Rust-Written Redox OS 0.8 Released With i686 Support, Audio & Multi-Display Working

[phoronix]

Phoronix: Rust-Written Redox OS 0.8 Released With i686 Support, Audio & Multi-Display Working

After more than a half-year of development work, Redox OS 0.8 released today as the newest version of this from-scratch, Rust-written open-source operating system...

https://www.phoronix.com/news/Redox-OS-0.8-Released

[zexelon]

RedoxOS has me excited! They are designing the OS around some really unique paradigms that on the surface at least look to take the best of POSIX and dump the parts that no longer make sense! Its also being built as a unified experience from low level to graphics. Sure they may have bitten of to much innovation all at once and none of it will matter much without a way to get solid hardware support working.

All that said, they have made really impressive progress!

[Shnatsel]

USB isn't supported by Redox yet. You have to use PS/2 mouse and keyboard. That's the issue Michael hit; it's even mentioned in the announcement.

[rclark]

How I see it is... This really will help test/'validate Rust for 'real' use. Hopefully a Rust ANSI standard will be coming soon that developers can compile against as not to end up chasing their tail so to speak when Rust language developers decide to break things for a better mouse trap.

Neat project!

[timofonic]

It's stupid they support 32bit hardware in a modern operating system aimed at innovation, they are wasting efforts instead improving essential hardware support such as USB.

I had high hopes in RedoxOS, but it seems it will end as ReactOS. An eternal experiment, to say the least.

[mirmirmir]

These people and their fans really underestimate projects they are working on

[kvuj]

RedoxOS has me excited! They are designing the OS around some really unique paradigms that on the surface at least look to take the best of POSIX and dump the parts that no longer make sense! Its also being built as a unified experience from low level to graphics. Sure they may have bitten of to much innovation all at once and none of it will matter much without a way to get solid hardware support working.

All that said, they have made really impressive progress!

If you want to see a really interesting OS with their own C++ library/PDF renderer/GL library/UI library/Web browser which can load discord, check out serenityos!

I think Michael should sometimes cover some update of it.

[ssokolow]

How I see it is... This really will help test/'validate Rust for 'real' use. Hopefully a Rust ANSI standard will be coming soon that developers can compile against as not to end up chasing their tail so to speak when Rust language developers decide to break things for a better mouse trap.

To be honest, I think people have an idealized perception of standards, ignoring things like:

1. How much developer time crafting and extending them draws away from development.
2. How compiler vendors extend them anyway (It's not as if the Linux kernel is written in ANSI C. It's written in GNU C. LLVM Clang support was primarily achieved by extending LLVM Clang to implement the necessary parts of GNU C.)
3. How compiler vendors fail to comply with them, either accidentally or intentionally (eg. How GCC and LLVM Clang can't agree on the ABI of __int128 despite it being specified in the relevant AMD64 SysV ABI PDF.)

Historically, programming language standards have come about for one of two reasons:

1. To bring things back into agreement after multiple vendors arose and failed to achieve intercompatibility (what happened with things like C, C++, Fortran, etc.)
2. As a political ploy to make the language look more legitimate to government or enterprise customers than its competition, regardless of their relative technical and interoperability merits. (what happened with things like .NET and, though it's not a programming language, Microsoft Office Open XML.)

Sure, there's work on the Ferrocene Language Specification to provide a paid-for verified compiler build for breaking into the market occupied by things like SPARK and MISRA C but, for the language in the broader sense, the Rust developers are already going above and beyond to avoid breakages.
I doubt they'd agree to any kind of standards that would pin them down further than they already do voluntarily through their RFC process and "so resource-intensive it requires donated Azure time from Microsoft" regression suite and crater runs and the like, and then you'd just wind up with "The Rust® compiler (and yes, like Linux® and Firefox®, Rust is a registered trademark owned by the relevant foundation) has chosen not to implement the Puccinia language spec" and the spec would either become a rubber-stamp for what the Rust compiler implements or wither and die as the crates.io ecosystem continues to follow the Rust compiler.

[Quackdoc]

It's stupid they support 32bit hardware in a modern operating system aimed at innovation, they are wasting efforts instead improving essential hardware support such as USB.

I had high hopes in RedoxOS, but it seems it will end as ReactOS. An eternal experiment, to say the least.

actually reactOS has been quite nice to me for using inside of VMs. I actually use it for retro gaming now. since it supports 1bit games via winevdm and win9x games pretty well

[cb88]

RedoxOS has me excited! They are designing the OS around some really unique paradigms that on the surface at least look to take the best of POSIX and dump the parts that no longer make sense! Its also being built as a unified experience from low level to graphics. Sure they may have bitten of to much innovation all at once and none of it will matter much without a way to get solid hardware support working.

All that said, they have made really impressive progress!

The problem with that... is it because just another almost unix that would be incompatible enough to be annoying.

And a better Unix was already written as Plan9... so frankly you gotta either do better than Plan9 or you aren't truely trying to one up unix you are just riding on its coat tails.