Google's Fuchsia OS Magenta Becomes Zircon

[M@GOid]

People tend to forget that Google has a long list of projects and services that started as the latest great thing, only to die in obscurity some time later. This Fuchsia can go somewhere? Sure it can. But right now, it is more a research project than a future Linux killer.

[danielnez1]

I find this project interesting, though like others have said I don't see Android or Linux going away anytime soon. However it is good to still innovate and try out new things.

[nomadewolf]

I expect it to replace the Linux kernel not only in Android but in desktop Linux as well. And after that the Linux kernel may serve the niche needs of computing devices where microkernel is not welcome.

I guess it depends a lot on what type license Google uses...
But for one, i'd much rather see a project like RedoxOS or GNU HURD succeed. Mostly RedoxOS because it's not owned by a corporation with comercial interests and it's written on a programming language that's supported by a company without comercial interests.

[LEW21]

It's high time to replace Linux with a modern, safe microkernel. Linux contains too much stuff and has too large attack surface, which makes sandboxing and containers difficult to implement securely. And it contains a lot of concepts - like user accounts - that are not relevant or not flexible enough in some cases - like containers. Microkernel will make it possible to turn them off, limit access, or switch for other implementation on a per-process basis.

Of course the replacement of the largest and most popular kernel is a giant project, that only the biggest IT companies can manage to do. Fortunately, Google accepted this responsibility.

[tildearrow]

Typo:

it's low-level Magenta core has been renamed to Zircon.

[starshipeleven]

I expect it to replace the Linux kernel not only in Android but in desktop Linux as well. And after that the Linux kernel may serve the niche needs of computing devices where microkernel is not welcome.

Rampant baseless speculation.

[starshipeleven]

With Zircon IHVs will have to support 1-2 versions(just like with Windows where this number is 3), while with Linux they have to support close to 12 (currently maintained versions). Guess which kernel they will support more eagerly.

FYI: with AMDGPU that is merged with mainline their proprietary driver can track easily the Linux kernel.

[starshipeleven]

I guess it depends a lot on what type license Google uses...

And you even need to ask? It's the usual opensource permissive license (aka closed-source firendly).

[starshipeleven]

Of course the replacement of the largest and most popular kernel is a giant project, that only the biggest IT companies can manage to do. Fortunately, Google accepted this responsibility.

The amount of rampant speculation is rising.

[jrch2k8]

This probably will not have much to do with Linux as a kernel but with Linux as drivers, upstream kernel require certain level of quality to accept drivers in the main tree and historically ARM hardware drivers tend to be fast coded trash(specially true with chinese brands) and most don't like FOSS at all either.

I will assume zircon will behave more like windows, Google will deal with the basic micro kernel and IPC(and might be FOSS) but let the drivers be closed with some sort of sandboxing so ARM providers can keep pushing trash unstable code fast and be happy while android can update more often(in theory at least) and claim they have as much penetration per version as Apple does while removing themselves from any responsibility outside the base OS, i would guess chinese phones security is going to be even more awesome(LOL).