For all his unfortunate abrasiveness one strength of Linus is and has always been his capacity to see the big picture, e.g. that usually compatibility/API,ABI stability/performances trumps extreme security measures and also he has always been able to accomodate with big players/corps in the industry.
I kind of find Linus refreshing, although I'm not sure that would survive working directly with him. I think you're begging the question though: surely compatibility/ABI stability/performance trumps extreme security (for some values of 'extreme') for people and in cases where that is true. I happen to agree with you and Linus on this (baring a known exploit of an unpatched security bug), but that heirarchy is nowhere engraven in stone. If I always recompiled a critical codebase, I wouldn't care as much about ABI compatibility. If I'm always targetting a very specific platform, I don't care about compatibility. If I'm sitting on millions of unused cycles in a single-threaded realtime environment, I only care about performance to a certain point. If I'm working on software to control a medical device or nuclear reactor, my priorites around extreme security change.
You're right about that priority for the vast majority of the Linux user-base, but that's a characteristic of those users and developers, not of kernel software in general.
I have never even tried to read the kernel source, but I read the mailing lists - quite frequently. Linus runs a tight ship. It is funniest when someone does something that will break userspace.
He sets high standards. Meet them. I don't believe any forks have really done well. His method works, as abrasive as it may appear.
Yeah that may have worked in the past when Linux was barely used by anyone. Now it is on a billion Android phones.
You can see why Google wants to move to its own OS when Linus has that attitude.
Besides, it's bullshit that you can't have ABI stability and security. It feels to me like Linus is still in the "I never write bugs" stage of denial, despite the evidence.
