aren't really picked up by the segmentation algorithm. Red and black have the same hue (though different saturation and value), but the hues of red and green are 120 degrees apart.

Calculating the segmentation masks with SLIC is about as fast as creating a rectangular mask (right now; if implemented in C the rect mask should be much faster).

File: 1473791634486.jpg (142KB, 1439x1374px) Image search: [Google]

142KB, 1439x1374px

>Week 6 or so of working a Java lab
>one kid STILL doesn't know how to input a number in Java
This job hurts sometimes my dudes

>>56936620
No. Leave it up. It's weak to hide it and looking like you haven't improved is a bad thing

File: Screenshot from 2016-10-05 17-36-47.png (690KB, 976x1127px) Image search: [Google]

690KB, 976x1127px

4/4
Here's what happens when you train the SVM based on SLIC segmentation and apply it to SLIC-segmented test image.

When applying the SVM to known data and comparing that to the groundtruth, I get over 99.99% accuracy with the SLIC segmentation; with rectangles it's something like 97%.

>>56936655
Hallo Junge, was ist das?

>>56936640
>Semester 6 of university
>Some half-functioning 25yos STILL doesn't know how to input * in *

>>56936655

What's the use case for this?

>>56936738
Automatic analysis of football games.

>>56936655
what acronyms are an SVM and a SLIC

>>56936655
is it doing this in real-time?

https://www.youtube.com/watch?v=zBkNBP00wJE

>>56936783
>C++
>C ++
>C
>++
What did they mean by this?

>>56936811
Postfix ++ adds more stuff but returns the same value.

>>56936765

I'm not seeing how it provides real analysis, aside from hot-spots on the field where players spend more of their time.

I really, really want to find work but I can't think of a single project to put on a portfolio. Should I just roll the programming challenges image?

File: no.jpg (186KB, 816x488px) Image search: [Google]

186KB, 816x488px

>>56936620
Hide. Every time i published something on dpt, i got ostracized/bullied.

>>56936872
It's best to contribute to already popular project. Do this so long until nobody else understands the code anymore because you've written so much of it. Then they'll have to name you maintainer and you're set.

File: 1462117744958.png (260KB, 563x542px) Image search: [Google]

260KB, 563x542px

>>56936892
>Do this so long until nobody else understands the code anymore because you've written so much of it. Then they'll have to name you maintainer and you're set.

File: 1466357136698.jpg (213KB, 853x910px)

213KB, 853x910px

>>56936873
I forgive your bad programming anon-san

>>56936872
pretend to be a girl

>>56936873
What is this manga? Looks sort of like nichijou
?

>>56936775
SVM:
https://en.wikipedia.org/wiki/Support_vector_machine
SLIC: Simple Linear Iterative Clustering
http://www.pyimagesearch.com/2014/07/28/a-slic-superpixel-tutorial-using-python/

>>56936778
Nope, I'm writing this in Python remember. Training the SVM on 24 images takes a couple of minutes and classifying 18 test images from different videos takes another few minutes. I annotated around 22,000 pixels in 5000 segments in those 24 training images; loading the ground truth array alone takes a good 30 seconds. Thankfully that's not a problem; you train the SVM once and then classify whatever. The classification is what needs to be optimized (and implemented in C) later, but for now I'm only doing research and long computation times aren't an issue.

>>56936865
It's part of a much much larger project. The real analysis is based on player positions but to find those you need to know where the camera is looking, which in our case is found by a) matching field lines with a model and b) tracing camera movement over a couple of frames. The current matching algorithm is slow and relies on reducing the possible area of the image where field lines could be (hence field masks), plus the masks can later be used to determine likely player positions.

>>56936872
Write a generic useful thing. Like a server application that is just a networked database manager. Try to make it easy to configure for a wide range of situations.
I'm sure you can find similar projects but for other areas if you think hard enough. If you come up with something you're interested in doing then do it.

>>56936937
ah neat
i'm doing a writeup on clustering right now. nothing this fancy

>>56935175
are you doing this on windows? from what I've read, windows and git don't mix

File: transition.png (946KB, 752x605px) Image search: [Google]

946KB, 752x605px

In real world (I mean in job where code quality matters)

Do you use using namespace std
or each time
std::

?

>>56936973
>windows and git don't mix
they mix just fine, this is nonsense.

anons adventures with git and scrum 3: wreckening

As you might remember, yesterday I reverted direct pushes to master and develop. I also created a feature branch containing the libraries needed for the project.

When I arrived at school today, our team had fucking created pull requests and then SELF-APROVED THEM, BYPASSING ME AS QUALITY ASSURANCE. THEY PUSHED BACK UP THE SAME SHIT I REVERTED.

On the upside, our teacher has been informed of the clusterfuck that is the current situation, and I've come to realize that the most important thing is that I know how to work the libraries in this, rather than to have a working product by the end of this.

I fucking hate these people though. There is exactly one person who I trust in that group, and ironically, he is a muslim.

>>56936973
>windows and git don't mix
What are you on about?

>>56936979
You would expect the latter but find the first.
Do the latter and don't be that guy.

>>56936294
thanks

>>56936386
can you web-dev in haskell?

>>56936989
>ironically, he is a muslim
Wow if you think that's ironic that's pretty racist and I can't even

Just...wow

>>56936979
Where code quality matters you avoid STD's just like real ones.
But you're asking what it's like in a workplace. I say do a scoped namespace if the std:: gets excessive. But I don't think anyone complains about a lot of scope operator use anymore.

>>56936979
In the real world, I don't use unsafe languages.

>>56937029
>racist
Now that's racist anon. He's clearly opposed the religion. Not 'race'.

>>56936989
> trusting muslims
RIP in pieces

>>56936979
In the real world code quality doesn't matter.

File: tumblr_GJ8CNmZ.png (14KB, 265x63px) Image search: [Google]

14KB, 265x63px

>>56937053
That's what YOU think:
https://youtu.be/eNRVL8tibOo?t=410

>>56936989
Just make sure that everything is clearly communicated with your professor.

Most profs will work with you on things like this as long as you let them know well ahead of time and are reasonable.

>>56937143
That guy is just being facetious/silly, he doesn't actually think I'm doing anything wrong at all.

He's a very cool guy, thank god, so I think things will turn out okay. I've been getting physically almost ill from the lack of sleep that has ensued from this shitstorm.

>>56936987
>>56936994
http://latkin.org/blog/2016/07/20/git-for-windows-accidentally-creates-ntfs-alternate-data-streams/
https://github.com/blog/1938-vulnerability-announced-update-your-git-clients
things like these, because of retarded NTFS

>>56936944
I'm just always a bit worried that my idea has been done a billion times and that I'll end up reinventing the wheel but worse.

>>56936973
No. It's a Debian box.

When creating regexes, are there any differences between "[abc]" and "(a|b|c)"? Thanks.

Racism/Sexism/Bigotry/etc are probably the greatest red herrings in history.

>>56937172
>putting colons in your file names
Either way, this would likely occur on OSX as well, considering HFS+ doesn't allow them either. Hell, NTFS and HFS+ aren't even the only file systems that don't allow colons.

The only issue I see here is the lack of a clear error if you're using an old version of Git. The config option that would alert you to this is now on by default.

Second link was fixed very quickly, and also affected non-NTFS machines, so again, your NTFS hate is unfounded.

>>56937233
Yes.

>>56937208
Oh, and when I finally *did* manage to fully clone the repo, it didn't build. Because the program wants Bison 2.something and Debian gives you 3.0.x. Something in the files Bison outputs have changed enough between the versions that it no longer compiles. I don't know enough Bison to fix the file and even if I did, I'd probably just break it.

I'd have to install Debian 7 to get old enough Bison, but at that point everything *else* breaks or won't install because it's too old.

The program's author is aware of this (judging by the comments on Github), but doesn't have time to fix it.

So everything was fucked up from the beginning anyway. The program exists in source form only, so it has to be compiled.

>>56937208
>>56937311
do some testing:
- clone the original repo locally
- clone the repo in github, then clone it locally
- diff the two directories
if they differ too much, you probably found a bug in github or something
btw, what FS are you using? have you fsck'd the partition lately?

>>56937233
No.

>>56937382
ext4, no errors found.

Can't be bothered to clone anything right now. I wasted too many hours with this already. And I nuked the directories already.

>>56937187
So? Writing something to learn to improve is just fine.
The idea of reinventing the wheel being taboo is horrible. With that mentality we'd never have any developments in the material sciences, we'd still not have quantum physics and overall mankind would be retarded compared to the society that accepts exploration as part of a furthering process.

>>56937515
it shouldn't take more than 5 mins
oh well, whatever. I wanted to know if the retards at github fucked up...

>>56932882
Jason from Stackoverflow helped me.

http://stackoverflow.com/questions/39871258/how-do-i-change-the-value-of-the-detailpage-parameter-to-so-that-it-actually-ope/39879149#39879149

>>56937172
>using Windows
>using OSX
I don't know this problem.

>>56932599
A "Job Search" tool

http://107.170.17.223/justjobs/

>>56937233
Depends on implementation. Probably yes, because you're using a capturing group vs a character set. But if the capturing group is never referenced the implementation may change what the effective expression is.

>>56937632
Thanks, I'm working on a Xamarin app now, and I might have run into this.

>>56934016
how is C++ a clusterfuck? I think the STL is very contained and works well.

>>56937753
>how is C++ a clusterfuck?
The ISO actively refuse to rewrite anything that could break existing code and instead choose to bolt new things on top.

new thread when ? :3

>>56932599
A CUTE !

>>56936979
you find std:: everywhere

when handling your own persistence of data and there is a complicated domain model - is it best to have a data mapping layer without any SQL directly in it? Is it really that big of a deal if all the SQL logic is mixed in with the mapping logic? I've heard that the mapping should have no coupling with any SQL, and even that using, for example, a JDBC ResultSet in the mapping is not good.

like:

domain <- mapping -> gateway -> database

what if tim cook was tim gook?

>>56937897
You've heard correctly.

The basic idea is to completely separate the logic retrieving the data with the application logic.

In this situation, the application should not care if SQL, or an API, or Morse code is generating the data in the background.

Or something like that. Basically, you should be able to completely rip out the section with SQL code in it and replace it with something else, and 80% of your application will be none the wiser.

>>56937949
but what if you know for a fact that you will only ever be using SQL? Are we just being dogmatic here? Is it really necessary to create another level of indirection because we might have a different way of getting the data?

>>56938002
>but what if you know for a fact that you will only ever be using SQL?
Hey, if you can see the future, you can do whatever you damn well please.

>Is it really necessary to create another level of indirection because we might have a different way of getting the data?
Generally speaking, yes, unless it's a trivial project that won't need to be maintained beyond 6 months or so.

>>56937684
No problem.

Programmers are antisocial robots in general. (Of course there are few exceptions like everything in the world.)

>>56937767
How does this manifest as a problem. What should they have done differently if you had a say?

>>56936631
You messed up your hand annotation of red players. The one in top right is wrong and your annotations of the field are very close in some cases. Looking interesting, why don't you use a fancier classificator than SVM like CNN?

>>56938309
>How does this manifest as a problem.
Are you serious m8?
Do you actually do extensive c++ work?
> What should they have done differently
The most fundamental thing I mentioned earlier. Re-write the language with the intention of progressing it further. Not stacking on more without removing the depreciated.

A good talk that explains it.
https://www.youtube.com/watch?v=ltCgzYcpFUI

>>56938282
While your statement is easily defensible due to vagueness and your personal definitions of "programmer" and "antisocial robot", I disagree.

I think you'll find that much of the workforce that programs aren't /r9k/-tier autists.

Most of the programmers I've worked with are closer to dudebros than robots.

File: 1437948970845.gif (1MB, 260x167px) Image search: [Google]

1MB, 260x167px

>>56936302
>that fucking video

>>56938396
>it's bad because it's bad. to make it good, they should have made it not bad, m'kay?
Good job spouting meaningless platitudes and not saying anything of substance.

>>56936302
Please, please just end computers already. It was a mistake, technology was a mistake, programming was a mistake. Erase all of it.

>>56938423
The percentage of normies are significantly lower among programmers than among people with other jobs.

>>56938527
I'll accept that.

>>56936302
Christ what a spaz.

>>56935886
It's a kernel?

>>56936302
>I'm a beginner
>I don't write production Rust software at all
>I fucking suck ass at public speaking. This is made clear by my near-incomprehensibility and complete lack of stage etiquette.

So why the fuck is it speaking?

/g/entoo devs everyone

https://www.youtube.com/watch?v=BRn1IGzDTYM

>>56938580
>>56938580
>>56938580

>>56938396
>Why C++ Sails when the Vasa Sank
>starts slide with picture of completely different ship sailing under completely different flag, from a completely different era

Thread deleted!

>>56938762
But why

>>56938762
but why

fucking bpshitters

>>56938777
>>56938777
>>56938777

>>56938583
pwned

>>56938762
Someone make one with an anime girl holding K&R or something

>>56938791
>>56938791
>>56938791

>>56938774
>>56938776
butwhy mind

same timestamp and everything. i will savor this moment

>>56938803

>>56938781
lol

>>56932615
PHP even has a hipster framework now, for "Web Artisans"