Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The Pixelbook is being used to test Google's Fuchsia OS

androidpolice.com

Post reply on HN
Print view View original

81–90 of 109 posts

Re: The Pixelbook is being used to test Google's Fuchsia OS

#81

Post by hacknat »

openfuture wrote: post #8
jkmcf wrote: post #2

Aside from control, what does this provide over Linux? IMO, the Linux desktop problem is due partially to relying on the antiquated X Windows system, but mostly a lack of funding a good vision, e.g. Unity. After growing up with OpenLook, then Motif, the plethora of Linux window managers, and unhappy years with Windows, I noticed most of the elegant apps were being written for OSX. I’m not sure how much of this is due…

Capability based security and microkernel, both pretty substantial differences.

Also seems like Magenta doesn’t have users and groups as a first class construct. It looks like there are primitives to create them in user space, but the kernel only has one construct (Jobs) by which security is administered, rather than the many different security paradigms of a nix system (i.e. a process can have access based on what user it’s running as, uid not, SecComp control, etc). This makes security much easier to manage at the kernel level as all security granted to processes is explicitly granted rather than inferred, as is often the case in nix.

Honestly, I think the multi-user assumptions that *nix started with are largely irrelevant now. Most people don’t have multi-user needs on their devices (you can’t even do it on iOS), and even servers are moving towards single-user constructs with containers, etc. I think the an operating system built around a multi-user model will be viewed as the edge case in the coming decades rather than the norm.

Re: The Pixelbook is being used to test Google's Fuchsia OS

#82

Post by dcow »

cbhl wrote: post #9
jkmcf wrote: post #2

Aside from control, what does this provide over Linux? IMO, the Linux desktop problem is due partially to relying on the antiquated X Windows system, but mostly a lack of funding a good vision, e.g. Unity. After growing up with OpenLook, then Motif, the plethora of Linux window managers, and unhappy years with Windows, I noticed most of the elegant apps were being written for OSX. I’m not sure how much of this is due…

Control is all of it. Security on Android is a joke for 80%+ of users. They can't run on the latest version of Android, because various vendors' drivers are in out-of-tree kernel patches that are un-upstreamable for non-technical reasons. (By comparison, Chrome OS is also Linux-based, but, IIRC, it requires all shipping devices to have drivers upstreamed.) Owning an OS with a stable device driver ABI would allow Goog…

But it's not just control, is it? Reading this sub-thread I'm not convinced some commenters have actually taken much of a technical look at Fuchsia. I'm not associated with the project but from what I understand there are plenty of technical merits:

1. Zircon, the kernel, is entirely capabilities based. No users, no groups. Those are higher abstractions built on the capabilities system. This is all around a win for security from executing 3rd party drivers to building a sandboxed multi-user system, not to mention is just generally more flexible and less error prone at the kernel boundaries and internally.

2. The WM is scene based. Applications draw to a scene which the WM composites and renders. This means a single draw list is sent to the GPU for each pass instead of many apps spamming their own updates. Also you can do cool things like have objects from one app cast shadows on another app.

3. The kernel API is not a bloated mess and will not become one. Keeping the number of syscalls on the order of 10s not 100s is a design goal. You can see the commitment to this in the way the project is structured. Which brings up:

4. Testability. The project was designed with testability in mind. Not Enterprise Java mock and verify the world shit. Just good practice separation of responsibility between the different layers and enforcement of a sane unidirectional dependency relationship from lower to higher layers. Good luck testing the Linux kernel.

5. The system will update from a single image rather than relying on N parties to pull new patches into their distro and update and test everything N times before releasing to people. Again layers.

This isn't all new by any means, but it's a pretty practical approach to an OS in 2017. This isn't an "academic microkernel", and also decidedly not *nix.

Enough evangelism. Point is, the project is not solely designed to give the big G more control over a platform. Not saying that is not one goal, but technically this thing checks out.

Re: The Pixelbook is being used to test Google's Fuchsia OS

#83

Post by fabrice_d »

sebe wrote: post #72
cmurf wrote: post #38

From Wikipedia: Chrome OS is an operating system designed by Google that is based on the Linux kernel and uses the Google Chrome web browser as its principal user interface. OK so why not Google Chrome web browser on top of Fuchsia, instead of Linux kernel and the usual user space stuff? Google can call that retrofit anything they want, so it could still be Chrome OS.

Fuchsia can display web content, it has a web view that currently webkit based not chrome/blink. I think the plan is to use chrome/chromium web browser, once/if arrives in fuchsia,that will be interesting.

There is actually a fuchsia port in the upstream chromium tree.

Re: The Pixelbook is being used to test Google's Fuchsia OS

#84

Post by endlessvoid94 »

rcarmo wrote: post #61
ricardobeat wrote: post #13

Earlier quoted context omitted.

Elementary is the OS you're looking for: https://elementary.io I wish it had more adoption. Right now there are not a lot of apps written using it's UI guidelines/framework as most projects are worried about portability and/or have moved to Electron. It is slowly getting better though, and seems to be developing an ecosystem geared towards quality as OSX used to be: https://medium.com/elementaryos/appcenter-spotlight…

I’ve been using Elementary for a year straight[1] and like it a lot, but I see it as Ubuntu with the “right” UX. (I also don’t see Linux developers flocking to it in droves, which is sad because it is a lot cleaner and easier to use than Gnome and KDE, IMHO, but then again Linux has always been more about diverging choices than unity—and I don’t mean the desktop environment here). The only thing Elementary needs to d…

I am so interested in this! Are you willing to answer some questions?

1. What hardware are you running? 2. Do you pay for apps? 3. Are you able to use this as your primary machine? 4. What kind of work do you do?

Re: The Pixelbook is being used to test Google's Fuchsia OS

#86

Post by gkya »

CharlesW wrote: post #17
Mister_Snuggles wrote: post #14

Earlier quoted context omitted.

One feature that differentiates Linux from other systems is the sheer amount of choice that you get. For example, with Linux I can choose GNOME, KDE, a selection of other desktop environments, or I can choose a simple window manager like twm and sort of roll my own environment. Or maybe I could set my system up so that it runs without a window manager and just gives me a full-screen emacs environment. With Mac OS, fo…

> One feature that differentiates Linux from other systems is the sheer amount of choice that you get. It's a difference but not necessarily a feature if, as a user, you don't have the background to make informed and safe decisions about those choices.

Why would everything have to cater to the lowest common denominator on the scale from computer-novice to power-user? Those can just use Windows or Mac OS, or maybe Ubuntu. But why would their use cases invalidate mine?

Also, IDK how things are on Apple computers but on Windows no two apps look similar, so the common user should not have more difficulty with diverse UI paradigms on current Unices than on Windows.

Re: The Pixelbook is being used to test Google's Fuchsia OS

#87

Post by lucideer »

dcow wrote: post #82
cbhl wrote: post #9

Earlier quoted context omitted.

Control is all of it. Security on Android is a joke for 80%+ of users. They can't run on the latest version of Android, because various vendors' drivers are in out-of-tree kernel patches that are un-upstreamable for non-technical reasons. (By comparison, Chrome OS is also Linux-based, but, IIRC, it requires all shipping devices to have drivers upstreamed.) Owning an OS with a stable device driver ABI would allow Goog…

But it's not just control, is it? Reading this sub-thread I'm not convinced some commenters have actually taken much of a technical look at Fuchsia. I'm not associated with the project but from what I understand there are plenty of technical merits: 1. Zircon, the kernel, is entirely capabilities based. No users, no groups. Those are higher abstractions built on the capabilities system. This is all around a win for s…

> Good luck testing the Linux kernel.

This particular line struck me more than the rest of your (very good) post. For those of us unfamiliar, can you recommend/reference critical resources on the testability of the Linux kernel?

Re: The Pixelbook is being used to test Google's Fuchsia OS

#88

Post by dmitriid »

jkmcf wrote: post #2

Aside from control, what does this provide over Linux? IMO, the Linux desktop problem is due partially to relying on the antiquated X Windows system, but mostly a lack of funding a good vision, e.g. Unity. After growing up with OpenLook, then Motif, the plethora of Linux window managers, and unhappy years with Windows, I noticed most of the elegant apps were being written for OSX. I’m not sure how much of this is due…

> I believe Google could do this.

I keep seeing people day this, and yet: Is there any evidence at all that Google can pull off designing something like this?

Re: The Pixelbook is being used to test Google's Fuchsia OS

#89

Post by andrewflnr »

lucideer wrote: post #87
dcow wrote: post #82

Earlier quoted context omitted.

But it's not just control, is it? Reading this sub-thread I'm not convinced some commenters have actually taken much of a technical look at Fuchsia. I'm not associated with the project but from what I understand there are plenty of technical merits: 1. Zircon, the kernel, is entirely capabilities based. No users, no groups. Those are higher abstractions built on the capabilities system. This is all around a win for s…

> Good luck testing the Linux kernel. This particular line struck me more than the rest of your (very good) post. For those of us unfamiliar, can you recommend/reference critical resources on the testability of the Linux kernel?

AFAIK fuzzing is the state of the art when it comes to testing Linux: https://lwn.net/Articles/677764/

Re: The Pixelbook is being used to test Google's Fuchsia OS

#90

Post by sitkack »

Linux is the last Unix and that is ok. Unix is a philosophy, not an implementation. Linux mocked Mach, but was usurped by the hypervisor, violently corralling Linux in a microkernel environment anyway. Lots of things are in the kernel that don't need to be, making them non-updatable, as someone else controls the keys. That past, present and future is awesome. But lets build from the past and build the future. Not saying Fuschia is it, but as Unikernels and Exokernels have shown us, Linux itself is just an app in the stack. Unix is a framework for running processes. Your domain problems are the real problems, the OS is an implementation detail.
Post reply on HN