Why do you guys make fun of Rust?

Why do you guys make fun of Rust?
What is wrong with it?

t. complete newfag

Other urls found in this thread:

github.com/rust-lang/rust/issues?q=is:open is:issue label:I-unsound sort:created-asc

Lurk more.

I only see you guys discuss the SJW boogeyman but never discusses the technical aspects of the language.

This is something i'd actually like to know about. It sounds like a huge security improvement over things like C and C++, but I imagine there may be a catch of some sort.

It's shilled to fuck, and all those claims are mostly unverified. They are all things that work in theory, and after Meltdown and Spectre there better be damn good proof that something is secure and not going to cause problems.

You can always tell when something is being shilled because it's really popular and controversial for a set amount of time, and then it suddenly falls off the internet. That's exactly what's happened with Rust and it's exactly how systemd got started.

For most critics here, that is the catch. Technical criticism tends to focus on instability and unsafety (which is not by design but due to compiler bugs), and long compile times, which also is something that can be fixed soon. IMHO what's much worse than that is Rust's type system which still lacks generics over constants, e.g. Matrix (see github.com/rust-lang/rfcs/pull/2000), and traits, e.g. Vec (a vector of anything that is an Iterator). Now this can sooner or later be implemented, too, but as this affects the language on a much more fundamental level, I wouldn't consider the language 'stable' nor 'as useful as C++' yet.

-It's sold has fail proof when it isn't and has proven itself to not be fail proof.
-It's VERY young
-Ada and D already exist and are more reliable
-It's shilled by horrible people who don't care about the technology and use a facade of caring about it to push their political agenda.

m8 you need to read more


But so is your criticism. So many strawmen in a single post.

There's no language that can protect from these two not even ada or D

Not the user you responded to but it his massively shilled, if you have been here the past two years the number of rust threads or pseudo rust thread have been the same amounts of web browser threads, plus the whole crustacean and supporters who shill it like if it was a new religion/cult is a fucking turnoff.
Has for "does it work" of course it works but it's young and ultimately has big errors and other problems that of course can always be fixed like any other languages before it like ada or D.

Most of the people here don't code

On 7billion people how much can code ?

Rust was originally shilled as a language that "fixes the problems" with C and C++. I know how to code reasonably securely in C so for me, that means fixing code readability. When I look at Rust code, it somehow manages to be even noisier on the eyes than C++.
To add to that, the cognitive load of writing even a single line of Rust code is through the roof.
It's as if it was made just to stress people out more and waste time.
I understand the benefits of lifetimes and ownership, but the cpu architectures themselves are flawed at the hardware level and that's something no language can fix. Assuming you know how to wield C and C++ properly and don't need ridiculous amounts of compiler errors to lead you to success, why not just keep using C and C++ ?

nice try rustfag
sage and report all rustfag threads

It's better than C++ (not really hard to do), but not than C. But yeah, if you do use it, you'll have to distance yourself from the gay community.

GC in the stdlib.
Go LARP elsewhere.

I just think Rust is too complicated for a general purpose language. As someone put it, the cognitive overhead every line has is just way too much to find it comfortable to use. It's also unlike C++, where things can get as hard as you need for specific parts of the project, but is generally comprehensible. Rust is hard all the time, and shit like vectors may need different type signatures depending on what you want to do with them, or worse, WHERE will you use them from. There is also the whole fucking type system, that may require nesting half a dozen types in a specific order you could achieve in two in a saner language; also, have fun unwrapping, or even worse, correctly handling every fucking possibility in a thousand levels of nesting. Also, did you just change a type signature because you realized you missed on a certain capability that you realized you needed for a certain function? Have fun rewriting your whole program.

Rust also has some gotchas with some of the features it teaches the user in its own fucking book. Enums are really fucking nice, but they never mention that defining struct type enums is akin to shooting yourself in the foot and nobody does it after their first time because accessing their contents is verbose as fuck. Then it has the whole deref operator inside match clauses that doesn't work like an actual deref operator but whatever. It also had some really basic features (like try-casting) as "EXPERIMENTAL DO NOT USE!!!!" months after the 1.0 release.

tl;dr mite become a nice language, but as is the language is amateurish and intentionally guides you towards bad practices that will make your code fucking unmaintainable.
The community is nice and helpful fuck off Holla Forums

The language itself is pretty good, I'd say it's better than C++ and slightly better than C, and it's getting better over time. Modern C++ is pretty good, but it doesn't matter because it still has to be backwards-compatible with the old code.

Yes, the syntax is terrible. Yes, the cognitive load is higher, but it's generally worth it. Just git gud.

Yes, there are some edge cases that are insecure, but it's better than the whole language being insecure, like C and C++.

The worst issue by far is the SJWs and their CoCs. One day, they will pay for their faggotry.

Not exactly true. docs.rs/typenum/1.9.0/typenum/
This is hacky and rapes compilation times even further though.
The size of the type has to be known at compile time. So if you want to have a Vec of things that implement a trait you have to use a trait object. This isn't a limitation of the language.

Wrong. Do you have any proof to back up your claim?
D has GC and is not memory safe.

Complete bullshit

Your whole post reads like you are a complete retard that has never used a statically typed language and decided Rust is shit because you couldn't understand it.
??? This makes no sense
Yes that is how strongly typed languages work
For example?
You don't unwrap in production unless you want your shit to crash.
What do you mean??? Are you trying to say that shit code is shit?
Again, that is how strongly typed languages work. If you don't want to deal with it use some shitty language that encourages runtime casting.
Every language has experimental features. What is your point? Also Rust doesn't have try casting. It only has statically checked casting.
Wtf? Now you are just trolling.
- Ashley Williams (Core Team, Community Team Leader)

Also this thread: reddit.com/r/rust/comments/7nx3cm/announcement_ashley_williams_joins_the_core_team/

Have YOU used Rust?
You clearly have not.

In C++, std::vector gives you a vector of smart pointers to T. It may look verbose due to the ise of the std namespace, but you can use std and forget about it. It's pretty clear what it does. Now, the equivalent in Rust is Vec. Rc provides the smart pointer, RefCell provides the mutability (because of course, mutability is a type; from a theoretical point of sense this is extremely cool, but it's just a pain in the ass in practice), and Box provides a pointer to heap because why would we couple that functionality into RefCell? But wait, if T happens to be a type implementing Copy (which is not Clone, because every single fucking small nuance spawns a new type; they can't affor suboptimal decisions after all the layers of pointer indirection they threw themselves into), which usually are primitive types and shit you will have to look in the documentation continuously? You could use Cell instead of RefCell, because OF COURSE there is a type for that. Except they are not the same and depending on whether you want to synchronize changes between different sections of your code reading the ssme object you may need to use RefCell over Cell, and also throw a couple more Rc in the type signature because why the fuck not. Not to mention naming conventions are overtly terse to overcome the fact they endlessly nest their types and having to linebreak less often during type definitions, and that makes what every shit does less obvious. The fuck is a Rc, and what's the difference between a Cell and a RefCell? I can kind of guess what a shared_ptr is, but Rc is unintuitive as fuck. You may also use Weak instead of Rc. What do these types do? Whenever you search when to use A or B, you usually get "they are totally different and have nothing to do at all" with a half assed explanation on why comparing things is bad that doesn't solve your question. Yes, you get used to it, but it's dumb.

Sometimes you do want your program to crash, but anyway, that's why I brought up the next point.
In the example above, Vec returns an Option, Rc may panic depending on how you access it (arguably not in our case, considering tbe other types are there to circumvent that) and RefCell may panic. And this is a relatively simple datatype for Rust, five levels of indentation or endless temporary variables are normal for Rust standards.
Except other languages don't need changing type signatures every time you realized you forgot a specific guarantee continuously and write thousands of lines of error handling code whenever you do.
It is dumb to have basic features as experimental features. It is simply not mature enough yet.
Don't remember exactly the function, but I think it may have been related to parsing or coercion. I just remember the generally taught and accepted function or trait simply assumed the input was valid and did not even assume it should return a Result instead of a T, and the trait that did was an experimental one. Which is stupid considering casting from an int to a C-style trait could fail just like any type with more possible values than those the target type accepts could and should fail. Although that was a year ago so I dunno.

Also, I must add the compiler has compilation stages and some really minor failures will halt the conpilation at all, while hiding more important errors in later stages that could have been detected even with the earlier failures. Tool support for Rust is shit, and their official vim plugins suck and don't even work, even though the fixes have been circulating around for a while but they can't be assed to accept the pull requests. I would like to shit on the type system again while we are at it, since it encourages thousands of indirections and RefCounting to the point your average program will be as slow as one made in a managed language,and the managed language will be more comfortable to use.
Go to their IRC and ask anything about Rust or programming in general. They usually answer pretty quickly. There used to be a claim that the whole Rust community was composed of retarded SJW who don't know how to program, but this is clearly not the case. Their Reddit is mostly cancer, but that's because it's Reddit.

Don't be surprised if you get banned if you go full kill all the niggers and the gays upon entering the chatroom, tho.

Box gives you a T on the heap. RefCell gives you dynamically checked borrow rules. Rc gives you reference counting. Vec gives you a Vector of Ts.
Yes please answer that question. Why should RefCell do a heap allocation?
Copy is a marker trait that signals that the type can be cloned by a memcpy. It doesn't "spawn" a new type.
I don't really get what you are rambling about. Maybe try learning some Rust first? Also if you use a lot of RefCell and Rc either you are doing something wrong or you should use a garbage collected language.
It literally stands for reference counted.
Have you tried reading the docs? doc.rust-lang.org/std/cell/index.html
If you don't know what weak in the context of reference counting means, maybe you should read a book about programming?
Vec is a type. It doesn't return anything.
Yes. It shifts the borrow checking to runtime.
If you say so.
Except they do.
The basic features are all stable though.
Do you mean TryInto/TryFrom? The Into/From traits are for infallible conversions. TryInto/TryFrom for fallible. But just because the Try* traits are currently unstable doesn't mean you can't have fallible conversions.
Yeah you are just retarded as fuck. Rust doesn't encourage ref counting nor indirections. It is you either writing shit code or using the wrong language.
Ok. But if I go full kill all men?

Yes, of course you can explain it. It's already coded, after all. Can you explain why is it so fucking overengineered compared to other languages, though?
Bitch, the "idiomatic" way to write n-trees is using linked lists. Even inside an arena that shit is going to be terrible.
Dunno. Why don't you try?

Ok. Could you please quote me where I did that?
That are two completely different thing. But I don't expect anything else from a faggot who doesn't know what weak references are. Go LARP somewhere else.
I don't have to. The leader of the Rust Community Team already did that. The Rust cucks responded with "stop harrassing her" and shut down all discussion over it.

Oh, what is that? A tree that uses linked lists for children and siblings. And just like that, you outed yourself as the real LARPer. Conversation terminated,

That's exactly what I mean with misquoting. Go read a book to improve your reading comprehension, nigger,

Yeah. That is not a tree. It is a Graph. Writing Graphs in safe Rust is hairy but Trees are no problem.
You literally said you didn't know what weak in the context of reference counting means. And just like that, you outed yourself as the real LARPer. Conversation terminated.

Keep shilling it here and we'll attach swastikas to everything rust to make the internet and FOX News think you're internet nazis!

here comes rustfag arguing with himself to bump his shit thread and generate false conversation

it combines all the worst ideas
that could have been solved instead if the programmer had spent a little more time on getting the fundamentals right

t. Holla Forums larper

It's syntax is perl-level of noise and It's downright unpleasant to program in. You will spend less time thinking about your actual program, and more time searching for the right "idiom" or mental contortion to tame the "borrow checker". Also, shit compile times ("incremental compilation" is supposed to assuage this, but I've only seen data from cherry-picked toy benchmarks to support that claim). I recommend it if you either:
a) don't actually enjoy the act of programming, but have an unhealthy single-minded obsession with MUH SAFETY that needs to be satiated.
b) are a hipster who wants to signal about being a big boy "systems programmer" while partaking in Mozilla's hip social justice community

Ada isn't a LARP, just because it's too hard for brainlets doesn't mean it doesn't exist and cannot be useful in the hands of a competent programmer.

Anti Rust shill spotted.

Lmao. Point at him and laugh.

Something tells me Steve is the product of a parent-child relationship..

Ok. Seems like you skipped the algorithms and data structures course in uni.

Not an argument.

There's two way to make massive shitposting in Holla Forums
Start a Rust thread.
Start a license debate.

Here's 2 of the many gripes I have that haven't been mentioned yet.
1. The borrow checker can get annoying from a USERs perspective. If I want to just make a small tweak to a program so that it works better for me, I sometimes run into the borrow checker. This means that I might have to restructure more code to just make the borrow checker happy. If I know that a change is safe, I don't need the compiler to complain that it isn't. I'd prefer if the patches I made were just a couple lines.
2. For some reason the Rust community seems to license almost everything under the MIT license instead of the GPL. It seems like they do not want their software to remain free by allowing other developers to take away the freedom aspect of the projects.

It's because rust is financed and shilled 100% by mozilla, a for profit company. It's in their best interest to get other companies locked into their shitty rust eco-system that they run and manage.

It threatens their oo-rah masculinity

Because it's natural and organic and gets slapped around every thread, like a benis.

You know which subset of graphs introduces directionality? :^)

directed graphs, duh

I'm firing up my meme machine and readying a queue of dank swastikas...

1. The borrow checker runs at compile time, so the USER doesn't have to care about it even if they're reverse engineering the binary. You're just a shitty dev that needs to git gud.
2. No, they want to make cargo usable to professional developers without some one line dependency of a dependency of a dependency suddenly GPLing millions of dollars and thousands of hours of investment. Rust's dynamic linking is complicated at the moment so even LGPL is difficult to comply with. I checked the dependencies of a game engine once and it had this very scenario: some snowflake had decided that their

But what if the user wants to make a modification to the program?
That's a good thing. Though it's a little misleading using the word suddenly since it only takes place when you want to distribute the binary.
There is nothing wrong with having to release your source code along with the binary. What if the users want to make small changes to it?
Those games could either remove the GPL software, or just do the right thing and become licensed under the GPL. With games it is actually not that bad becoming GPL since when you sell a game you are selling a lot of assets along with it.


why do you think programmers should give their labor away without remuneration but artists must be remunerated for their labor?

art doesn't run botnet on your computers.
it's not about the money.

The programmers made something and then their company sold it.
For example, a craftsman uses his labor to make a piece of art and then he gives that art away in exchange for money. The programmers also make something with their labor and then they would give it away in exchange for money.
Just like with code, I would prefer if they would give away the project files instead of just the final copy, but it's up to the company.

You could also imagine the opposite. What if all art assets were free, but the game binary wasn't. It's not like people would be able to enjoy your product by just having the models and not having anything to make them interact with each other.

Programmers have no value. Pajeet can do it for less.

Fuck niggers get AIDS, Steve.:^)

Some user put it this way:

Rust is a tool like any other, use it for good. It's a good language. Don't however, contribute in any way to the project or core team faggots with AIDS.

They're just a bunch of larpers.
Theres litterally nothing wrong with Rust.

It isn't though. doc.rust-lang.org/reference/linkage.html

disgusting tbh

what about 'impl Trait' or whatever that's called?

impl Trait is just another way to specify generic bounds in function signatures.
You can't have a Vector of structs that implement a trait because the structs could have different sizes. So what you have to do is use a trait object. That is a pointer to the struct and a pointer to the vtable.

Programmers produce reusable modules and use others' work in their work, and in turn receive developer support and feedback from users of the program. An artist produces unique works that aren't improved or used by other artists and thus stands nothing to gain by giving work away in the same manner.

Yeah, I used the wrong words. It's not a C replacement; too big and high level.


A language that is supposed to replace C++ should not be more complicated to learn/write than C++.
Just git gud at C++ then?

He meant those bugs are proof that even the things you think are absolutely safe and cannot possibly be unsafe can one day be found to contain one fatal flaw that makes the whole thing a mess.
So, he's right. Rust is way to young to go replacing anything. Formal proofs need to be produced first. Rust has to prove itself in practice, not only in theory.

C++ is memory unsafe though

I'm using Firefox Quantum right now.
Also I have replaced grep with ripgrep and ls with exa.

He meant those bugs are proof that even the things you think are absolutely safe and cannot possibly be unsafe can one day be found to contain one fatal flaw that makes the whole thing a mess.

So, he's right. Rust is way to young to go replacing anything. Formal proofs need to be produced first. Rust has to prove itself in practice, not only in theory.

Why are you reposting your post with reddit spacing???

What are paragraphs?

So is Rust and they still haven't bothered patching it. Face it the memory aspect is just for marketing which makes it a bad choice for production.
How many non-american, non-SF companies use Rust right now?

see pic

Rust is not memory unsafe. Stop spreading FUD.
It is not though. Memory unsafety is a huge problem in big projects. Not that you would know that of course. I'm sure your FizzBuzz implementation is 100% memory safe. But try writing something more complex before you talk shit.
There are a few in Germany.

Firefox Quantum is pretty damn fast and secure.

Separating long lines consisting of multiple sentences by a single line break doesn't make sense imho.

use furryfox

Show us.

Firefox quantum leaks even more memory than old firefox.



So there aren't any big players using rust? Sad.

Shut up retarded anti Rust shill. First you excluded USA and now you exclude everything that isn't a big player.
The list at rustjobs.rs isn't exhaustive. There are a lot more companies that are using Rust in Germany. They just didn't out on that list.
Source: Employed Rust developer in Germany

Anzeige ist raus Bernd

Viel Erfolg! Ich bin hinter 7 proxies.

user, Ich...

Nothing. But this is Holla Forums, where honest discussion on the technical merits of technology is not allowed. Muh feels or fuck off. this is the Holla Forums credo.

How so? It has no runtime, just like C (unlike C++), and real-world rust programs are faster than their equivalent C programs (see ripgrep, leaf or colemchyma for instance). "Big" doesn't really mean anything and it lets you do operations from as low-level as you need to high-level constructs which are faster and significantly easier to write. If anything it's what C should have been.

He meant those bugs are proof that even the things you think are absolutely safe and cannot possibly be unsafe can one day be found to contain one fatal flaw that makes the whole thing a mess.
So, he's right. Rust is way to young to go replacing anything. Formal proofs need to be produced first. Rust has to prove itself in practice, not only in theory.

Because judging by your response, you didn't read it or intentionally ignored the point. And because copying and pasting it inserted an extra line break there.

Are you mad? Rust has already proven itself in practice and nobody said that Rust would eliminate all seurity issues.

By "in practice" I mean formal proofs. If there are any, I'll be very interested to read them. Please throw me a link.

The only relevant discussion I've seen always devolves into "it's not necessary" and "sure, check out my blog where I explain it to you", but nothing academic.
For example: github.com/rust-lang/rust/issues/9883
I'll be very interested because I think Rust has potential, but as I said in my thrice-posted post, it's way too young and relatively unproven (your examples are not enough yet).

Please stop being clinically retarded thank you.


nod an argumend :--DDDDDD

Not an argument. Actual examples were provided. How about you debate them? Oh wait you can't because they're undeniable.

Formals proofs are "in theory" though.

Protip: Nobody cares about formal proofs. Companies don't pay you to jerk yourself off to some theory shit. They want you to get shit done.
But here, look at this: ralfj.de/blog/2017/07/08/rustbelt.html

error: (no arguments given)

How big is your company?
What kind of product does it develop?
In what capacity does it use rust?


No, I want you to verify your bogus claims.

How about you verify your bogus claims first?

How would answering those questions dox you? Stupid much?

High quality discussion.


Why would you say that to me?

Nice. Thanks for digging it up.

They both have a runtime that is the standard library (unless you statically link it) that is a lot bigger in Rust, of course.
Never said Rust couldn't be as fast as C.
Yes it does. It mean compiler can stay simple and not become stupid behemoths. Find me something as simple as tcc/pcc for Rust. In fact, there's only one compiler for Rust, LLVM.

Another thing is that C is probably a lot simpler when you want to do a lot of unsafe memory stuff. Stuff like pointer aliasing, for example. Don't get me wrong, Rust does seems promising (especially that nice ML inspired syntax) but it needs maturation and will probably never completely replace C.

Wrong: github.com/thepowersgang/mrustc
Also, LLVM is just the backend. There is currently another backend in the works: github.com/stoklund/cretonne

I see, it's still fat as fuck, though. And stop LARPing as this faggot, you'll have more serious discussions.

ML has a mostly clean syntax. Rust has more of a Perl/C++-inspired syntax that tries to use every character on the keyboard just because they're there.

In unsafe blocks, rust has the exact same operations as C, so that point is moot at best.
Neither have a runtime.

LOL. I didn't always LARP as Steve Klabnik and the "discussion" was as shit then as it is now.
Also you are the retard complaining about muh bloat. There is no reason to keep something minimal at all cost only to appease your autism.

How big does your project need to be before it's worth it learning Rust over C++?

If your project is big, you probably shouldn't use rust. Unless your time is worth little and you enjoy inordinately long compilation times, that is.
Using rust is never really worth it.

Incremental compilation and multiple codegen units are in the beta right now which means it will be in the next stable release.
Rust's incremental compilation model is vastly superior to C++'s btw.

HOw is it superior?

Here's that meme again. The only support for it I've seen are cherry-picked toy benchmarks based on recompilation after trivial changes. C++ modules will make rust compilation times look even worse when they're added in C++20 btw.

Educate yourself:

It hardly is a meme when it is working right now.

It's a meme because rust fanatics are way overstating whatever benefits it provides.
no u
The best part is that C++'s compile times (for normal code that doesn't retardedly abuse templates) are already better than rust's. And they're just going to get even better.

If you're new to programming should you learn c++ or rust?

Rust. Since it rejects programs that have memory errors, you get into the habit of writing memory-safe programs regardless of which language you'll use after.

The best part is that Rust's compile times (for normal code that doesn't retardedly abuse generics) are already better than C++'s. And they're just going to get even better.
Anyways, kill yourself anti Rust shill.

C++. Not only does rust's borrow checker impose a worthless cognitive load on a beginner, but it also prevents learning through mistakes about the underlying machine, meaning you will never be as good a systems programmer as someone who started with C++. C++ can also provide a more sane and gradual learning curve for certain concepts also used in rust, like move semantics.

Wrong. As long as overhead from the borrow checker exists, C++ will dominate in that regard. In other words, the only way for rust to catch up is to abandon its reason for existing.

Random thoughts about the evolution of this thread :

1. You're all just a bunch of non-rustaceans

2. The worst elements from the Ruby community, which was already a cesspool, moved to Rust

3. You will never make the ladies wet like Steve does

4. Steve is an insufferable personality yet he still has more success than all of you

5. You will never write the reference book for a language unlike he/them

6. A good sign you CS curriculum is weak : your operating system class is not in Rust



neither nor
you should pay attention to your data structures and algorithms class if you are in a strong CS curriculum otherwise this will be a Java enterprise class and then move straight to an academic functional language so you will become unemployable for your own good

t. NSA


That's a new one.
Also, you can use smart pointers, static code analysis and some basic discipline to write safe code in C++.

I win.
Have you considered making a C++ instead of shitting up this one? Kill yourself anti Rust shill.

So why not asm then? C++ is also way to high level. Kill yourself anti Rust shill.

The reason I won't use it is because it won't use the OS's native toolkit, whether that is GTK, QT, or Win32. Even fucking Java with SWT can do that. Consistency is important to me, and a application that sticks out like a sore dick has no place on my machine.

Modern C++ is not memory safe though.

C++ fag here, modern C++ is a trainwreck. Any spec past C++11 is cancer.

Prove it

Like pottering.

What's the point of arguing? Everyone here will just have an autistic shit fit when you point out anything, there's no point anymore.

For one, asm is not portable. Although, I do think writing asm is a good exercise.

It doesn't need to reach your retarded standards of perfection to be safe enough, though.

qt widgets bindings are hard

C++ is a shit language that gets even shittier with every new version. You literally can't defend this fucking trainwreck.
Now either leave this thread or kill yourself, anti Rust shill.

Then why does the new firefox leak memory out the ass?


There are portable assemblers as well as asm-like languages such as Z# or whatever that was called. You literally have no excuse. Not that I expect better from sophist prime the master of all illogic.

Defeats the point of writing asm in the first place.
Defeats the point of writing in higher-level languages.
Maybe you misunderstand; (zero-overhead) abstractions are fine as long as they don't try to make the underlying machine completely inaccessible, as rust tries to do.
wew lad

Literally any C/C++ codebase

Kill yourself anti Rust shill

Wrong. Any proof to back up your claim?

Ever heard of heartbleed? Goto fail? Ariane5/cluster? The two mars rovers? Maybe you should use google sometimes, or stop living under a rock.

So you say.
How about the fact that every problem rust hides under the carpet has to do with the reality of the underlying machine? e.g. those spooky scary segfaults
Of course, you can use unsafe blocks, but at that point you may just as well use C/C++.

No, not very modern.


I wasn't implying. I was stating a fact. There has been no attempt
to move the smallest parts of the ecosystem, to provide replacements
for base POSIX utilities.

As a general trend the only things being written in these new
languages are new web-facing applications, quite often proprietory or
customized to narrow roles. Not Unix parts.

Right now, there are zero usage cases in the source tree to require
those compiler tools. We won't put a horse into the source tree when
society lacks cart builders.

So rather than bothering to begin, you wrote an email.


Yes, now I am implying something: you won't bother to rewrite the

And I understand, why would anyone bother? It took about 10 years for
gnu grep to be replaced sufficiently well in our tree. This stuff
doesn't happen overnight.

However there is a rampant fiction that if you supply a new safer
method everyone will use it. For gods sake, the simplest of concepts
like the stack protector took nearly 10 years for adoption, let people
should switch languages? DELUSION.

In OpenBSD there is a strict requirement that base builds base.

So we cannot replace any base utility, unless the toolchain to build
it is in the base. Adding such a toolchain would take make build time
from 40 minutes to hours. I don't see how that would happen.

Are they POSIX compliant? No. They are completely different programs
that have borrowed the names.

By the way, this is how long it takes to compile our grep:

0m00.62s real 0m00.63s user 0m00.53s system

Does Cgrep compile in less than 10 minutes?

Such ecosystems come with incredible costs. For instance, rust cannot
even compile itself on i386 at present time because it exhausts the
address space.

Consider me a skeptic -- I think these compiler ecosystems face a grim
bloaty future.


So you're just shitposting then.

You could at least have the decency not to post the most basal of lies that can be discredited even by 5 years old children with an internet connection.

Post your smug faces to react to this

The conversation was about modern C++, so why the fuck did you post C-specific examples in the first place?

Let me check those goalposts:

Looks like the goalposts have stayed put.
No magic needed, just



The reddit spacing meme doesn't make sense to me. Why would you press space only once? When the fuck would you ever want a line of text to be on the next line, rather than having a space between them? That's not how paragraphs work. If anything, it would be reddit spacing if they're pressing enter 3 times, not 2 times.

Because an army of plebbit faggots injecting dozens of spaces into their blog posts makes threads long af to scroll through.

> As a response to this, Theo asked rhetorically "Where's ls, where's cat,> where's grep, and where's sort?", implying that noone so far bothered to> write implementations of even the basic unix utilities in such a> language.

Well, I fritter away some more idle minutes running anotherten or so iterations of running my shell script, adding andremoving switches, carefully checking that no commands arealiased, that my search path is correct, that I'm actuallyrunning grep on the correct files and that I am reallyinvoking grep without the "-h Do not display filenames"switch. I check the non-setting of "-h" many times, sincehaving it set would produce exactly the sort of lossage Iwas experiencing.Finally, about to tear my hair out, I happen to type "c-L"in the Emacs buffer in which the grep options are displayed.This recentres the window such that three additional linesappear at the top of the buffer's window. I read---------------------------------------------------------------------- want the filename to appear, use /dev/null as a second file in the list.----------------------------------------------------------------------ARRGGGGGGGGGGGHHHHHHHH.I scroll back and read the whole screen:---------------------------------------------------------------------- When any of the grep utilities is applied to more than one input file, the name of the file is displayed preceding each line which matches the pattern. The filename is not displayed when processing a single file, so if you actually want the filename to appear, use /dev/null as a second file in the list.OPTIONS -b Precede each line by the block number on which it was found. This is sometimes useful in locating disk block numbers by context. -c Display a count of matching lines rather than displaying the lines which match. -h Do not display filenames.----------------------------------------------------------------------The iron spike of enlightenment finally punches through intomy cerebellum "... ... ... so... so... so... youmean... there's ANOTHER reason why Emacs always appends/dev/null to its m-x grep command-line! ... Well f__k meharder again and again! ... errgghhhh... FEELINGS arecascading over me!!!"Incidentally (or not -- YOU be the judge!) this is EXACTLYthe kind of bullshite which makes the "ls" command lose sobadly (and which, incidentally, contributes to every unixFTP server flagrantly violating the FTP spec): the defaultfilename argument for "ls" (the current directory) istotally different from specifying "ls * .*" (which one mightthink would be the same thing) because bloody cretinous "ls"descends into subdirectories which are specified on thecommand-line, even those without a trailing "/" (even ifunix had some command-line convention to discriminatebetween "the directory" and "the files of the directory",which it doesn't) and even those which are a result of awild-card expansion (even if unix commands knew anythingabout wild-cards, which they don't.)(I won't even start to rant on the matter of needing ".*" inaddition to "*" to get all of the files in a directory. Iwon't. I won't. I won't.)Anyway, I now feel so much better for knowing why Emacs doeswhat it does.I'm now a grep wizard! I can probably get a job as ahigh-powered unix system administrating consultant, the sortwho just recently hosed three day's of my mail.

Nice meme, but I don't think C++ can even be compared to C at this point anymore.

The lines are too short to be called paragraphs. Look, they don't even wrap to a separate line.

Looks longer in the reply box.


It feels wrong to have it any size other than default.

security through obscurity

then just be conscious on what your post will look like after you post it.

github.com/rust-lang/rust/issues?q=is:open is:issue label:I-unsound sort:created-asc

the catch is that it's still not there yet, and the development team is too impotent to fix some unsoundness bugs from 2012 (it's been almost 6 years!):

oops I mean 2013 of course


it's not /b dumbass

sage negated





sage reapplied rustkike

Imagine if the core queers weren't wasting time on crap like buttfucking and microaggressions.

Imagine if the Holla Forums LARPers weren't wasting time on crap like browsers and language wars.

Same in MSM.

why is everything deleted?



whos this ashley person? a tranny?

No. She seems to be a real woman.

With a nice group of white knights


read this
if you still disagree you fucking woman hating racist mail the rust moderation team

So she never got reprimanded but you're no longer allowed to yell at her?
The issue has never been addressed but it's over now?

She wasn't serious about it so there is no need to reprimand her.
You are not allowed to yell at anyone in spaces where the CoC applies to.
The issue has been addressed.

I hate women.
j/k i wasn't serious about it :^)

If the roles were reversed and a man did what she did, he would've been thrown out of the industry so fast his head would spin.
The fags who drank the Kool-Aid aren't the ones you need to convince, it's everyone else who sees shit like this and thinks twice about using your language that needs convincing, and repeating the same line of bullshit over and over again isn't going to do the trick.

repeat after me:
I shall never post screenshots in JPEG

is it readable? yes
fuck off autismo


where the fuck do you think you are, fucking normie?


disregard trip, i suck cocks
mods pls delet

leave it to the turkroach to be even more retarded than the klabnik imposter


Beta cucks in search of a mother. Not like they want to fuck her look at her face

This. Double standard douchebags.

That's where vigilante justice comes in. See

to be fair, looks fuckable. not 10/10 of course, maybe 6/10

you can delete your own posts, dinghole

says the fag who doesn't consider how his mental diarrhea looks with a sane line length

No. Don't base your rating on its profile picture. Look its talks up on jewtube. 3/10 max.

This. Seamonster for sure.

Rust uses the method len to get the number of characters right? No, faggots in Rust made it the number of bytes. So rather than using size and making sense, they used len in a confusing way. They admit it was a mistake but won't fix it because Rust has to remain backwards compatible forever.

C counts the number of bytes too.

it's not wrong to give the length of the string in bytes, it's just not what your Pajeet Java brain expects because you're mentally inferious to the Rust programming Ubermen

see you in 2020. bye.

String in Rust's std are stored in UTF-8. That means that getting the number of codepoints in the string takes O(n) time. Getting the number of bytes is O(1). In the whole std len() is always O(1). So it would be confusing if it wouldn't be the case for strings.
If you want the number of codepoints do this: my_string.chars().count()

I'm another user, and you're still a faggot. If it looks like a fucking paragraph on my screen it is a paragraph.
user, I...

You are retarded. I did not mention aspect ratios.

Hey there cuckchan /g/ refugees! Please discuss Rust. Thanks!

Rust and Ruby start both with the same 2 letters.

Just saying.

You can't write Trust without Rust.

Just saying.

Kill yourself, reddit spacer.

bump XD

2 good names for fork langs if the SJW queers ever force it, Trust and Just.