If Rust is like a better C++, why has it not picked up momentum

>>57541726
>why has it not picked up momentum in the work environment.
Industry is very inert.
>Also would it be possible for someone to design an AAA video game engine with it?
Have been done in assembler, can be done in rust as well.

The language is still young, but it's slowly getting used by more places, just give it time.

>>57541829

You don't believe it will become another dead language that no one uses? I would like to learn it, but I want to be able to use it for work

no libraries

>>57541865
You can use C libraries or write your own binds

>>57541882
no C++ libraries

>>57541919
Due to name mangling you have to write "extern C" when you're writing a library in sepples, thus there's no difference.

>>57541955
D can link with C++ libraries

>>57541963
I just told you that there's no difference.

>>57541974
>Forced to jump through hoops
>No difference

>>57542322
Yes, we get it, sepples suck.

>>57541852
Within a certain probability it will be a stillborn language like 99.9% of them.

>>57542475
Qt is the only competent cross-platform GUI library and it's for C++.
>Gtk
Fuck off.

>>57542567
What's the problem with GTK?
Honestly wondering.

>>57541919
It's in the plans for 2017 to add easy bindings to C++.

>>57541852
Doubt it, the community is already pretty big and as a language it actually gives useful properties.

Also they're slowly integrating Rust component into Firefox, so I don't see it dying.

>>57541726
>If Rust is like a better C++, why has it not picked up momentum in the work environment.
Because it's worse than C++ in all the ways that count, such as not being backwards compatible with C, and not being established in the industry for 30 years. The compiler is slow, the additional safety checks make your program slow, and the community is full of toxic identity politics that instantly disqualify it from being considered by any corporation.

>>57541726
No standard
o

S
t
a
n
d
a
r
d

>>57541726
unproven + not ISO = high future risk

small user base + volatile specification + minuscule library collection = high current use costs

Also, you basically already have to be competent in C++ object lifetime management to have a prayer at using it effectively without feeling like the borrow checker isn't some cruel arbitrary overlord stopping your project from building.

>>57541726
>If Rust is like a better C++, why has it not picked up momentum in the work environment.
Because of inertia

>>57542739
1. Poor support for anything that's not Linux. It gives the end-user a hard time if you're not on a Linux machine.
2. It's also ugly if you're not on Linux
3. Bad ideas
https://blogs.gnome.org/desrt/2016/06/13/gtk-4-0-is-not-gtk-4/

>compiler updates also change semantics and syntax
A hard riddle indeed

File: ay.png (525KB, 745x596px) Image search: [Google]

525KB, 745x596px

>>57542988
>and the community is full of toxic identity politics that instantly disqualify it from being considered by any corporation.
this 1000000%

File: 1478633092405.png (73KB, 2048x1024px) Image search: [Google]

73KB, 2048x1024px

>>57541726
stop shilling this cuck language. maybe if you are too retarded to understand pointers this >>>/trash/ language is for you

Rust is dead though. It had a lot of hype behind it and then it failed to deliver on so many parts.

rust is infested with sjw's and lawyers, most poisonous community atm

>>57541726
rust is dead, finished. wait for the gc changes in d lang and migrate there.

Go > D >>>>>>>>>≥>>>>>> Rust

it has some nice ideas, and some terrible ones.
i dont think it will become more popular then c++, specially with standard actually getting updated now.

>>57544019
>garbage collection for AAA-games

nice try kevin

>>57544050
If you're making a game you're doing it in C++, there is no alternative to C++ for games

>>57543426
>It's also ugly if you're not on Linux
It's ugly for me even on Linux tbqh

>>57544215
What about Rust?

>>57544215
Is the only requirement an openGL library?

>>57544215
Unity

>>57542739
As someone who had the pleasure developing with GTK:

- there are like 5 people working on it, of which 3 are from red hat
- good luck with the people who want the 'old' look and feel back and start campaigns in their distro to unGNOME3ify the applications
- heavily underdocumented - in particular the newer components
- in the same spirits, Glade is shit and mostly outdated for the newer widgets
- the layout system sucks (most do, but GTKs layout system is completely inane)
- bindings: Good luck with all the incomplete bindings and their non-existent documentations (except for the Python one). You basically have to know how to develop GTK applications in C to use it in Vala - which is one major reason nobody but the elementary devs use it
- >>57543426
- because GTK is really only the high level widget set, creating custom widgets sucks
- retarded design decisions, e.g. combining buttons is a theme setting
- regarding themes, GTKs theming story sucks. According to the devs, the old system sucked and according to everyone else, good luck with Adwaita

Theres probably more, but that is all I remember right from the start. That said, it's probably still better than the enlightenment toolkit.

TL;DR: Qt completely destroys it in every aspect and if I can't have nice bindings in other languages it's a price I'm willing to pay. However, since Python bindings exist there is no reason it shouldn't work with Rust as well. Providing a complete binding to Qt is a matter of manpower.

>>57543875
>there are no pointers in Rust

>>57544396
I am interested to setup my next project with vala, but after reading this my stomach hurts. I have already notices, that the documentation is horrible. Would you recommend Qt instead?

>>57544019
The war on these languages is going to be the death of all of them. They're all backed by big companies.
Facebook backs D
Mozilla backs Rust
Google backs Go
But everyone's fighting to be the best language, and they are all constantly changing and so they are losing their supporters. Meanwhile, people who are presented these three options won't choose and decide to wait on it and keep using C++.

>>57544497
>Would you recommend Qt instead?
Yes.

It's not perfect; I don't like the way you'll need QTypes and how QtQuick is more and more priorized by the devs.
But at the end of the day and compared to the alternatives it wins hand down.
Most likely because developing a complete Widgetset is a number game, so it's good there is a complete team dedicated to creating Qt.

File: 1470341172739.jpg (45KB, 333x333px) Image search: [Google]

45KB, 333x333px

> tfw Ada/C

>>57544497
>my stomach hurts
Stop living in India

>>57544520
>But everyone's fighting to be the best language, and they are all constantly changing and so they are losing their supporters. Meanwhile, people who are presented these three options won't choose and decide to wait on it and keep using C++.

This, same with Swift and C#. Which is good, because it forces the C++ commit tea to finally put modules into it.

That said

some D faggot working and writing D tools at fecesbook != fecesbook backs D

>>57544631

Now if only there was a LLVM frontend, so I could enslave people with my unfree software.

>>57544607
thanks mate
take this (You) as a sign of friendship

>>57544239
You're not reading the thread, Rust is dead on arrival

>>57544607
pinned to my nipple

bonus: On linux, unless you do KDE development, don't use your repos version of QtCreator as it will be too old (and not stable). Use the Qt installer from their website and don't forget to skip the botnet part.

>>57544727
>according to /g/
But let's be honest, when was /g/ even remotely right about things?

File: 1473661982427.jpg (68KB, 960x540px) Image search: [Google]

68KB, 960x540px

>>57544777

>>57544833
/g/ predicted Trumps finger movement?

>rust community is filled with sjws
Tell me more please.

>>57544663
I don't think gcc restricts that

>this thread
>enterprise programmers who are completely impotent to do anything without a library ecosystem
I could talk about Rust, but it would be wasted on idiots who cant understand the benefits without a thousand hand holding beginner books, thanks for reminding why I dont frequent /g/ for programming threads anymore

>>57544937
Seems like a miss and hit:
http://stackoverflow.com/questions/23653913/what-is-the-difference-between-the-adacore-gnat-version-and-the-standard-one

>>57545049
Ya,
fsf=outdated but free to use
Gpl version = almost new but gpl
Pro = new, free to use, but expensive

>>57542988
It's written by SJWs

>>57544520
Well no, if consumer electronics has taught us anything, it's that when lots of comparable things compete with one another, the one with the nicest name wins. Therefore, D is going to win.

Imagine explaining it to a manager:
>We want to use Go because Google does it, it has nothing to do with Pokemans
>We want to use Rust because blah, blah, blah, technical stuff...
>We want to use D because it's one language better than C

Clearly D wins.

>>57541726

>If Rust is like a better C++, why has it not picked up momentum in the work environment.

> Falling for the SJW nu-males propaganda about a "better cpp"

>>57545211

Don't be ridiculous, they're incapable of actually making anything. They only know how to destroy, that's why they need to parasite themselves onto existing movements.

>>57542988
>safety checks make your program slow
>they're statically ensured checks done at compile time

Anon, you don't know what you're talking about.

>>57542567
Tk.
>inb4 it looks like shit
No. Learn how to use it properly and it will look okay. Do NOT follow two decades old tutorials.

http://tkdocs.com/

>>57541726
Code written in Rust must be ten times faster to write and run ten times faster before it would be worth the enormous costs of moving to it.

>>57542805
>Also they're slowly integrating Rust component into Firefox, so I don't see it dying.

Firefox itself is dying.

Also who the fuck think Vec is a good name?

>>57544520
D will never happen though.

Just look at their document: https://dlang.org/spec/spec.html, compare to a younger language (write by 3~4 programers) like crystal: https://crystal-lang.org/docs/

One introduces all features of the language, one just simple rantings about garbage no beginner want to care.

>>57549050
Are you saying D's specification is it's downfall because it's not catering to babies first language, or the other way around?

>If Rust is like a better C++

Only a complete and utter retard would actually think this is true.

>>57542567
>competent library
Nice buzzword you have here.

>>57549625
I meant D documents are crap. Simple as that.

And since it's crap, beginners (to the language) will just give up trying to learn it.

And since no one will try to learn it, it will be forever remaining the creators' toy.

I keep hearing about D since 2008 or so, but have never heard it was used to create anything worthwhile.

>>57550459
I only skimmed it, but it seems quite good. Any non-novice programmer should be fine with it. It's not supposed to be an introduction book.

Is Rust worth learning? I've been hurt before. (Golang.)

>>57550819
It's an interesting language for sure, but I'd start with D.

>>57541955
No you don't

>>57541726
because if everyone moves to Rust, who is going to maintain C++ code?

>>57550819
I think the other anon discouraged you enough when he answered "learn D (a dead language)" to the question "is Rust worth learning?", but repetita iuvant....
Don't fall for memes.

>>57544520
>they are all constantly changing
Go has maintained the 1.0 backwards compatibility guarantee for several years now, since the release.

Because of the autistic type system.

>>57541726
Firefox is on its way out, so is Rust. It's just another meme language.

>>57550819
Rust will be dead in less than a year. Don't bother.

>>57550848
Rust is the only new systems language with any real momentum behind it, the other one is Golang. Both are awesome but only Rust has some truly cool shit like memory safety and also generics.
I don't hate Go, in fact I might use it but if Go isn't good enough to do a fast rest API in, then fuck it I'll use rust.

>>57552870
Go hits the sweet spot between performance/productivity. I won't be surprise it will be next Java few years from today.

>>57550565
It's garbage since they put a specification where a reference should be.
Beginner mistake.

C++ has over 3 decades of active development and design. on top of that it has the most support out of any other language out there, including libraries and APIs.

You can't expect everyone to jump ship to the first trendy meme language that these SJW companies shit out.

>>57552870
Rust isn't really a systems language, they thought they could get away with advertising it that way after Go did the same thing with a similar set of features, but for low level you still need C/C++ as much as you ever did, that's never going to change

>>57552870
Go isn't a systems language, but a server language.
>>57553154
Can you explain what Rust is lacking in comparison to C/C++?

>>57553206
>Can you explain what Rust is lacking in comparison to C/C++?
ability to get easily around all the safety features that make it unusable for many purposes that people use c++ for

>>57544497
Qt is garbage. Gtk is much better. It has bindings from many languages and looks better than shitty Qt. Many large companies use gtk. For example, bloomberg.

>>57550848
D stands for DOA. It has an optional GC so you have to sift through the libraries to find out which ones are using it and try to use different libraries to thin down the runtime costs. What a joke.

>>57553248
Except it looks like shit on anything but Linux and breaks theming every other release.

>>57553271
Why would i NOT want to use gc?

>>57553878
A GC is completely fine for normal use. Retards just meme about it. For a real reason to not use a GC is if you need to meet hard, real time constraints.

>>57553248
>It has bindings from many languages
Of which most are incomplete or not documented.
> and looks better than shitty Qt.
Only on a unmodified, nonfunctional utopic GNOME desktop
>Many large companies cripple along with gtk.
Many more use Qt. More important, there is third party support. Best you get for GTK is indirect over Wx

>>57553216
>>57553206
Which is a way of saying he can't into Rust.

>>57554189
Rust's safety features are completely redundant, you should just write better code

>>57554221
t. fizzbuzz expert

>>57541726

It feels unstable as fuck. They keep adding new language features in minor versions and release a new one of those every freaking other month. They are supposed to be working on implementing a well thought-out stable spec and not be changing both the spec and the implementation all the time.

>>57554353

That's just embarrassing. How do big projects like Servo cope with their code breaking every couple of months when random parts of the language get changed around with no warning? Maybe if they weren't constantly trying to keep up with such an unstable language, they'd get something done.

Why are use rust when Ada does the same thing but better?

>>57554467
ADA is used by the army with their scary guns, no SJWs will touch it

Learning new toy languages only makes sense when you already know reasonably well C, C++, Java, Python and Fortran.

Because Go covers 99% of its uses cases in a much more pleasant manner. They also de-emphasized CSP-style concurrency, which was originally a strong point of Rust, so again it loses to Go.

>>57542567
Ttk

>>57544050
Unreal (hyper super AAA engine) has a slow as fuck GC.
Unity (less popular in AAA industry) has a slow as fuck GC.
Go has a fast as fuck realtime GC that you can tell to fuck off until such a time that you're comfortable with running mark and sweep on these niggers.

It's fine.

Like D, Rust is just another sepples.

>>57554467
a) it doesnt the same thing
b) just look at your options regarding compilers (and therefore ABIs), standard libraries and documentation
c) probably >>57554496 as well
>>57554784
>Tk
is a PITA for more than simple dialogs
>has a slow as fuck GC
only used in their script shit that doesn't do the heavy lifting

>>57555204
a) you're right, one can do unions
b) the compiler options are great. Even the shit outdated one is faster and produces better binaries than rust's main. Standard libraries are well documented and best of all, don't have sweeping changes every month.

>>57541726
>If
big word


>...Rust is like a better C++, why has it not picked up momentum in the work environment.

IF it is better, then inertia is a big part of the reason. All the best developers who might use it already use c++.

Also, it's competing against a huge catalogue of libraries, tools etc. that are mature and that developers already know, are already embedded in their workflow etc.

So yeah, inertia.

>Also would it be possible for someone to design an AAA video game engine with it?
not my area, but I guess so.

If I like C but hate C++ will I like Rust?

>>57555812
Do you also like Haskell?

>>57541726
I wonder if it will work for web development.
I'd really love a language that is modern and allows for nice abstractions without straight up murdering performance - and it seems like this is what rust is designed for.

>>57554799
explain this sepples meme

>>57555852
Never used Haskell but I'm always up for new things baby :3

>>57556022
is being made to suck cock new for you?

>>57542988
>the additional safety checks make your program slow

No. All the non-compile time checks go away. You'd have to opt-into run time stuff.

>>57553216
lol. What a load of shit. I've written kernel level code in Rust just fine.

>>57555812
You might, but you might not also. I hate C++ but Rust is tolerable. It doesn't have all the warts C++ has, but it's also a bit more restrictive around the edges than C. If you like pointer arithmetic, a currently stable ABI, or make a lot of direct system calls, it will probably feel too strict. But then again, iterators and ml style matching are nice. Just stay away from Steve. He's a massive faggot.

>>57544396
>- the layout system sucks (most do, but GTKs layout system is completely inane)
Worse than CSS's where there hasn't actually been any layout functionality (until flexbox and grid become usable) and you had to misuse a completely unrelated feature (floating) or in very desperate situations display: table-cell?

Because I honestly can't believe it gets any worse. The people who were in charge of the web in the 90s and early 2000s should be murdered brutally.

>>57555937
If it ever get's a production-ready asm.js/WASM compiler, I would be very tempted to use it for the client-side. Go still seems like a better choice for servers, at least until Rust gets a comparable standard library networking stack and some more third party libraries.

>>57541726
Because people like C++

>>57556097
Does the WASM-Specification already include web-apis or will we always need JavaScript to actually do stuff?

>>57541726
> Also would it be possible for someone to design an AAA video game engine with it?
Technically possible but who would waste time and money in doing it instead of just using one of the existing ones?

>>57556097

I thought it had a wasm /asmjs target now

>>57556132
You mean bindings to DOM stuff? No, not a part of WASM afaik. (but I'm not a webdev and it shows)

>>57556139
It does but it's still very hacked together. It's coming along, though.

>>57556132
AFAIK the current spec does not have the ability to call browser APIs, but it is planned.
>>57556139
Point me to it, if it does. Googling only gives me prototypes and Coming Soon™ notices.

>>57556163
I wonder what web-assembly will do for the next generation of frontend frameworks.
I wonder if it will help at all.

>>57556090
No, not that level of bad.

>>57556209
Oh thank god, you saved my sanity.
Just the thought of a layout system more insane than this almost destroyed my mind.

>>57556139
>>57556188
It doesn't exist except in theory, someone made a WASM version of LLVM and rustc uses LLVM so people think it automatically gets WASM output, but it's a lot more complicated than that, and they aren't compatible with each other due to being based on different LLVM versions.

>>57556240
You know what is the most stupid thing about Cascading Style Sheets? The syntax is not even fucking cascading.

>>57556188
>Point me to it, if it does.

$ rustup target list | grep wasm
wasm32-unknown-emscripten (installed)

https://users.rust-lang.org/t/compiling-to-the-web-with-rust-and-emscripten/7627
https://github.com/rust-lang-nursery/rustup.rs
https://github.com/tcr/rust-webplatform

Still a WIP, though.

>>57556400
First link never loads with JS enabled, because it blocks on a Discourse CDN script. Stupid SJWs.

>>57556395
Is there even a single thing about CSS that isn't stupid? Even a single feature that works as advertised with no weird side effects?

The scariest part is that I've worked with it long enough that it's beginning to make "sense" to me. All these hundreds of workarounds you need to make something interesting.
But then I had to prepare a CSS training course for the apprentices at my company and while making that I realized just how insane everything is.

>>57556476
>js enabled

Works On My Machine™.

>>57556506
Does not work on my machine in any browser™. You probably already have it cached.

>>57556501
Font styling works as intended, so do flexboxes. Not much else.

>>57556561
Nope, that wasn't it. Deleted cache then ctrl+F5, still loads in a second.

I'm not defending that browser software, btw. I think Jeff Atwood is a fat piece of shit and his software is bad.

>>57556395
I think the most stupid thing about it is that despite its name it's for

- style
- layout
- animations
- probably more

>>57556634
*forum software

>>57556634
It loads now, so was probably Discourse's fault.
>>57556636
- customizing mouse and keyboard interaction
- conditionally hiding content
It's also turing complete.

>>57556581
Flexbox does have some inane naming, though.
Like Flex-Flow. What the fuck? Why does it even exist when setting it to auto and using width or height does the same?
Want to align items along the flex direction? justify-content.
Want to align items perpendicular to the flex direction? align-items.
Want to align multilines perpendicular to the flex-direction? align-content.

That. Naming.
And then there are several cross browser inconsistencies with it, especially when you want to use the relative height of parents.

And fonts aren't really easy either. Doing fonts correctly with cross-browser consistent fallbacks, caching and all that is a surprisingly difficult thing. Just look a stuff like this: https://github.com/guardian/frontend/blob/d4422b4537165424e70a898d150db4e806ba04d6/common/app/templates/inlineJS/blocking/loadFonts.scala.js (Code used by the guardian's website to handle fonts).

>>57541726
Is Rust just C++ with a fancy compiler?

>>57556924
It's closer to Swift or Go.

>>57556924
It's C++ if it had a bastard kid with ML.

I'm scared if I pick up rust it'll be a dead language/wasted time.

>>57557499
It already is. Just forget it happened.

>>57557499
If you use C++ and friends it should be easy to pick up, and dealing with the borrow checker will probably make you a better C/C++ programmer. If you only use GC languages don't bother unless it takes off.

>>57557550
>dealing with the borrow checker will probably make you a better C/C++ programmer
This is a myth, the borrow checker doesn't give you anything you don't already have with Valgrind and friends.

>>57557603
I never found valgrind very useful, because there were always memory leaks from the libraries I was using and I couldn't do anything about it.

>>57557650
They'll still have memory leaks when you use them in Rust.

>>57557603
> the borrow checker doesn't give you anything

Immediate, compile time feedback.

>>57557691
Rust at NO point claims to prevent memory leaks (although safe rust prevents a subset of them). Not sure what you're on about.

>>57557727
Then what's the point of the language?

>>57557727
>Rust at NO point claims to prevent memory leaks
Then it is built on a lie.

>>57557851
It prevents a rather large subset of common memory bugs, including things like use-after-free and it has automatic de-allocation of boxed (heap allocated) types.

The reason it doesn't guarantee "no memory leaks" is certain kinds of leaks can happen even in languages that are memory managed like python or java. (for example reference counted loops will always have a reference and therefor never be freed, even once they are no longer in scope)

>>57557852
No it's not. You made assumptions that weren't true and then blamed it on the language.

Frankly, I don't know of a language in existence that can catch ALL memory leaks.

>>57558041
Rust is advertised by the SJWs as a silver bullet for systems programming. I'm not making assumptions. They're telling lies.

>>57558110
>memeing buzzwords

Link me to where the official documentation claims that it will never ever leak.

Rust is slow lmoa just use assembly

>We do not demand that Rust run on "every possible platform"
Wew C++ already wins

>>57558282
>It must eventually work without unnecessary compromises on widely-used hardware and software platforms.

I'm don't see the issue. It lets them remove a bunch of "implementation defined" ambiguity and the only real trade-off is you can't use it on some extremely exotic outmoded hardware.

>>57558110
They never claimed to prevent leaks, only memory UNSAFETY, memory leaks are not memory unsafe.

There's really no way to prevent leaks without having a useless language, even garbage collected languages can have memory leaks.

>>57558188
Rust actually allows optimizations that would require a lot of care and then explode anyway if done in C or C++, i.e. zero-copy parsers.

I wrote an mbox parser that can parse a 300MB mbox file with more than 60_000 emails in under a second.

In comparison mutt takes at least 5 or 10 seconds, mine is basically as fast as the disk because it avoid ANY copies or reallocations using an RcRef from owning_ref, the same buffer that was allocated to read a line from the input stream is used all throughout each layer of the parsing, try and do that with C or C++ without having to kill yourself.

>he's not a wizard at memory management

>>57558463
The only way to get memory leaks if you're not an absolute retard is when destructors/finalizers fail to execute for whatever reason.

>>57558908
Yes. What's your point?

>>57558908
The same applies to Rust leaking memory, you either have to do it explicitly (using mem::forget) or you need to create a cycle.

>>57559183
>There's really no way to prevent leaks without having a useless language
is simply wrong.

The syntax isn't enough like C so it's dead :^)

>>57559230
Yeah, as dead as Python.

>>57559206
It's obvious from the context he meant there's no way for the compiler to enforce it. Of course it's possible to write a program without leaks, ffs. Who would suggest otherwise?

D looks great, much better then rust.
shame about the gc and some other design choices.

>>57559259
Doesn't Ada ensure a program will never run out of memory at compile time? That implies it detects possible leaks.

>>57560883
Not with regular Ada, you can only catch stack overflow's from exceptions and failed allocations. Spark probably has a tool for it, since recursion is not allowed.

>>57558337
I mean, call me dumb, but I always found it fun to port an application to DOS or a 3DS or a printer. Lack of portability is kind of a deal breaker.

>>57549050
Crystal is never going to happen though, because it's just different enough from ruby to feel like an abomination to a ruby programmer, and nobody else in the space is going to want that syntax.

>>57561403
>I'm an ignorant faggot.

You know what, there are tons of languages inspired by ruby syntax.

>>57541726
just wrote a multithreaded hello world in rust.
can confirm it is better than c++

>>57561452
>jellyfag at D's relative popularity to Crystal

You're just jealous that D hit #21 on the TIOBE Index for November 2016 while Crystal is relegated to the 51-100 range of "unimportant bullshit nobody would ever use" along with VBScript and whatever the fuck Visual FoxPro is.

Oh God .. I did not know about the SJWs lmao https://www.reddit.com/r/MozillaInAction/comments/3ij392/rust_developers_discuss_rephrasing_the_dining/

>inb4 reddit
just googled that shit

>>57545798
>compile times over 1s in current year
wew lad

bamp

>>57562448
An second long compile can save an hour of debugging

>>57561358
Not dumb at all. But you can do all that in rust right now, actually. You can also compile to AVR micro-controllers and the like. When I'm saying outmoded an esoteric I mean 1960s mainframe. Modern hardware just isn't that much different from each other these days.

>>57541726
It has very little libraries, while C++ has pretty much everything past a certain level. I've seen some people mess around with making UI frameworks for it. But why would someone use it in industry to make a game for example, if they have no library, engine or experienced devs to make a game with it? In addition the entire project still needs a lot of development time to be up for that.

>>57564477
But there ARE libraries. C can be used rather easily.

>Rust is facing some interesting challenges ...1. A disharmonic personality. Reading any amount of Rust code evokes the joke "friends don't let friends skip leg day"
RUST BTFO BY THE CREATOR OF D

>>57554107
Also any work on embedded systems and other situations with limited memory. GC affects more than just speed, it also takes control from when and where your memory can get allocated.

I'll admit though in most cases it won't matter and a GC is preferable as at least half of all programmers shouldn't be trusted with freeing their own memory.