Redox has published the summary of development covering February, and there’s quite a few interesting leaps forward this month. First and foremost, the operating system got a major file read/write speed boost by implementing records in RedoxFS. The migration to UNIX-format paths is ongoing, Boxedwine is currently being ported, and more and more programs are getting ported, including complex applications like Audacity, Celestia, KiCad and Neothesia.
There’s a lot more this month, so be sure to read the whole report.
I’m really excited about Redox. Linux compatibility without the reliance on the whims of the single individual at the top. It goes to show how, with some effort and backing, Linux could be quickly usurped
“the whims of the single individual at the top” – that’s not a thing. Any company can fork Linux entirely and do whatever they want. Some do. Many maintain their own custom kernel patches and use them in their own distros or kernels. Most rely on “the whims of the single individual at the top” because they trust that specific actor to make good decisions. Linux is developed in a web of trust system.
IMHO, this is a massive misread of the way things actually work in Linux kernel development.
I’m also excited by Redox, because I like alternative operating systems. More is better!
“Linux compatibility without the reliance on the whims of the single individual at the top. It goes to show how, with some effort and backing, Linux could be quickly usurped”
Yeah I’m not sure where you’re going with that but as a Linux user for 25 years and without intimate knowledge of how the people behind Linux do what they do these days, I can only be forever greateful and appreciative of what they achieved and continue to do behind the scenes to allow me to work and play Windows and Microsoft free. Couldn’t care if “the single individual at the top” is good/ bad/otherwise.
Likewise, I’m looking forward to what Redox achieves. It looks like a fantastic OS moving in the right direction!
Redox may have laid an intriguing cuckoo egg in Linux’s nest. I think this is a better approach than attempting to use rust in linux, but has it gone far enough?
Look for instance at the R9 kernel for plan 9, currently the main focus of what was HarveyOS distro. Writing plan 9 (starting with the kernel) in Rust seems eminently sensible because the OS is quite small and there is the opportunity to make genuine improvements in the rewrite. I would say this is not the case with mainstream free Unix’s which already contain countless developer-years of refinement..
If I might be so bold as to suggest, OS news seems not to cover the Plan 9 scene as thoroughly as it deserves.
Squizzler,
I agree completely that a new clean rewrite is better suited for picking up rust and really embracing safe code designs from the start. That said, I think the reason linux is significant for rust is not because it’s ideal for rust development, in fact it’s far from ideal and the C code base will be an ongoing impediment for safe rust, but because linux is a significant player such that even a little bit of rust adoption by a major player could prove more influential than a niche OS that fully embraces rust. This is not fair, but there’s not much a small player in the long tail can do about that.
Years ago I looked upon plan 9 as a sort of “unix v2”. They cleaned up some of unix’s problems, but plan9 never became more popular than it’s predecessor and at this point I don’t see much change happening in a mature market.
I hardly think that the so-far low degree of success of Plan9 counts against it being reported here. OSAlert is supposed to report on niche OS. There should be at least an item on each of the 9Front releases. The Register (UK IT news site) manages that.
THEir correspondent for such things, Liam Proven, recently presented at Fosdem on Plan 9 as a way to address linux bloat serialized (along with a related previous Fosdem presentation), link below.
https://www.theregister.com/Tag/One%20Way%20Forward