News for hackers. Stuff that matters.
Re:Microsoft's Linux Kernel by cameronbrown on Friday June 28, 2019 @07:32PM (#20309711)
This is probably an unpopular opinion but.. there is a lot of value in having multiple competing kernels. In the same vein as browser engines, any monoculture is a bad idea. Not to mention there's been a lot of good work done on the NT kernel itself. Under the hood it's pretty advanced, even with all the user-space cruft on top.
Re:Microsoft's Linux Kernel by AnIdiotOnTheNet on Friday June 28, 2019 @08:05PM (#20309884)
Sadly, much like with the web, things have evolved such that it is virtually impossible to create a competitor due to all the accumulated complexity. The driver problem on the PC platform is pretty much insurmountable at this point, so any competitor would have to begin life, and gain massively in popularity somehow, on a much less flexible hardware platform.

It's a shame and makes me sad for modern computing.

Re:Microsoft's Linux Kernel by mamcx on Friday June 28, 2019 @09:25PM (#20310219)
I have think the same but maybe TODAY is not as hard as before:

1- We have 4/5 mainstream OSes today: Win/OSX/iOS/Android + Linux. In the case of iOS/android: Some users alread switch between the 2.

2- If it have apps is almost all people need. Drivers are not THAT significant IMHO.

3- Chromebook is a thing, despite it sound stupid: A OS that is just a browser.

Web is the 6 truly mainstream "OS".

4- What people need on drivers? USB (and with usb-c), hdmi, bluetooth (maybe?) + wifi. With this alone and you get even better than iPad: Monitors, wireless mouse, keyboards, external storage, and by extension you catch the rest.

5- Printers? By proxy of iOS it can work full wireless with no drivers on device.

So I think a true desktop OS could launch with less worry about legacy hardware.

Where IS the problem is the apps. It need to be better than chromebook, have very great first party apps that cover basics and hopefully, a VM layer so you could run linux and catch the rest.

Re:Microsoft's Linux Kernel by AnIdiotOnTheNet on Friday June 28, 2019 @09:34PM (#20310247)
> What people need on drivers? USB (and with usb-c), hdmi, bluetooth (maybe?)

Do you honestly think there is a single USB-C driver that covers literally everything that can connect over USB? And anything with an HDMI port uses the same driver!?

Re:Microsoft's Linux Kernel by mamcx on Friday June 28, 2019 @09:49PM (#20310310)
No, but I suspect the array of devices is more narrow than suspected?

I think that nail the software (app + dev experience) is a bigger challenge and priority than worry about the (external) hardware.

P.D: One thin I forget to articulate is the possibility to leverage linux as a bridge for drivers (possible)? so the new OS ship a smallish linux just for get compatibility.

Re:Microsoft's Linux Kernel by AnIdiotOnTheNet on Friday June 28, 2019 @10:13PM (#20310375)
Then you're just making a Linux distribution, or something like Android I suppose. You may as well say extending Chromium would solve the web monoculture issue. That isn't how it works.

> I think that nail the software (app + dev experience) is a bigger challenge and priority than worry about the (external) hardware.

You are ludicrously wrong. Using the Linux kernel as an example, because drivers are in-tree, as of 4.19 (late 2018): https://raw.githubusercontent.com/udoprog/kernelstats/master...

As you can see, drivers are the vast majority of the kernel. And Linux's hardware support is often called out as a reason people use Windows or MacOS!

Re:Microsoft's Linux Kernel by mamcx on Saturday June 29, 2019 @12:51PM (#20313542)
I stand corrected :)
Re:Microsoft's Linux Kernel by toyg on Sunday June 30, 2019 @06:57AM (#20317589)
Considering that the alternative to “OS in a browser” these days seems to be “every app is a (different) browser” (Electron), the Chromebook concept is not that stupid.

It’s sad that commercial realities are pushing people to throw away 30 years of progress on desktop UIs, but here we are.

Re:Microsoft's Linux Kernel by cacarr on Monday July 01, 2019 @01:11AM (#20322618)
> 3- Chromebook is a thing, despite it sound stupid: A OS that is just a browser.

It's not really "just a browser" these days. Runs Android apps, and regular old GNU/Linux desktop apps.

Re:Microsoft's Linux Kernel by pjmlp on Monday July 01, 2019 @09:15AM (#20324932)
Only on some selected devices.
Re:Microsoft's Linux Kernel by cameronbrown on Friday June 28, 2019 @08:23PM (#20309963)
I wouldn't be so cynical. As with all things, tech comes and goes in cycles. I think with the limit to Moore's law rapidly approaching for current silicon, there's a good chance that efficiency & simplicity will become front and centre again at some point.

Not sure if this will mean new architectures, operating systems & better tools.. but one can dream!

Re:Microsoft's Linux Kernel by AnIdiotOnTheNet on Friday June 28, 2019 @08:40PM (#20310032)
It is difficult to continue to hope for that while watching people cheer on additional layers of complexity like the garbage fire that is the modern web.
Re:Microsoft's Linux Kernel by DoreenMichele on Friday June 28, 2019 @09:08PM (#20310136)
Try roasting marshmallows over it once in a while. See if that helps you feel better about it.
Re:Microsoft's Linux Kernel by efreak on Saturday June 29, 2019 @02:37PM (#20314057)
I did. I abused the history API on my website when it was new, and then they went and fixed it. As it turns out, sometimes people care more about the marshmallows than they do about the fire.

I don't have the website anymore, but the basic idea is detailed here: https://grack.com/blog/2011/03/07/abusing-the-html5-history-...

Re:Microsoft's Linux Kernel by DoreenMichele on Saturday June 29, 2019 @03:19PM (#20314243)
Less flippant answer: To some extant, I feel it is on me to try to access good things via the Internet, figure out how to interact with it reasonably safely and satisfyingly and not get overly bent out of shape about the inevitable problems.

All things have their downsides. There are no perfect things that are nothing but goodness and light.

As they say: Sunshine all the time makes a desert.

Re:Microsoft's Linux Kernel by sjruckle on Friday June 28, 2019 @09:06PM (#20310126)
Yes, but if garbage never caught fire nobody would bother inventing better fire suppression techniques.

Also, nobody ever saw a fireman by hiding inside with their window blinds shut. Take a peek outside, what you see might bring you hope.

Re:Microsoft's Linux Kernel by meruru on Friday June 28, 2019 @10:20PM (#20310393)
We don't need NT to have competing kernels. All of the following have their own kernels:

  + FreeBSD  
  + OpenBSD  
  + NetBSD  
  + DragonFlyBSD  
  + Minix  
  + Haiku  
  + Redox  
  + Plan 9  
  + SeL4  
  + Fuchsia
And many more.
Re:Microsoft's Linux Kernel by cameronbrown on Saturday June 29, 2019 @02:43AM (#20311147)
Dozens and dozens of UNIX kernels (with Fuchsia being the exception I believe..? Not sure). Yes I know kernels have far outgrown POSIX, but it's still a form of monoculture. Diversity in all dimensions matters imo.
Re:Microsoft's Linux Kernel by meruru on Saturday June 29, 2019 @02:57AM (#20311179)
I'm not sure anything below Minix can be considered a UNIX kernel.
Re:Microsoft's Linux Kernel by __s on Saturday June 29, 2019 @04:46AM (#20311514)
Unix kernels could use more cohesion, not less, in terms of API. Even if that API is implemented mostly in userspace

View the kernel API like HTML. Browser diversity is good, but they need to be implementing the same interface if the smaller players want to support any hardware that works with Linux

Re:Microsoft's Linux Kernel by __d on Sunday June 30, 2019 @08:13PM (#20321454)
Not to disagree with your point, but Plan9 is not really a Unix kernel, despite being written by some of the same group at Bell Labs.
Re:Microsoft's Linux Kernel by CameronNemo on Monday July 01, 2019 @11:42AM (#20326454)
Less than half of those kernels are derived from UNIX. Do they replicate the same interfaces? Sure. But so does NT.