Google has finally – finally – truly and honestly confirmed Fuchsia is a thing.
Fuchsia is a long-term project to create a general-purpose, open source operating system, and today we are expanding Fuchsia’s open source model to welcome contributions from the public.
[…]Starting today, we are expanding Fuchsia’s open source model to make it easier for the public to engage with the project. We have created new public mailing lists for project discussions, added a governance model to clarify how strategic decisions are made, and opened up the issue tracker for public contributors to see what’s being worked on. As an open source effort, we welcome high-quality, well-tested contributions from all. There is now a process to become a member to submit patches, or a committer with full write access.
In addition, we are also publishing a technical roadmap for Fuchsia to provide better insights for project direction and priorities. Some of the highlights of the roadmap are working on a driver framework for updating the kernel independently of the drivers, improving file systems for performance, and expanding the input pipeline for accessibility.
It has been a very, very long time since any of the major technology companies built a new operating system from the ground up. Windows 10 is Windows NT, a project started in 1989 and first released as Windows NT 3.1 in 1993. The Linux kernel was first released in 1991. macOS grew out of NeXTSTEP, development of which started in 1985, seeing its first release in 1989. These operating systems are old.
Fuchsia is truly new, and developed by one of the biggest companies in the world, and while Google has a spotty track record when it comes to corporate attention span, I doubt they’d roll out the red carpet like this after four years of sort-of-but-not-really open development if they intend to kill the entire thing two years from now. And even if they do – the code’s out there anyway.
There’s a guide on how to build Fuchsia and set up an emulator (for Linux and macOS), so you can start poking around today.
Riverlane, a spin off from Cambridge University, has a quantum OS called Deltaflow.OS. Camridge or Oxford (I forget which) have next generation experimental secure OS too.
https://www.riverlane.com/products/
Deltaflow demo for the National Quantum Technologies Showcase 2020
https://www.youtube.com/watch?v=bnpAXypi1kM
I guess I don’t see how a quantum OS relates to a more traditional OS like Fuschia. Still cool thought
I mentioned it because it addressed the issue of new operating systems Thom raised.
There are other new OS worked on by Oxford and Cambridge but they don’t see much life outside of the lab. Almost all current OS have a last generation security model and need rewriting from the ground up and vendors are not mad keen on busting their marketing hype and doing an Osborne to themselves.
I don’t have a problem with old OS. In some ways I wouldn’t complain if we went back to Gem or RiscOS with a bit of tidying up. I honestly cannot stand bloat.
There is a lovely simplicity to GEM.
Is riverlane a major tech company?
I hear people complain about the current OS’es being “old”, but I don’t hear a lot of people providing explicit details on how they could be improved, or much beyond generic comments like “better security” or “sandbox everything” without in-depth analysis on how or what impact it may have.
I don’t see perfection in computing and technology when I look at the world around me, but I also don’t see it crashing down. It all seems to be chugging right along, fine for the most part, using those “old” OSes. I guess I’m not convinced we `need` something new written from the ground up, or that it would really be all that much better.
friedchicken,
You aren’t wrong, our operating systems can likely continue to be used in perpetuity with just maintenance support at this point. But many of these systems were designed by engineers who just didn’t have as much experience or hindsight as we do. When you really dig into it, POSIX and win32s carry with them a lot of legacy baggage. Legacy decisions in the linux kernel have held it back. I’d highlight asynchronous programming as an area where support is hacked in and kludgy at best because the entirety of the kernel was written under different assumptions. The programming APIs aren’t always as good as they should be. The C programming language is still responsible for tons of bugs and exploits every year. A big reason we keep using old tools is because they are popular, which is often more important than merit. Another big reason is that there’s trillions invested in today’s platforms that nobody wants to pay to replace when it’s already “good enough”, this same sort of momentum has kept mainframes around forever.
The irony is that we’ll pay many times the price of replacement in in aggregated maintenance costs, but short term incentives leads project managers almost universally to push off long term investments in favor of short term fixes. We have the same problem in politics, Long term thinking puts the burden of cost on those currently in office but rewards a future administration, possibly from the opposing party. It’s why fiscal responsibility is so damn difficult to achieve politically, but I digress. The point is I think there’s a lot that can be fixed, but the world isn’t ideal enough a place for it to happen!
@friedchicken @alfman
I agree pretty much with everything both of you said. Both can be right. the old OS were good for their time but we have learned more and they are now used for different things so updating would be an issue. This is where bad decisions and technical garbage all begins to snowball so the new OS have their own problems. A middle way learning from both is possible. The analogy with politics is absolutely correct. Governance is simply another system and the same kinds of mindsets and concepts can be found in politic and admin as much as in technical areas.
C/C++ really irritates me. I personally feel C/C++ needs a major redesign and simplifying. The argument I’ve heard against this by vendors is compilers wouldn’t be able to compile legacy code which is of course complete nonsense.
What maybe is a surprise is why the OS is still a thing. I think it is the persistent general nature of computing devices i.e they insist on staying multi purpose, that makes the OS layer stay visible to the poor user. On any appliance with a limited function set the OS would vanish from the users eyes.
Weirdly, even the microprocessor is still a thing. Arguably nobody whomsoever interacts with a microprocessor anymore, yet it is still marketed as a feature to users. Honestly, even most professionals should not notice microprocessor attributes beyond the generic computing performance. It’s all hidden behind compilers. And I wonder sometimes if compliers even notice the finer details the silicon architects labored over. Unless of course they screwed up by creating vulnerabilities like searching protected memory using pipeline flush statistics. Probably did it deliberately to get noticed :-).
Well, AFAIK Fuchsia has at least 7 years under active development. Not only that. In a long term, Google want it to grow as an eventual substitute for Android.