[endorphone]

In that case it wouldn't have replaced Android at all -- It would have simply replaced the Android kernel, which happens to be Linux. And at that point you have to ask what is gained, and at this point the answer is "nothing".

[wolfgke]

> And at that point you have to ask what is gained, and at this point the answer is "nothing".

Not having to care about the permanent breaking of internal kernel APIs is not "nothing".

[endorphone]

How, exactly, will they "not have to care"? The identical ramifications occur with Fuscia as they happen with Linux! This is farce.

Every single problem that Android has had, from low-latency audio issues (they've rebuilt that a dozen times in a dozen cartoonish ways) to driver stagnancy, is completely and directly a result of Google choices and implementations (and they do the same thing again and again! It's remarkable). The notion that Google is going to fix all of their own self-sabotage by starting anew is comedic in a sense, and is the folly of countless foolish projects. "We keep fucking up again and again...let's start from scratch and this time we'll surely do it right!"

This time, however, it'll be different...

[wolfgke]

It is a difference whether the fault for this lies at the Google engineers or the kernel maintainers.

[endorphone]

Google has never been required to commit upstream, but they've chosen to do so because of the ease of integrating changes downstream. They could have forked off and bashed the code into whatever form they wanted to. Breaking anything and everything.

But they didn't. They kept keeping it in sync. They probably had a reason for doing that.

[pjmlp]

Google has chosen to cherry pick commits to upstream, there is still plenty of stuff on AOSP, and much more on internal Googleplex repos.

[monocasa]

Well a change of the kernel's license from GPLv2 to BSD is certainly a change. I'm sure the OEMs think that gains them something.

[endorphone]

Android OEMs are not in the business of writing kernels, though, and the changes they do are minimal. And their HAL/chipset code -- the thing they might actually care about as IP -- is not governed by the GPL at all, nor is any of the enormous volume of system and userspace code they write.

It's a neat initiative and might yield something interesting, but if the Linux kernel was replaced by Fuscia the ramifications are seemingly very minor. Android's many issues have never been at the kernel level.

[monocasa]

The components of their HAL/chipset code in the kernel are most of the time governed by the GPL. We can take some examples apart of need be.

[pjmlp]

After Project Treble that only applies to the legacy HAL code.

[monocasa]

Even fully after Treble, there are absolutely still kernel drivers.

[endorphone]

Qualcomm has no obligation to GPL their kernel drivers. Nor does nvidia, or any other KLM maker. This has never demanded that kernel drivers be open sourced. That's why it's impossible to make a runnable AOSP image for many devices.

Vendors still make generally minor changes to the kernel, but these are not unique or special or some competitive edge. They've just nuisance necessities.

[pjmlp]

Sure there are, they are known as "legacy HALs" on Project Treble documentation.