Google polishes Chromium code with a layer of Rust

Google plans to support the use of third-party Rust libraries in its open source browser project Chromium, a significant endorsement of the programming language and its security characteristics. In a blog post published on Thursday, Dana Jansens, from the Chrome security team, said Google's software engineers have begun …

[Anonymous Coward]

No Rust please

Rust? No thanks! The time has come to remove Google Chrome from my pcs.

[Hubert Cumberdale]

Re: No Rust please

That's your problem with Chrome?! There are so many other reasons not to use it.

[merlin27]

Re: No Rust please

Congrats on removing spyware from your PC. So which browser are you going to move to?

Brave? Thats based on Chromium which will contain Rust.

Firefox? It'll definitely have rust.

Safari? Good luck using the browser Apple crippled so PWAs won't compete with apps on iOS.

Unfortunately for you no amount of FUD will stop intelligent people from using Rust to write safer code.

[bazza]

Re: No Rust please

Clarification : Firefox is already using Rust.

If someone is planning on avoiding Chromium because of Rust, there is not many alternatives...

[Hubert Cumberdale]

Re: No Rust please

I just have no idea why anyone would really care about the language used to build an application of they don't have to actually work with it...

[bazza]

Is Bjarne Stroustrup Right?

Genuine question - I don't know. Bjarne Stroustrup is a clever guy, I just need help absorbing the full meaning of what he's saying about modern C++.

I've had a look at the materials produced by Bjarne Stroustrup in saying that modern C++ can match Rust for memory safety.

I ask specifically about the exclusive mutability that Rust has. I'm not entirely clear from what Bjarne Stroustrup has emitted exactly how that is recreated in C++. Are we talking about unique_ptr? And does that completely match Rust's exclusive mutability?

From what I've read by Bjarne Stroustrup, it seems that he's basically saying that using some of the modern features of C++ and being disciplined (no cheating, use a linter, etc) and you can achieve the same end result.

For me, the advantage of Rust would be that provided the unsafe keyword didn't appear anywhere the code review burden is a lot lower. If one set up a C++ writing environment and checking toolset to enforce the same discipline and rigour, fine, but it sounds like making that flaw tight is tougher.

It could be quite interesting if there were a flavour of c++ compiler that applied Bjarne's prescriptions.

[Anonymous Coward]

Re: Is Bjarne Stroustrup Right?

Of course C++ can. Achieving it is dependent upon extensive knowledge of the APIs you are using, their data structures, interactions, permissions and record locking for example.

Bjarne's C++10 book is a necessarily heavy affair that describes features relatively extensively, but not so good at describing good practise. And then the dive into mutexes is unintelligible unless you actually live with those issues.

Rust has certain features that help, but it cannot address the library dependency, or the inevitable C or C++ like interfaces up to the host OS.