Http://www.kroah.com/log/linux/android-kernel-problems.html
To be honest, Google is indeed a very arrogant (or even more arrogant) Company. My personal opinion is:
Basically, it doesn't matter if you just separate them. Based on Google's experience, can I manage my own branch?
I am not especially optimistic about it. So far, everything has been going smoothly, but there are many prerequisites for this success.
For example, the hardware devices are relatively unified. For example, most of the modifications made by Google are just a small amount of effort.
So many practices that lack consideration and are not recognized by kernel developers, can they continue to develop after splitting them separately?
I don't think it will work, but there is always a doubt: There is a cloud in ancient Chinese saying, there is no worry, there will be a worry.
In fact, I do not agree with the current maintenance method of kernel developers, and think that this is an inactive thirst, and it will inevitably be unable to continue one day.
However, if Google is so large that it thinks it can have its own set, but it is not really ready, it is even more at risk.
I believe that the attitude of improper things will accumulate more and more resistance as we move forward.
I mean, there will always be problems with kernel design, coding, and management, and they are all unnecessary.
To be honest, some people often say that Google has many technologies. In fact, it doesn't really look at any technology: people's business development doesn't rely on this.
I don't know if I have to maintain such a comprehensive software stack for a long time and want to develop it.
I don't know if Google will pay the price if it determines if it is wrong. It depends on whether others' responses are correct and timely.
But why do we have to sell flaws? As far as my recent experiences are concerned, there is only one idea: Google is so cool to live...