People who say C++ is a good programming language are dumb. Y'all motherfuckers need to shut the fuck up and go back to CS 101. C++ an ass of a language and is still doing nearly as well as it does out of sheer inertia and literally nothing else. Every single real developer who says C++ is good is either a lying little kid who thinks he's cool for using a "real" language or a crusty old fart who've been too busy manually hardcoding his own diabetes medicine to look outside in the last 15 years.
>b-but, but... NOT BUTS. C++ is shit. Every single possible positive thing you could say about C++ involves C++ acting like some language other than C++. Even now I can hear your filthy cheetos-stained mechanical keyboards rattling away, typing those same old tired bullshit arguments about "performance", "optimization" and how it gives "good coders" the power to do amazing things (and make lots and lots of hair-pulling mistakes). Except you pissheads forget to mention one fatal detail: All these amazing things are only possible because C++ allows you to drop down into C and manually reinvent the wheel every fucking time. C++ is at its best... when it's actually C. In contrast, when C++ is used As Intended™ you get literally nothing out of it but a compiled version Python with irritating syntax, annoying header files, frustratingly obtuse compiler errors, aggravatingly small standard library and a rage-inducingly fussy manual memory management.
>b-b-but doesn't that m-mean C++ p-possesses flexibility that other langua-- I SAID NO BUTS YOU STAMMERING DICKSHEATH. C++ is only that way because it's a frankenstein's monster of a cruft-ridden idiotic attempt at modernly squaring a 1967 circle. If you had any sense you'd realize what a goddamn mess it is and switch to Rust already.
Grad student here. You don't know anything. Nothing will convince me you're not a junior in CS.
>C++ is at its best... when it's actually C. C++ is C with features to build great APIs. You write unsafe C code and wrap it up in a little package, complete with templates that the client can use instead of void *. It's a good language.
Jackson Hall
My money says you're a grad student in something like systems, databases, graphics, computational natural science (eg bioinformatiics) and not one in compilers, PL, or theory.
Oliver Campbell
>switch to Rust already Nah I'll stick with C. Isn't it time for your dilation?
Dylan Lewis
>everyone knows slow interpreted dog shit is the master race
Aiden Carter
>theoryfag can't actually code for shit surprise
Brayden Cook
Do you realize where you are?
Jace Sullivan
He's so tilted, he doesn't even recognize his fatal ERROR.
>HAHAHAHAHAHA >PRESENT DAY >HAHAHAHAHAHA >PRESENT TIME
Jose Parker
The only reason C++ ever got popular is because of the name. If it wouldn't have C in it's name, nobody would have looked at it much.
Nathaniel Carter
All that text, and yet you wrote literally 0 (zero) arguments for your position.
Kevin Myers
> theoryfag can't actually code for shit As an undergrad researcher, I set a record for number of apps ingested for an API usage study. That's on the order of 100k loc that have to work correctly in Scala and C++. Techniques I helped develop for several phases of the extraction are the most efficient that exist right now. My paper comes out in March.
What compels you to assume you have nothing to learn from me? I never assumed the same about you.
Samuel Bell
Wow, that's a real waste of time. Become a NEET and use your powers for something good instead.
Blake Thomas
>lemme sperg out some random big words to masquerade as a great researcher
Logan Watson
>avatarfag trying to make himself sound worth listening to Nothing new here. What does any of that shit have to do with the topic anyway?
Adam Diaz
> Getting second author on a paper at a good conference and getting an expenses-paid trip to present my work to a DARPA panel before turning 22 while securing stellar letters of rec is a waste of time. lolok
Carter Gray
Admittedly, it's pretty far from what OP was mentioning.
I just find it weird that sepples people immediately assume anyone who doesn't share their enthusiasm for it have never had to maintain a large codebase in it, as if the only reason someone could disagree with them is just not being informed.
Tyler Gomez
If you could look past your narcissistic need to tell everyone shit they don't care about, you'd see that his post wasn't serious.
Hunter Howard
Quite ironic to say when the exact opposite is true.
Jason Rogers
t. webdev it just werks
Christopher Morales
>it just werks There is literally nothing wrong with this. In fact, most normal people would consider this an advantage.
Anthony Young
rust doesn't have manual memory management (only optional with unsafe)
Grayson Price
Like being in grad school and therefore knowing everyone else hasn't finished undergrad because they don't like C++?
Chase Collins
yeah. whenever a normie ask me what 1st language to pick i say python because in the slight chance that he or she will actually try to learn it at least he or she will become more productive.
Camden Ortiz
Yeah quit being Sup Forumsay
Lucas Sanchez
My post was serious. I think academic CS work is a waste of time in 90% of the cases. Nobody needs that shit. Work on something that has an immediate use for yourself and those around you instead.
Who cares about that shit? Seriously, the other user said it well; It's just narcissistic bullshit you tell yourself to justify spending your time on such unnecessary garbage. Nobody will ever use it. Create something useful and creative instead. Also second author? Lol. If I was into this academic bullshit I would never do anything that wouldn't put me as first author.
Samuel Watson
You'll probably never use my (incremental) work. I'm completely to terms with that.
But I think dismissing 90% of "academic CS" (as opposed to...industrial CS?) is really odd. The model checker MS uses to test driver patches for correctness, the abstract interpreter that verified the A380's flight controllers, FlashFill, elliptic curve crypto, fully homomorphic crypto, backpropagation are all recent "academic" (by which I mean theoretical) CS, indispensible and used daily. Google, MS, and IBM wouldn't agree that 90% of academic CS work is a waste of time.
And believe what you want: I obliged this dick measuring contest because "I'm in grad school" was used to stone-wall all other opinions. It's fascinating how salty people get when you challenge their claims to being peerless.
Josiah Lee
Graphics, you got me. I think compilers people belong in the first group, by the way. I guess you like PL... what book should I read, then? I've read Sebasta and the Dragon Book, but that second one doesn't touch on PL at all, really.
Eli Lee
Consider this: if people who say c++ are dumb, they will be dumb in any language. Thus, it is better that they keep writing C++, so that their idiocy is kept contained.
Hunter Flores
okay but riddle me this: how could an undergrad student who's busy doing nerd shit all the time cultivate such a chinese cartoon pictures collection? i'll just let that sink in....
Aaron Nelson
*who say C++ is good
Liam Martinez
Rust is such a great language that Rust faggots have to spam every single place on the Internet begging competent developers to adopt their meme language
Brody Moore
I hate this meme defense. I also hate the Rusties who spend more time talking about Rust than writing Rust. What the Rust community really needs to do is keep plodding along and building an ecosystem until companies start adopting it en masse, or they don't. It's pretty binary.
Jaxon Barnes
OP I am interested to know what low level languages you actually think are superior to C++ and can back it up with evidence?
Jordan Johnson
What you listed constitutes not even a percent of CS work. And they aren't recent at all. And with academic CS I didn't mean it's mathematical aspect, but rather the semi-practical CS part. >(as opposed to...industrial CS?) No, as opposed to every other form. >Google, MS, and IBM wouldn't agree that 90% of academic CS work is a waste of time. They would, since they don't even read a percent of the shit that gets produced all over the world in CS. Also they often just tend to do things themselves instead of relying on academic CS.
Stop doing this shit, and start working on yourself. What would you develop if you had the time to develop anything you wanted? Probably not the shit you are doing right now.
Julian Morris
I grant you the compilers people are more applied. A lot of their work revolves around clever heuristics rather than proving theorems. I have plenty of respect for this (and other applied disciplines; I hope I gave no other impressions).
You're right. I do like PL. I wish more people knew the basics. I honestly think they'd write better software.
As for book recommendations, it depends on what you want.
For a PL-oriented take on compilers, I think Andrew Appell's "Modern Compiler Implementation in ML". It goes hard into dataflow analysis, which I think is very valuable. Standard ML really shines for compiler implementation.
The most underappreciated aspect of language design is probably type system design. I recommend Benjamin Pierce's "Types and Programming Languages". The best all-around type system-oriented PL book is probably Robert Harper's "Practical Foundations for Programming Languages", but it's so theoretical that it turns people off. The best way to read it is probably to watch OPLSS crash course lectures Harper gives on it and try following along. There are lectures like it every year.
Abstract interpretation and model checking are more practical fields because they underpin all deep semantic analysis of programs, which BTFOs everything less powerful, but only a few people really know how to do it well. I wish there would a good textbook to recommend. Patrick Cousot's lecture material is highly recommended: cs.nyu.edu/~pcousot/
Everyone I've mentioned is part of the DeepSpec project to formally verify a whole software stack for correctness. The lectures online should be good.
I also like to watch POPL paper talks summarizing new research in the field. They're intuitive and fast-paced rather than super technical.
While I have your ear, what graphics texts would you recommend? There seem to be a lot. Always found graphics interesting but never got into it.
Wyatt Foster
>superior This notion does not exist. Superior in what regard? The point is, that C++ might seem to be a good solution (to someone who doesn't understand yet), while in fact it is a badly constructed mess of anti-patterns and bullshit in it's core. Just programming C is superior. Just pick plan9 C and you even have a modern Version.
Cameron Barnes
> What you listed constitutes not even a percent of CS work. 90% of CS publications are unremarkable, but so is 90% of everything. That's hardly reason to dismiss everything. >And they aren't recent at all. FlashFill is recent (2016). So is work on concurrent models of C++ (2016) and the most recent breakthrough on bootstrapping homomorphic encryption (2016). > Rather the semi-practical CS part I'm not sure what this means. >They don't even read a percent of the shit that gets produced all over the world in CS. If so, why do they sponsor all the good conferences? > Also they often just tend to do things themselves instead of relying on academic CS. They do a lot of in-house research because they see the value in it. They also hire lots of people from the academy. To give an example I know of, Facebook and MSR gobble up PL reseachers like no one's business. > Stop doing this shit, and start working on yourself. What would you develop if you had the time to develop anything you wanted? Probably not the shit you are doing right now. I also worked for a bretty cool VR startup. And why assume I don't work on myself? I still read (news, textbooks, manuals). I still practice programming. I think the shit I was doing in undergrad was way more interesting than writing CRUD apps or data serializers (which, be honest, is like four fifths of all programming jobs), and I feel I was lucky to be involved with it.
Alexander Ross
>90% of CS publications are unremarkable This is pretty much what I said. I see, so you seem to agree with me. >Flashfill if this is that excel shit, it is really unremarkable and nothing special at all. >concurrent models of C++ C++ is a garbage language, and there are good, useful, and performant parallelism models since ages. Actually the most important paper is from 78. Adapting those to the piece of shit that is C++ is nothing special. >the most recent breakthrough on bootstrapping homomorphic encryption Useful, but also more math than CS in a sense. >I'm not sure what this means. CS that pretends to be practical. I don't know what it's called in the States. Possibly applied CS or something. The branch where they will produce applications that nobody ever uses or would even want to use. >They also hire lots of people from the academy. From where else should they get people instead for their own bullshit?
Jaxson Nguyen
at least it's fairly consistent not jit just kicked in now yo ram is gone
John Johnson
I don't think we disagree all that much. We were using language differently. > if this is that excel shit, it is really unremarkable and nothing special at all. It's a much harder problem than you're giving it credit for. microsoft.com/en-us/research/wp-content/uploads/2016/12/popl11-synthesis.pdf This paper is older than I remembered, but there were lots of follow-ups generalizing this further. > C++ is a garbage language, and there are good, useful, and performant parallelism models since ages. I personally don't like C++, but the problem is an important one. cl.cam.ac.uk/~pes20/cpp/popl085ap-sewell.pdf Slightly relatedly, here's new work proving math.h correct automatically: youtube.com/watch?v=KH5fCmOVglc
Obviously, not all academic research on topics like this is practically applicable, but more is than lots of people realize. That's all I was saying.
Kayden Myers
what I see here is basically "C with classes", which only reinforces faggot's, err, I mean, OP's point.
P.S. OY VEY
Juan Edwards
I did! And I like C++.
Easton Harris
retard here Why the fuck do I have to do function prototypes in c++? Why can't the compiler just check and compile in the right order like other languages do?
Benjamin Mitchell
Why did you make this thread if you don't want any discussion on the topic of C++? What kind of hive-mind spits out mongoloids like you OP?
Jason Gray
>Why can't the compiler just check and compile in the right order like other languages do? No reason at all.
Brody Evans
Not OP, but you can't discuss with people who like C++. They are lost. It's just better to let them go. Just informing newbs to not go down the C++ path is enough to justify this threads existence.
Dominic Robinson
OP is a faggot, but C++ is still shit.
Lucas Myers
muh single pass compiler even though e.g. classes discard this rule
Landon Kelly
>rust >not java Opinion discarded.
Grayson Howard
The whole reason C is considered a 'systems language' compared to other languages is that C is extremely weak typing, it will try and cast anything to make it compatible with the types it is being applied to. It has absolutely no restraints on pointers and bounds of any container. The fact that the language has no type rules means it crashes a lot less than a strongly typed or bounds checked language. This is the exact opposite of where every modern programming language is going toward which is strong typing, strong bounds checking, and controlling wrong types instead of just building systems that are broken by design and just held together by debugger fixes
Jonathan Allen
This is correct. C was designed so it could be compiled in a single pass, back in the paleolithic when computers had about 8 bytes to work with. Because of muh backwards compatibility, it's still necessary to forward declare function prototypes for mutually recursive definitions even though C++ can't be compiled in single-pass fashion and Clang and GCC will fucking exhaust system memory on any platform where single-pass is necessary. If you're targeting such platforms nowadays, you'll just cross-compile on a powerful box anyways.
Kevin Evans
This is mine. I'm sorry for the confusion. I meant to reference this:
Jackson Rogers
>rust
can i feed my family with that?
Levi Walker
You certainly can, but it will be harder to find a job.
Nolan Green
>not a single job posting in my country
Kevin Myers
I mean, if that's the case, you probably live in a country where there aren't jobs writing OS kernels or program analyses either.
I'd move to a different one for work desu.
Camden Stewart
I'd shut up you retarded tranny. Systems programming is dead. There won't be another big mainstream Kernel anymore since Linux is enough (worse is better), and even if there came a new commercial one into existence, it wouldn't be written in Rust. >inb4 Redox It's shit. The designflaws could only come from someone who thinks writing Rust is a good idea.
Joseph Cooper
t. triggered third-worlder
Andrew Perry
>OH LOOK IT'S ANOTHER C++ THRED
Sup Forums - Battle stations, Vidya Gayms, C++
Daniel Barnes
Wow, you sure showed me. I am really triggered by Rusts unpopularity. Damn I really WISH I could program Rust for a living, fighting that retarded kiddyglove compiler every damn day.
Btw, I am not the one who can't find a Rust Job in his country.
Ryan Moore
Oh fug, the song started playing in my head.
David Howard
> I can't do systems programming, let alone with an affine type system, therefore no one can or needs to.
Blake Sanders
The point is that I can do (and actually DO) systems programming without the need for Rust. I have tried it out, and at one point I stopped and just thought it was retarded unnecessary garbage. Who knows if that has changed though, since the Rust Devs can't even keep their gay language stable.
Dylan Garcia
>Systems programming is dead. >The point is that I can do (and actually DO) systems programming. user, assuming these are both you (or at least both of these posters inhabit the same nervous system Fight Club-style), you do realize at least one of these has to be false, right?
Landon Clark
obviously rust is a white male's language, no one else is so heavy with guilt to mewl about diversity constantly
Julian Jones
COBOL > everything else.
Levi Watson
We're getting tears every single day this week.
Owen Gutierrez
>more plebby ad homs Color me impressed.
Dominic Anderson
...
Bentley White
Your shit thread doesn't deserve anything better.
Ryder Gray
>Your shit thread. Not my thread. I'd still be embarrassed if the best argument I could give against a programming language is calling its userbase fags.
Owen Butler
Thank you for all these great resources.
Aaron Davis
All dynamically typed languages are fucking awful.
Thomas Rogers
i agree with OP desu
t. haskell user
Ayden Roberts
...
Julian Lopez
No problem! Very happy to help in your auto-didactic project.
It's always great to meet people interested in learning about new shit on the Korean cartoon/neo-nazi mouthbreather and Marxist-Leatherist with Somalian characteristics BDSM enthusiast website.
Joseph Jenkins
>compilers Not him, but what are you implying about students studying compilers?
Aaron Garcia
Shut up and acknowledge C++'s superiority!
Dominic Long
If they have a serious education in the foundations of compilers, then you understand the value of a good type system and know what one looks like.
Granted, you don't need that to do babby stuff, but you're not going to get far into high-octane shit like monomorphization or escape analysis without some actual PL theory.
Logan Baker
>an actually funny thread on Sup Forums wow, that's been a while
Ryan Green
Oh, I guess that doesn't really apply to me then. My focus is on the backend.
