is C even worth learning in 2017?

Yes.

Of course not. The trillions of lines of C code worldwide disappeared overnight in 2016, so no new or legacy software require C developers whatsoever.

>>60397519
>implying there won't be a javascript to C interpreter and virse-versa

>>60397474

If you plan on doing any sort of embedded systems programming, absolutely.

C is the best. All other languages fail in comparison.

>>60397603
>interpreter
You've triggered me

>>60397665
t. C zealot

The only reason to learn C is for legacy purposes since it's still basically everywhere, there are more modern languages that do what it does better these days however.

>>60397474
Maybe if you want to do Linux system programming or work with embedded systems but otherwise not really.

What about C++?

Not really but it doesn't hurt if you're interested in it. Just don't force yourself to learn it because you think it's useful.

File: 1458371477884.png (61KB, 160x256px) Image search: [Google]

61KB, 160x256px

>>60397474
Reasons to learn C:
>Do you want to work on a project that's written in C?
>Do you want to write something that C is a good fit for?
>Do you think it'd be fun and interesting to write something in C?

This is much the same as for any language, but there you go.

No. No one will hire a self taught C developer for the 100 jobs available nationwide.

I can't take a programmer seriously if they don't even know C.

>>60398754
W-what about c++ ?

>>60398793
C++ is for people who can't write good C.

>>60398811
Triggered

File: image.jpg (85KB, 582x377px) Image search: [Google]

85KB, 582x377px

>>60398811
>C++ is for people who can't write good C.

File: 1491432359096.jpg (48KB, 450x720px) Image search: [Google]

48KB, 450x720px

>>60397474
>C

>Can't wrap head around pointers
>Gets frustrated
>Says C is a bad language

File: going.jpg (111KB, 1058x705px) Image search: [Google]

111KB, 1058x705px

>>60398811
>C++ is for people who can't write good C.

>>60398308
Please tell me what modern language does systems programming better than c?

File: o0t0dyCt7y1umi3qro1_500.jpg (55KB, 500x375px) Image search: [Google]

55KB, 500x375px

>>60398811
>C++ is for people who can't write good C

>>60398308

Nope. C is close to the metal and thus faster. It's practically portable assembly. All other higher level programming languages use runtime resolution and by extension, are slower. If you want speed, you want C, hands down full stop.

>>60397474
>>60397519
>>60398308
>>60398811
>>60398890
>>60399628
>>60399684
>>60399713

C++ is a horrible language. It's made more horrible by the fact that a lot 
of substandard programmers use it, to the point where it's much much 
easier to generate total and utter crap with it. Quite frankly, even if 
the choice of C were to do *nothing* but keep the C++ programmers out, 
that in itself would be a huge reason to use C.

In other words: the choice of C is the only sane choice. I know Miles 
Bader jokingly said "to piss you off", but it's actually true. I've come 
to the conclusion that any programmer that would prefer the project to be 
in C++ over C is likely a programmer that I really *would* prefer to piss 
off, so that he doesn't come and screw up any project I'm involved with.

C++ leads to really really bad design choices. You invariably start using 
the "nice" library features of the language like STL and Boost and other 
total and utter crap, that may "help" you program, but causes:

 - infinite amounts of pain when they don't work (and anybody who tells me 
   that STL and especially Boost are stable and portable is just so full 
   of BS that it's not even funny)

 - inefficient abstracted programming models where two years down the road 
   you notice that some abstraction wasn't very efficient, but now all 
   your code depends on all the nice object models around it, and you 
   cannot fix it without rewriting your app.

In other words, the only way to do good, efficient, and system-level and 
portable C++ ends up to limit yourself to all the things that are 
basically available in C. And limiting your project to C means that people 
don't screw that up, and also means that you get a lot of programmers that 
do actually understand low-level issues and don't screw things up with any 
idiotic "object model" crap.

>>60397603
>virse-versa

>>60397474
Any remotely decent programmer would be able to figure out C in a day, so knowing any C-like language is good enough.

>>60399745
Wrong.

File: 1494693613997.gif (232KB, 510x662px) Image search: [Google]

232KB, 510x662px

>>60399772
>Wrong.

>>60399766
This. Although C programs written pre-90s look insane.

>>60399772
It's just as true now as when he said it.

>>60397474
C isn't good, but necessary.

File: absolutely disgusting.jpg (74KB, 720x540px) Image search: [Google]

74KB, 720x540px

>>60398811

>looking for a programming job in my area

>search for 'java'
>48 results

>search for 'C#'
>52 results

>search for 'COBOL'
>32 results

>search for 'C'
>1 result

>search for 'C++'
>2 results

>>60400278
now do python you quadruple nigger

File: 1469470485929.png (72KB, 236x231px) Image search: [Google]

72KB, 236x231px

>>60400278
>fewer results for 'C'
>even though 'C++' contains 'C'

>>60400291
6 results.
4 of them for data science jobs, 1 for an "ethical hacker", and 1 for a sysadmin with scripting experience.

File: c.png (1MB, 1000x1400px) Image search: [Google]

1MB, 1000x1400px

>>60397474

YES

File: 1493426124376.png (1MB, 1920x1080px) Image search: [Google]

1MB, 1920x1080px

>>60398308

> " there are more modern languages that do what it does better these days however. "

Do what it does => YES
Better => Proof?

Name one.

>>60400435
https://benchmarksgame.alioth.debian.org/u64q/rust.html

>>60400535
>C beats out rust in almost every field
Really makes you think.

>>60400535
>C
>Lets the programmer do what needs to be done
>Rust
>Gets in the programmer's way as much as possible

File: 1493608787339.gif (110KB, 346x339px) Image search: [Google]

110KB, 346x339px

Its like all those modern language are saying "we can't trust programmer using our lang" yet were suppose to trust the ones who made the language in the first place.

> productivity turned on its head
> innovation stifled

>>60400435
>>60399696
Rust

Rust is a much better systems language with modern semantics and an actual package manager. Soon, with compiler optimisation (and ditching llvm), rust will out perform C if it doesn't already.

Go may even out perform C soon, fucking pathetic language

>>60400673
You clearly have no fucking idea what you're talking about.

>>60399736
fallacy

File: 1491641421503.png (130KB, 315x330px) Image search: [Google]

130KB, 315x330px

C > K-On

>>60401570
Well of course, you can earn money with C to buy more Azunyan figurines.

>>60400363
Got eem.

>>60401507
What fallacy?
Also, your post is in itself a fallacy: the fallacy fallacy.

>>60397474
No

>>60400673
This

>>60400590
>>Gets in the programmer's way as much as possible
Only if your programming style is hacked together skid tier

>>60397474
Is Jesus a jew?

>>60400673
A programming language is as good as the people programming with it. The Rust community is full of SJW idiots who jerk off to how "safe" Rust is over C or C++ and how Rust is the future rather than doing something useful.

>>60402976
>A programming language is as good as the people programming with it.
You using language X and sucking at it doesn't make X bad.
Take your shitty 14 y/o tier logic somewhere else dude

>>60397603
>c javascript interpreter running emscripten running a c javascript interpreter running enscripten running a c javascript interpreter
wtf

>>60402986
That applies to C too.

>>60402993
>reddit
Hmmmmm....

File: How do communities work.png (580KB, 800x2810px) Image search: [Google]

580KB, 800x2810px

>>60403223
Remove yourself from this entire website.

>>60397474

Yes, so you can learn C++17.

>>60398308
>only reason to learn C is for legacy purposes

So there's no new
>firmware
>embedded software
>OS kernels
>hardware drivers
>low-level tools & utilities
? Or are the new ones now implemented in <insert meme scripting languge>

>>60401570
fuck off Mike

>>60397474
I don't know. Do people still hire C programmers? Or do you have some task that can best be solved by C? Those are the only real ways to answer your question.

>>60405957
literally all legacy reasons

>>60406118
>firmware, embedded, drivers and low-level tools are legacy
Neo /g/ everyone

>>60406162
it's mostly done in C for legacy reasons you imbecile

>>60406350

what would they write instead?

File: 1457290198609-0.png (215KB, 500x720px) Image search: [Google]

215KB, 500x720px

>>60406369
C#

>>60398811
you're right

>>60397474
not 4 u

>>60406350
>ffmpeg is written in C for legacy reasons
>libpcap is written in C for legacy reasons
>your shitty microcontroller code is written in C for legacy reasons

File: what the fuck.gif (2MB, 433x297px) Image search: [Google]

2MB, 433x297px

>>60406489

>OS kernels, embedded software and hardware drivers in an interpreted language

>>60397474
No.
Only Python3 is worth learning in 2017.

>>60397775
interpreter

>>60406522
Stop posting trap horses

C++ but only as far as structs with methods and constructors goes, everything else is shit.

>>60406593
Why would you pull in all of the fuckery that is C++ and all of its inherent disadvantages for something as trivial and useless as that?

>>60406593
>"C with classes"
Fuck off.

STL containers alone is a reason to use C++ over C.

>>60406616

this

>>60406616
>>60406630
C++ Templates and the STL are a cancer.

>>60406608
>>60406616
Haha, I'm a retarded and I've only used C++ in the method that I've described. I baited you into giving me more reasons to form an opinion, negative or positive, over C++, you fools!

>>60406518
>software with over a decade of history is written in C
that's what i'm talking about you fucking idiot

>>60400369
HOW ABOUT PHP

>>60406647
ffmpeg is a library with extremely high performance requirements. Even if it was to be started from scratch today, it would make absolutely no sense to use anything else.
If you've ever looked at multimedia libraries, including very recent ones (libaom is a good example), they are all written in C.

>>60406616
this

>>60406639
fuck off, schoolkid

>>60406639
>compile-time dependencies is cancer
Enjoy a less typesafe and more error prone garbage relic from the seventies then.

>container types are cancer
Enjoy reinventing the wheel every second then.

>>60406641
>I baited you into giving me more reasons to form an opinion, negative or positive, over C++, you fools!
I'll give you some more then:

1. Containers are good, you want standard implementations for commonly found data structures

2. C++ has (proper) const-correctness and type correctness

3. C++ has RAII, which allow you to have memory safety (see smart pointers) and other resource handling (for example lock guards)

4. C++ allow you to do generic programming at compile time (templates and constexprs), increasing type safety (point 2) and detecting potential errors at compile time rather than in run time

5. Exceptions offers a decoupled error handling, allowing you to write code in a natural way where you assume error conditions are literal exceptions, rather than having to write convoluted code where you have to deal with errors inline. In fact, decoupling error handling from the expected code path allows the compiler to optimise for the expected path


>>60406647
They're written in C because there is an abundance in open source community of people who know C.

>>60406756
>Praises exceptions
Opinion completely discarded. You're clearly retarded.

Sadly C is the only somewhat reliable choice of interface language for FFIs, that is gluing together code in different languages. You can't just reliably call C++ or Rust code for performance reasons e.g. from Python, Node or PHP without a C intermediary.

>>60400363
meh, knowing C++ implies you know C so I don't think it's anything personal to C.

off-topic question:
I've never scripted... is it easy to pick up and where should I start?

>>60406766
Ran out of arguments already, I see.

>>60406790
>knowing C++ implies you know C
Not really. Knowing C++ doesn't mean you know how to write good C.

So, C is here to stay or is it becoming like COBOL?

>>60406766

Exceptions are good if you respect that they're not general purpose error handlers. They handle exceptional cases, and they're good at that.

>>60406818
This.

Idiomatic sepples with modern move semantics and whatnot is very different from what's considered idiomatic C code aka "caller is responsible for resource management"

>>60406812
>>60406824
There is no point arguing with retards.
Exceptions are a cancer. The fact C++ even has them IMMEDIATELY makes it completely inappropriate for replacing C, amongst other things.
Enjoy your fundamentally broken ABI.

>>60406820
Ask again in 5-10 years.

>>60406820
People have been claiming that C has is dying/dead for decades now, yet here is still is, as strong as ever.

>>60406839

protip: don't throw exceptions if you don't want them.

>>60406681
x256 is C++ (+ASM)
tensorflow is C++
numerical libraries often make heavy use of fortran
libmir is D and outperforms all FOSS alternatives
tsv-utils is D and outperforms other C software

>>60406839
>The fact C++ even has them IMMEDIATELY makes it completely inappropriate for replacing C, amongst other things.
First of all, C++ has a concept of nothrow and other exception-free semantics.

Secondly, I never stated that C++ was a "C replacement". They are two different languages with two different use-cases.

Also
>muh broken ABI
It's funny how this non-argument gets tossed around on /g/ like it should be some inherent requirement of any language to be able to interface with code generated by MSVC or GCC from before 5.0.

You can compile your code using GCC and link it with llvm-ld or compile it with clang and link it with GCC, it makes no fucking difference.

>>60406860
>>60406884
That's not the point. The feature exists, so it must be supported.
As such, you cannot throw an exception over an FFI boundary. Due to however the fuck a C++ implementation lays out exceptions or whatever, it becomes nigh impossible to link code from different C++ compilers together.

>>60400673
t. General Niggerowicz of Rust Evangelism Strike Force™

>>60406904

>That's not the point. The feature exists, so it must be supported.
Are you writing C++ or are you writing a C++ compiler?
A feature being there doesn't mean you have to use it.

>>60397474
If you want actually safe (verified) software, yes.

>>60406945

>C
>Safe

>>60406923
ABI issues are extremely important. It's one of the massive things that C has over C++, and why C++ is utterly inappropriate for a lot of the things C does.
Do you know why basically know languages can link against C++? Because C++'s ABI is FUCKED.
Anybody who wants to write a portable library that they actually want people to take seriously, they're more than likely going to write it in C, or at least use a C API/ABI.

>>60406904
>That's not the point. The feature exists, so it must be supported.
If you plan on implementing a standard complying runtime, yes.

>As such, you cannot throw an exception over an FFI boundary
Uh, yes you can. CORBA's IDL does, for example. Of course, if you define FFI for languages that doesn't support exceptions, like say C, but you shouldn't really be using a language designed in the early 70s as the default for modern language design, that's just all forms of retarded.

>Due to however the fuck a C++ implementation lays out exceptions or whatever, it becomes nigh impossible to link code from different C++ compilers together.
It's literally not, see >>60406884

You can use llvm-ld to link objects compiled with GCC together, and you can use GCC to link code compiled with clang together. In the latter case, it's necessary to specify flags to clang, but who fucking cares.

In fact, the very existence of NVCC for CUDA C++ and it's concept of host-compilers vs device code disproves your statement.

>>60406966
>Do you know why basically know languages can link against C++?
>basically know
>know

Lol

Also, you're wrong. It's possible to call C++ code from Python, it's possible to call C++ from Java and so on. See >>60406973

>>60406863
>x256 is C++ (+ASM)
That's ONE example. I can list shitloads of others that are written in C.
Looking back at my previous statement, I meant to say "almost all" instead of "all".

>>60406904
>muh abi and ffi
JNI supports C++, ergo you can interface with Java
Python supports C++ extensions, ergo you can interface with Python

What more proof do you need that your argument is fallacious?

>>60407013
>ignores all other examples

>>60407086
>>60406998
Now explain how I can do it portably, with any C++ implementation, with basically any language, on every major operating system. This is the sort of concern a library writer would have.

>>60407164
Also, I should add that you shouldn't add any dependencies for the user and/or make them jump through a bunch of annoying hoops.
Adding unnecessary dependencies is bad, and nobody is going to use my library if it's difficult to use.

>>60407091
Not him, but your "other" examples weren't for video encoding/decoding/transcoding.

>>60407164
>Now explain how I can do it portably
Use an IDL that generates stubs for C++ and your target language you fucking sperglord.

I mean, it's like you literally believe languages cannot possibly call C++ code, despite even you can even call C++ code from fucking FORTRAN.

http://www.yolinux.com/TUTORIALS/LinuxTutorialMixingFortranAndC.html

>>60406790
I usually write desktop applications in C++11+ and was tasked to start a project for some old platform which only had a C89 or something compiler.
It is not even near the same language.
Everything from string handling to making a for loop is different.
90% of that project was making the drivers and the tools I needed from C++.

It always annoys me when people write C/C++ as they are not identical.

>>60407164
>>60407211
Make a C extension to your Java code without adding a JNI dependency or jumping through the bunch of annoying hoops it is to define your JNI extension then, retard.

File: c-1.png (357KB, 454x603px) Image search: [Google]

357KB, 454x603px

>>60398308
>there are more modern languages that do what it does better these days however
The one factor that distinguishes C (and to a degree C++) from virtually every other non-machine language is that C doesn't try to invent its own programming paradigms. Every other high-level language in existence tries to define execution semantics or data structures that simply don't exist on the hardware level, and thus when using them you constantly have to second-guess what the compiler does with your high-level constructs to extract optimal code.

C, in contrast, is really just an easier way to write machine code. It just abstracts away the boring grunt work of register allocation and adds some simple structure to the code, but it doesn't try to invent things that the hardware doesn't do. That's what makes it pleasant to use, since you don't have to guess what the compiler will do to your code, reverse its work and just hope for the best.

>>60399745
PLS SIR DELET THIS

C is utter pile of crap and I'm not talking about point of view from 'modern languages' perspective, C did sucked even in times when it was made.

It's not a close to metal as people do say, it's only fast because of its very aggressive optimalizations.

It's not portable. It is way too difficult to write portable C code, which would be compilable by every C compiler.

It makes you write lots of boilerplate, for example you have to make your own container for every fucking type.

C is everything *nix nerds represent, with all its advantages and disadvantages. For some the main advantage is, that C allows you write as bad code as you want, there's nothing that can stop you from writing the worst code the world has ever seen.

>>60406954
>what is Isabelle/HOL and Coq
Get back after you graduate

>>60407741
So basically, C is a portable assembly.

>>60397474
short answer: no.
long answer: nooooooooooooooooooooo.

>>60407869
Yes, that is its strength.

>>60407847
>It's not a close to metal as people do say, it's only fast because of its very aggressive optimalizations.
So much this, although it should be added that C's simplicity (read: complete lack of any modern programming feature) also makes it very easy to optimise.

>>60407847
>It's not a close to metal as people do say, it's only fast because of its very aggressive optimalizations.
That's not true. The data model underlying the C standard (memory objects &c) is entirely modeled on real hardware, and that is exactly what makes it easy to optimize aggressively.

>>60408121
The C standard makes no assumption on the underlying hardware, which is why it is possible to port it to vastly different architectures. If it instead was, then optimisations would literally make no sense in the first place.

>>60408222
>The C standard makes no assumption on the underlying hardware
It assumes randomly addressable memory, but manages to avoid many of the particularities of that addressing.

>>60408248
Okay, yes, it makes almost no assumptions then.

My point is, you can have an architecture with no stack pointer and still implement C, or your architecture can have stack frames and the compiler will be able to do tail call optimisations for you. Optimisations isn't because C is so close to the metal, it is because it is so simple (as in, barely any features at all). See >>60407993

>>60408303
>Optimisations isn't because C is so close to the metal, it is because it is so simple
My very point is that "simple" and "close to the metal" are more or less synonymous in this case. The reason C can be considered "simple" is because the hardware that defines it is simple, and it doesn't define features that the hardware lacks.

C would be much more difficult to implement and optimize on such hardware as a dataflow architecture or a PIM (of those actually existed in any meaningful capacity). In the case of a dataflow architecture, Haskell would probably be the "simple" and "featureless" language.

>>60408351
Well, you have a point.

>>60408372
>>60408351
please dont agree with each other this is a discussion

>>60400278
>jobs that hire based on language expertise
Why? You can learn any of them in like 2-3 weeks easily just by having the documentation open while working on your first project. The concepts of computer science don't change between them, aside from functional programming to object oriented.

It's so irritating when looking for work.
>oh you have a masters and 10 years experience?
>you don't use java? Sorry, cant hire you even though it's piss easy to learn.

>>60408654
In 2-3 weeks, you can learn a language (that is not forth, lisp or go) on a superficial level. Nobody wants to hire a C "programmer" that doesn't know the standard by heart for example. Only for junior positions where they will train you anyway

>>60408715
In my first job I was hired as a C++ devoloper, despite not knowing C++ and also despite me being honest about this in my interview. I was told that my knowledge was what mattered, both the language and the framework they used would be easily learned with my background and understanding of their technology stack anyway

>>60408715
I just don't understand how you can learn LISP which is way more complex than C faster than C. Same for go.

>>60408882
I'm pretty sure the Anon you replied to tried to say that you can't learn Lisp faster than C, but I might disagree anyway. Even though Lisp may be more complex in the end, there's arguably less you need to learn just to get started than there is in C. Lisp's basic semantics are pretty simple.

>>60406118
You have no clue whatsoever of what you're talking about.

>>60399680
Typical comment of the kid who just recently "understood" pointers and thinks he's a genius.

Bro, pointers are not complex rocket science shit, stop trying make people feel bad about it.

>>60410219
>Bro, pointers are not complex rocket science shit, stop trying make people feel bad about it.
While I agree with you, it's also an obvious fact that many people are having genuine problems with them. I myself knew assembly prior to learning C, so that made them completely obvious to me, but perhaps OO language are ruining people who learn them prior to C.

>>60410328
>many people are having genuine problems with them
why do you even care lol

a person shaming people who don't understand pointers looks like a kid shaming another for not knowing how to change a lightbulb

should i learn WASB or DART?

>>60397474
More than you think, if you are interested in shit like Arduino and playing around with sensors and actual hardware.

If you are not interested in any of that, go for C++ in my opinion.

>>60406639
The only thing good about C++ is templates and classes, everything else is garbage, If I had proper templates in C it would be literally perfect.

>>60397474
No programming language is worth learning. I've done programming for 13 years now. Don't waste your fucking life. Do something else for the love of god.